family problem solving scale

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• HHS Author Manuscripts

Development of Mastery during Adolescence: The Role of Family Problem Solving *

A sense of mastery is an important component of psychological health and well-being across the life-span; however, relatively little is known about the development of mastery during childhood and adolescence. Utilizing prospective, longitudinal data from 444 adolescent sibling pairs and their parents, our conceptual model proposes that family SES in the form of parental education promotes effective family problem solving which, in turn, fosters adolescent mastery. Results show: (1) a significant increase in mastery for younger and older siblings, (2) parental education promoted effective problem solving between parents and adolescents and between siblings but not between the parents themselves, and (3) all forms of effective family problem solving predicted greater adolescent mastery. Parental education had a direct effect on adolescent mastery as well as the hypothesized indirect effect through problem solving effectiveness, suggesting both a social structural and social process influence on the development of mastery during adolescence.

Mastery, defined as a sense of having control over the forces that affect one’s life, is an important component of psychological health and well-being across the life-span (e.g., Mirowsky and Ross 1999 ; Pearlin et al. 1981 ; Shanahan and Bauer 2004 ; Thoits 1995 ). Research across multiple domains and ages documents a linkage between a sense of control and individual differences in mental and physical health (e.g., Lin and Ensel 1989 ; Pearlin and Schooler 1978 ; Thoits 1995 ). For example, Mirowsky and Ross (1998) find that personal control is associated with a healthier lifestyle. Rosenfield (1989) finds that personal control in the workplace is linked to better mental health. Keyes and Ryff (1998) include ‘environmental mastery’ (managing the demands of daily life) as one of six dimensions of psychological well-being in adulthood. In a review of control-related concepts, Skinner (1996) states “a sense of control is a robust predictor of physical and mental well-being” (549), and for some, perceived control is viewed as a “more powerful predictor of functioning than actual control” (551). Thus, whether labeled mastery, personal control, perceived control or environmental mastery, a sense of mastery is seen as central to how well individuals respond to challenges and situations encountered in everyday life 1 .

In particular, mastery is considered part of an individual’s array of personal resources that enables a person to weather negative life events and other stressful conditions, such as job loss, economic pressure, and relationship problems ( Conger and Conger 2002 ; Mirowsky and Ross 2003 ; Pearlin et al. 1981 ; Wheaton 1985 ). Indeed, “people with high self-esteem and a sense of personal control may have the skills to avoid or prevent negative events or chronic difficulties” ( Thoits 1995 : 62). Conger and Conger (2002) found that adults rated high on mastery actually demonstrated decreasing economic problems over time. Furthermore, mastery may promote good social functioning as demonstrated by a more rewarding job, a healthier lifestyle, and more satisfying relationships, (e.g., Pulkinnen, Nygren, and Kokko 2002 ; Rosenfield 1989 ). Thus, mastery appears to function as an important personal attribute that is both an indicator of positive adaptation and a resource that promotes individual well-being in adulthood.

Despite its central role in people’s lives, there is little understanding of how mastery develops. Such understanding is essential if this important characteristic is to be promoted in an effort to foster individual health and well-being. The limited knowledge regarding the development of mastery likely results from the fact that most studies linking control, stress, and mental health have focused primarily on the adult years (see Avison and Gotlib 1994 , Eckenrode and Gore 1990 ; Thoits 1995 , 2006 ). However, research is increasing on adolescent health and well-being and its implications for adult development (e.g., Colten and Gore 1991 ; Hauser and Bowlds 1990 ; Schulenberg, Maggs, and Hurrelmann 1997 ). For example, Lewis, Ross and Mirowsky (1999) propose that children from higher SES homes will develop a greater sense of control as they move into adulthood due in part to the higher level of problem solving and life skills they develop in such family environments. This view is consistent with a life course perspective which suggests that individual development unfolds in the context of family interactions and family socioeconomic circumstances ( Caspi 2002 ; Elder, 1998 ). The life course notion involving “linked lives” proposes that parents may help their children make good choices (i.e., become more effective agents of change in their own lives) through the acquisition of constructive problem solving strategies. The current study adds to this research by examining the developmental course of mastery during adolescence and the importance of family characteristics and interactions for such development.

THE DEVELOPMENT OF MASTERY

Development of mastery over self and social situations is a key element of the self-exploration and self-evaluation that takes place during the adolescent years (e.g., Demo and Savin-Williams 1983 ; Feldman and Elliott 1990 ; Harter 1999 ; Masten et al. 1995 ). Adolescents increasingly take on new social roles as peers, co-workers, and romantic partners, and must develop a sense of control during social interactions. In these roles they are expected to handle challenges and situations that arise in multiple domains such as school, work, and family where interpersonal interactions take place ( Caspi 2002 ; Colten and Gore 1991 ; Gecas and Seff 1990 ; Mortimer and Larson 2002 ). We expect that the quality and consequences of these interactions significantly influence adolescent mastery. Indeed, Lewis and colleagues ( 1999 :1575) propose that, “An individual learns through social interaction and personal experience that his or her choices and efforts are usually likely or unlikely to affect the outcome of a situation.” Consistent with this idea, when adolescents learn that their efforts will affect the course of events and may resolve difficulties in interpersonal relationships, their sense of mastery should increase.

Based on this hypothesis that mastery is acquired in part through social interactions and their outcomes, we propose that social processes in the family significantly influence the development of mastery. We hypothesize that the interactions and negotiations that occur within the family help socialize adolescents’ mastery, and a key dimension of this socialization process involves the nature of family problem solving interactions. Also important and consistent with the life course perspective, however, is the fact that a parent’s approach to socialization practices and problem solving strategies will be influenced by their place in broader social structures. One important marker of socioeconomic status (SES) involves parents’ education, which serves as the single exogenous variable in the conceptual model that guides this study ( Figure 1 ).

THE CONCEPTUAL MODEL

Parents’ education serves as the starting point for the conceptual model because it is an important component of SES that helps identify a family’s social class or position, and social class has been linked to the socialization practices of parents and strategies for handling conflicts in social relationships ( Conger and Dogan 2007 ; Gecas 1979 ; Oakes and Rossi 2003 ). Research suggests that individuals from higher SES backgrounds may have more flexibility and more resources to deal with problems as they arise (e.g., Kohn and Schooler 1982 ; Pearlin et al. 1981 ). Mirowsky and Ross (2003) state that education is the key factor for understanding the link between SES and psychological well-being. For example, people with higher levels of education tend to have greater skills to solve complex problems, jobs with more autonomy and creativity, and more opportunities to make decisions. Parent education also plays an important role in promoting self control as children transition into adulthood ( Lewis et al. 1999 ).

Bradley and Corwyn (2002) suggest that parental education may be the most important marker of SES in terms of socialization practices and child adjustment. Education enables a person to acquire the knowledge and skills (i.e., human capital) that may influence parents’ strategies for childrearing. We would expect, therefore, that more years of education will predict more effective strategies for handling problems that arise between parents and adolescents ( Cox and Paley 1997 ). Based on this reasoning, Figure 1 proposes that (1) parents with more education will engage in more effective problem solving strategies in marital and parent-child interactions, and (2) parental education will positively impact problem solving interactions between siblings as a result of observing more highly skilled parents (see Bandura 1997 ). It is through these interaction processes that family SES indirectly promotes a sense of mastery for adolescents.

Next we build on research which suggests that experiences with parents may play an important role in children’s development of mastery and self-confidence (e.g., Gecas 1989 ; Whitbeck et al. 1997 ). Parents are viewed as the primary agents of socialization through daily interactions (e.g., Demo and Cox 2000 ; Hokoda and Fincham 1995 ). A particularly salient aspect of family interactions for the development of mastery may be conflict resolution or problem solving interactions. Our conceptual model proposes that problem solving interactions within family subsystems (marital, parent-child, sibling) serve as key contexts in which children observe, learn, and practice skills associated with managing problems (e.g., Rinaldi and Howe 2003 ; Rueter and Conger 1998 ; Shantz and Hobart 1989 ).

Research on marital conflict suggests hostility and anger between spouses may have a direct, negative effect on children’s adjustment (e.g., Cummings and O’Reilly 1997 ). When parents fail to amicably resolve conflicts, children will suffer reduced psychological well-being and, presumably, a poorer sense of mastery. Furthermore, poor relationships between parents may create problems between siblings (see Conger and Conger 1996 ) and between parents and children (e.g., Fauber and Long 1991 ; Reese-Weber 2000 ). That is, when marital problem solving skills are compromised, so too are parent-child and sibling problem solving skills; consistent with the paths shown in the conceptual model.

Regarding the parent-child subsystem, we expect that adolescents learn communication skills and strategies such as negotiation and compromise during problem solving interactions with their parents (e.g., Barber 2002 ; Noller et al. 2000 ). Adolescents who perceive their parents as supportive and fair should be more accepting of parental suggestions (e.g., Davies and Cummings 1994 ; Whitbeck et al. 1991 ). Furthermore, constructive, compared to destructive, interactions may impart a sense of confidence about handling problem situations, and promote feelings by parents and children that they can effectively deal with mutual concerns and problems ( Rueter and Conger 1998 ). These feelings of effectiveness are expected to lead to increased mastery for adolescents.

Interactions with siblings also may contribute to the development of mastery. Unlike interactions with parents, which are by definition hierarchical, interactions between adolescent siblings may be more egalitarian due to their more similar stages of verbal, cognitive, and social development ( Furman and Lanthier 1996 ; McGuire et al. 2000 ). Furthermore, adolescent siblings are expected to emulate their parents’ problem solving strategies, and when these strategies effectively resolve disagreements, adolescents will experience increased mastery in dealing with daily difficulties.

The model conceptualizes problem solving as an important skill that is acquired over time and affected by family experiences. Specifically, adolescents exposed to constructive problem solving experiences in multiple family relationships should learn to resolve problems as they arise, contributing to a sense of mastery. Such experiences stand in sharp contrast to letting problems develop into larger, unmanageable difficulties that intensify feelings of helplessness and impede positive mastery development (see Thoits 1994 ). In the following analyses, we empirically evaluate the causal paths proposed in the conceptual model, and consider related issues that may modify or extend the basic conceptual framework.

RELATED RESEARCH ISSUES

Over time adolescents increasingly become active agents in their widening social world, striving to develop an increasing sense of mastery as they assert their place in the family and autonomy from parents (e.g., Barber 2002 ; Steinberg 1990 ; Thoits 2006 ). Thus, chronological age is one factor that determines mastery (e.g., Chubb, Fertman and Ross 1997 ). Another factor is the participation of adolescents in decisions that affect their lives ( Liprie 1993 ). Most parents increasingly involve their adolescents in decisions that concern them, such as buying clothing, family activities, and weekend curfews ( Bulcroft, Carmody, and Bulcroft 1996 ; Conger, Conger, and Scaramella 1997 ). For most individuals then, we would expect to see mastery increasing over the course of adolescence due, in part, to age as well as to experiences in multiple social relationships and situations.

In addition to the effect of age and experience, gender may be associated with the developmental course of mastery. For example, parents typically place fewer restrictions on the behaviors and activities of adolescent boys compared to girls due to concerns about personal safety, sexual activity, and deviant peers ( Brown and Huang 1995 ). Lewis et al. (1999) found that girls, on average, reported a lower sense of control than boys; they suggest that boys perceive a higher sense of control compared to girls as males are typically considered to be an ‘advantaged group’ in American culture. In addition, girls tend to have a “somewhat more dependent relationship with parents during adolescence” ( Brown and Huang 1995 : 154), which may inhibit the sense of control for adolescent females. However, results from other studies of mastery and control, have reported either no effects or inconsistent results related to gender (see Chubb et al. 1997 ; Whitbeck et al. 1997 ). Based on these findings and the fact that gender might modify the impacts of the processes proposed in the conceptual model, we take gender into account in the following analyses.

Participants

The present investigation included a total of 444 adolescent sibling dyads and their parents participating in a study of family functioning and adolescent adjustment in rural Iowa. In 1989, each family included two parents, a seventh grade adolescent (the target), and a sibling within 4 years of age, either younger or older (69% of the pairs were within 2 years of age). For the present study, one of the two siblings in the dyad is treated as the younger sibling (mean age = 13.52 years, range = 10.4 to 15.58); and one as the older sibling (mean age = 15.39 years, range = 13.00 to 18.92). The younger sibling sample was 45% female and older sibling sample was 51% female.

Families were recruited from eight counties in North Central Iowa; 78% of those eligible agreed to participate. Given the ethnic composition of rural Iowa at that time, all families were of European origin. Parents completed 13.52 years of school on average; the range was 10 th grade to post-graduate work. Average per capita income was $8,475, comparable to that observed for two-parent, white families in the United States in 1988 ( U. S. Bureau of the Census, 1989 ).

Interviewers visited each family’s home annually from 1989 (Wave 0) to 1992 (Wave 3). Two 2-hour visits, about two weeks apart, were conducted each year. During the first visit, the four family members completed a set of questionnaires. During the second visit, family members participated in four videotaped interaction tasks which are not used in these analyses. See Conger and Elder (1994) for additional details regarding the study. All cases with at least one wave of data during those years were included in the analyses; 92% of the original sample participated in 1992. In order to preserve the time ordering of the data, we used mastery data for both siblings from 1990 to 1992 (Waves 1, 2, and 3) and used data for the family problem solving variables from 1989 to 1991 (Wave 0, 1, and 2), a one-year lag.

Parent education

The measure was calculated as the average years of school completed by mother and father as of 1989 (Wave 0), the first year of the study. The combined average education was 13.52 years.

We used the 7-item scale developed by Pearlin et al. (1981) ; mastery was defined as “the extent to which people see themselves as being in control of the forces that importantly affect their lives” (p. 340). Each sibling independently responded, 1 = strongly agree to 5 = strongly disagree , to items such as “I have little control over the things that happen to me”; “What happens to me in the future mostly depends on me”; and “There is little I can do to change many of the important things in my life”. The average score was used; items were coded so a high score indicated high mastery. Internal consistency ranged from α = .67 in early to α = .80 in later adolescence.

Problem solving behavior in family dyads

Problem solving (PS) was measured in three family subsystems: marital, parent-child, and sibling, using a measure created for this study ( Conger, 1989 ). For sibling PS , the younger sibling reported on his or her older sibling’s behaviors and the older sibling reported on the younger sibling’s PS behaviors. The question prompted, “Now think about what usually happens when you and your sibling have a problem to solve. Think about what your sibling does.” Questions asked how often the sibling: “listened to your ideas”; “just seemed to get angry”; “had good ideas about how to solve the problem”; “criticized you or your ideas”; “showed real interest in helping to solve the problem”; “blamed others”; “insisted that you agree with him or her”; and “changed his or her point of view to help solve the problem”. Participants answered 14 questions, 1 = always to 7 = never , about behaviors their sibling demonstrated when attempting to solve a problem. Typical problems between siblings involved personal items, chores, sharing the bathroom or the computer, and interpersonal style. All items were coded such that a higher score indicated more positive PS behaviors.

Problem solving measures for parent-child and marital dyads were constructed in the same fashion; each person responded to the same set of 14 questions worded specifically for that dyad. For marital PS , wives reported on their husbands’ behaviors and husbands reported on wives’ behaviors and these reports were averaged together for a measure of overall marital PS. For parent-child PS , each child reported on the behavior of first mother and then father (comparable data on parent report on each child was not available); reports were averaged together for a younger sibling report on parents’ PS and an older sibling report of parents’ PS. Cronbach’s alpha for the 14 item PS scale was greater than α = .81 for each dyad type across the years of the study.

Table 1 provides the descriptive statistics for the study variables. As expected, the mean level of mastery increased across time (i.e., by age) for older (3.84 to 3.96) and younger siblings (3.86 to 3.93 on a scale of 5). The time-varying covariates for parent-child and sibling -child PS interactions are shown as the mean level averaged across three measurement occasions (Wave 0, 1, and 2). Correlations (available from the first author) among the study variables were in the expected direction and were consistent with the hypothesized associations.

Means and Standard Deviations for Study Variables

Note. Sample = 444 sibling pairs (888 adolescents) in 1989 (Wave 0); 92% of the original sample participated in 1992 (Wave 3)

Analytic Approach to Model Testing

Hypothesis testing involved a model building process as illustrated in Figure 2 . First, a bivariate unconditional growth model of younger and older sibling mastery was examined. Results (not shown) indicated that both older and younger siblings demonstrated significant variability in their levels (intercepts) of mastery, and there was evidence of growth; therefore estimation of subsequent models with predictors was warranted. The intercept factor for the younger sibling growth model was centered at age 13, and for the older sibling growth model it was centered at age 15, the approximate mean ages in 1990 (Wave 1). While data collection occurred on an approximate yearly basis, mastery over time was modeled as a function of chronological age (in years), utilizing the exact age at each wave of data collection for each adolescent in the sample. For the three-year study period, ages of younger siblings ranged from 10 to 19 years and ages for older siblings ranged from 13 to 21 years. Thus, although the analytic model ( Figure 2 ) appears to suggest that all adolescents were measured at the same three measurement occasions, each adolescent was actually measured at a unique point in time, contributing a minimum of one and a maximum of three measurement points (92% had three points). Growth models are designed to handle this type of unbalanced data ( Bryk and Raudenbush 1992 ), an advantage that allows the current study to model trajectories of mastery on a time scale of chronological age rather than calendar time.

The analytical model showing the associations between parents– education, marital problem-solving, parent-child problem-solving, and sibling problem-solving and adolescent mastery over time (age) controlling for sibling gender

This approach maps on to a traditional hierarchical linear model or linear mixed model and we use the “Level I/II” notation for the equations that follow where Level I represents the within individual variability across time and Level II represents the between individual variability. However, Figure 2 reflects the fact that we specified our growth models in a larger latent variable framework using the Mplus software ( Muthén and Muthén, 2006 ) that allowed us to estimate the growth models for the older and younger siblings simultaneously along with the path analysis relating the various predictors both directly and indirectly to the growth processes.

In the unconditional growth model and all subsequent models, the intercept factors for younger and older siblings were allowed to covary freely, to compensate for the shared variance between the two siblings within each family ( Khoo and Muthén, 2000 ). Both a linear and quadratic growth factor were included in each growth curve model but no random effect was estimated for the quadratic term because no individual child had more than three occasions of measurement. However, since there was a significant quadratic fixed effect for age in the younger sibling growth model, the quadratic factor with zero variance was retained in both the older and younger sibling models for comparison. It was possible to estimate a random linear effect of age but because of the small amount of variability in that effect, all covariances with the two linear slope factors were fixed to zero. The variance structure of random effects (growth factors) for the older and younger sibling models of change in mastery as a function of age is displayed in the Level II equation given below.

Once the effect of age was taken into account (see Level I equations), the family PS variables were added to the model as lagged time-varying predictors. Time-varying predictors are allowed to take on different values at each measurement occasion, but the effects of these time-varying predictors were assumed to be constant over time ( Bryk and Raudenbush, 1992 ). The present model therefore captures year-to-year fluctuations in parent-child and sibling-child PS, while estimating time-invariant effects. Consistent with the conceptual model ( Figure 1 ), the effects of marital PS were modeled as both direct effects on observed mastery at each year, and as indirect effects on mastery through parent-child and sibling-child PS. We also modeled the hypothesized indirect effects of parents’ education on mastery through parent-child and sibling-child PS as well as through marital PS. Finally, for comparison purposes, we estimated the direct effects of parents’ education on mastery at each year.

Initially, the effects of PS (marital, parent-child, and sibling) and parents’ education on mastery were allowed to differ for younger siblings and older siblings. Then, a series of constraints were included to test whether the effects of the variables within each dyad on mastery could be considered equivalent for younger and older siblings. Finally, gender of each sibling was added as a predictor of the intercept and linear growth factors, as indicated in the Level II equations. Thus, the effects of PS and parental education were estimated while controlling for age and gender.

The analytic model for the conditional parallel growth processes is given by the Level I and II equations below. In the interest of space, only the linear equations for the older sibling outcomes at Level I and random effects at Level II are given. The equations for the younger sibling are the same at Level II and at Level I differ only in that the centering for age is at 13 instead of 15.

Level I ( t = 1, 2, and 3 and i = 1,…, n=444):

Results from Growth Modeling

Results for the final model are presented in Table 2 . All models were estimated using full-information maximum likelihood (FIML) under the missing-at-random (MAR) assumption with Mplus V4.0 ( Muthén and Muthén, 2006 ). The results are presented as unstandardized estimates of effects of predictors on growth in mastery. Initial results suggested the younger siblings have a somewhat faster rate of increase in mastery; however, when constrained to be equal, both younger and older siblings demonstrated comparable linear increases in their levels of mastery over time (b = .05). There was a small, significant, negative quadratic effect (b = −.02) in the trajectories of mastery for younger siblings, suggesting a slight deceleration or leveling off in growth of mastery. That is, growth in mastery could still be occurring but at a slower pace than earlier ages.

Conditional Growth Model with Unstandardized Path Analysis Results

LL = −7708.90, # of parameters = 101, n = 444 sibling pairs

Findings in Table 2 show that gender was marginally related to the intercept (b = −.09, p = .06) and was significantly related to the linear slope of mastery for younger, but not older, siblings. Specifically, younger girls demonstrated lower levels of mastery than boys at age 13 (b = −.09) but they increased in mastery at a faster rate over time (b=.05). Next we consider the associations among the hypothesized predictors and mastery. For the time-varying covariates involving family PS, we report a single coefficient for predictors because their effects are held to be equal over time. For example, the relationship between parent-child PS from wave 0 to mastery at wave 1 is constrained to be equal to the same association from wave 1 to wave 2.

Of the remaining covariates, only PS interactions with parents had a significantly different effect on mastery for younger compared to older siblings. For both younger and older siblings, positive PS interactions with parents predicted higher levels of mastery, with the expected change in mastery being larger for younger (b = .18) compared to (b = .11) older siblings. On the other hand, constraining the effects of sibling and marital PS and parents’ education on mastery to be equal for older and younger siblings did not significantly worsen model fit compared with allowing these effects to be freely estimated, based on a likelihood ratio test for nested models (χ 2 = 11.28, df = 7, p = .13) (see Singer and Willett, 2003). Thus, the results in Table 2 are presented with equality constraints for younger and older siblings for these predictors of mastery. Positive PS interactions with siblings equally predicted higher levels of mastery during each subsequent year for older and younger siblings (b = .04). Positive marital PS interactions had a significant positive direct effect (b = .06) on mastery as well as a positive indirect effect through parent-child (b = .03) and sibling interactions (b = .01). Similar results were found for parents’ education which has a significant direct effect on mastery (b = .02) with comparatively small indirect effects through parent-child, sibling-child, and marital PS.

Variations in the parent-child and sibling PS interactions were explained by PS interactions within the marital dyad and by parents’ education (e.g., b = .30 for marital PS predicting parent-child PS). We did not find a significant association between marital PS and parents’ education (b = −.01).

The present study evaluated a conceptual model which proposed that parental education would promote effective family problem solving interactions which, in turn, would foster mastery across the years of adolescence. In addition, we expected that mastery should increase with age and that gender might influence the development of mastery. We consider the findings from the study and their implications in turn.

The Role of Family SES

Consistent with the conceptual model ( Figure 1 ), parent education had an indirect effect on adolescent mastery through its positive association with effective PS interactions between parents and adolescents and between siblings. These results suggest that family social status in the form of parents’ education has a pervasive effect on family interactions that facilitate the development of mastery. Lewis and colleagues (1999) suggest that “better educated parents may … help their children develop skills and habits that make the children more effective” (1578). This tendency of better educated parents to engage in more effective socialization practices is consistent with research on childrearing strategies (e.g., Conger and Dogan 2007 ). Parental education also had a direct relationship with sibling problem solving; this may reflect a process whereby siblings adopt patterns of thought and action similar to those used by their parents.

In addition to the results predicted by the conceptual model, two findings deserve special mention. First, we found a significant direct effect of parents’ education on mastery; this suggests that family PS behaviors do not entirely account for the impact of family SES on the development of mastery. It is possible that if a wider variety of parenting behaviors had been included, the influence of parents’ education might have been largely attenuated. For example, a broader array of socialization practices involving control strategies, direct tutoring and affective processes not considered in this report may be influenced by parental education and also affect the development of mastery (e.g., Conger and Dogan 2007 ). Furthermore, the influence of parental education may be genetically mediated to some degree which could be addressed with a genetically informed research design ( Conger and Donnellan 2007 ). Finally, parental education likely affects the broader social environment to which the adolescent is exposed, and which may affect the development of mastery. These possibilities merit attention in future research.

Second, we did not find a direct effect of parents’ education on marital problem solving. On first reflection this result seems contradictory to the general arguments in the conceptual model. If better educated parents are more skillful and adaptive in handling family problems in a constructive and effective manner, why aren’t these skills reflected in their interactions with one another? The literature demonstrates a robust relationship between parental education and the socialization of children ( Conger and& Dogan 2007 ). In marriage, however, the findings appear to be more complex (see Faust and McKibben 1999 ). It may be that our measure of problem solving may not adequately capture the complexity of PS style between these long married couples (on average 17 years). They may have well developed styles for handling and avoiding problems. It could also be that parents at this stage of the life course are more child-focused and their interactions revolve around helping their offspring face the challenges of adolescence. Finally, the emotional tone expressed by the couples during PS interactions may be important to consider. Further study will be needed to see if these factors help explain the absence of a significant association between parental education and marital problem solving in the present report.

Family Problem Solving and the Development of Mastery

Consistent with expectations, effective marital PS predicted more effective PS interactions between parents and children and between siblings. Marital PS also had a significant indirect effect on adolescent mastery through its effect on PS in both the parent-adolescent and sibling family subsystems. These results are consistent with earlier studies that find an indirect effect of marital conflict on child adjustment through parent-child relations (e.g., Fauber and Long 1991 ; Reese-Weber 2000 ). Our findings extend this research, suggesting not only an indirect effect of marital interactions on adolescent outcomes through the parent-child dyad but also through the sibling dyad. The robust influence of marital PS is also reflected by its direct relationship with adolescent mastery; consistent with studies which find a direct effect of marital dynamics on child adjustment (e.g., Harold et al. 1997 ). These findings suggest that exposure to effective PS between parents indicates to adolescents that difficulties and disagreements can be resolved in relationships in general, thus giving them greater confidence that they can control events in their lives.

Also as predicted, effective PS interactions with parents were related to individual differences in mastery over time for both older and younger siblings. These results suggest that adolescents’ mastery increased when they felt listened to and had an active role in solving problems and making decisions. These findings also are consistent with earlier research documenting that children learn to resolve problems and negotiate solutions most effectively under conditions of warm and supportive family relations ( Davies and Cummings 1994 ; Little and Conger 2007 ; Rueter and Conger 1995b , 1998 ).

Problem solving with parents had a larger effect on younger compared to older siblings’ mastery, perhaps reflecting the fact that parents may provide less guidance to older siblings who are in their late teens and approaching the transition to adulthood. Support for this interpretation comes from previous research which finds that, as adolescents increasingly participate in decisions that affect their lives, their sense of control increases ( Conger et al. 1997 ; Liprie 1993 ; Bulcroft et al. 1996 ).

We also found that PS experiences with siblings explained unique variance in adolescent mastery, which provides new insight on the possible consequences of sibling conflict resolution. Previous studies with younger children have found that most sibling conflicts ended with parental intervention ( McGuire et al. 2000 ) or that siblings’ resolution strategies were inferior to those proposed by parents ( Tucker, McHale, and Crouter 2003 ). The results reported here, however, are supportive of the notion that adolescents’ positive PS interactions with their siblings contribute independently to their sense of mastery. Moreover, results from this study suggest that both older and younger siblings contribute to one another’s development of mastery across the years of adolescence. Future studies should examine reciprocal influences between siblings at different stages of development to further our understanding of this process.

Effects of Age and Gender

As expected, we found that mastery increased throughout adolescence for both older and younger siblings, consistent with prior research which finds that mastery increases with age ( Mirowsky and Ross 1998 , 1999 ). Younger siblings also demonstrated a slowing rate of change in mastery over time. It may be that these younger siblings experience an increase in mastery during early adolescence, when parents begin to grant them more autonomy but that the growth in mastery levels off somewhat as parents retain control over certain areas. In contrast, the rate of change for older siblings does not slow, perhaps reflective of an increasing sense of independence, particularly for those who have left home to attend school or start work. This would be consistent with findings by Lewis et al. (1999) who suggest that the sense of control increases significantly during the transition to adulthood.

Gender was not related to the intercept or rate of change in mastery for older siblings. However, for younger siblings gender was marginally associated with the level and significantly associated with linear growth of mastery. Female younger siblings indicated a slightly lower initial level of mastery (age 13) which may be related to several factors. First, a lower sense of mastery may be related to the generally lower levels of self -esteem that manifest themselves about the time that girls are undergoing the pubertal transition in early adolescence (e.g., Brooks-Gunn and Warren 1985 ; Harter 1990 ). Lower mastery in early adolescence also may be related to stressors encountered during other normative life course transitions such as changing schools, dating, and having conflicts with parents ( Call and Mortimer 2001 ; Colten and Gore 1991 ). However, we did not find this same gender difference for older siblings, thus mastery may increase as girls accommodate to the challenges of early adolescence. This interpretation is based in part on the significant interaction effect of age and gender that suggests that although younger sisters start lower, they demonstrate a higher average linear growth rate compared to that for younger brothers. That is, they tend to catch up with boys over time. This issue deserves further examination in future research.

Contributions, Limitations and Future Directions

This study advances earlier research by examining family influences on the development of mastery at an earlier age than has typically been done in previous research (e.g., Lewis et al. 1999 ). It also specifically investigated family influences that have been presumed to be important in earlier studies but were not directly examined (e.g., Lewis et al. 1999 ). In addition, it is one of the rare studies of mastery during the years of adolescence and the only study of which we are aware that considers sibling as well as parental influence on mastery. Taken together, the findings illustrate one set of processes through which family SES (education) promotes family interactions that advance the development of mastery during the adolescent years. Presumably these early advantages will lay the groundwork for a healthier individual more capable of successfully negotiating the stresses and strains that characterize the life course.

The present study makes promising contributions to our understanding of the links between family experiences and adolescent mastery; however, there are a few limitations that must be noted. Due to data analytic requirements, measures of both mastery and problem solving behaviors for parent-adolescent and sibling dyads employed adolescent self-report which may contribute to some shared method bias (see Lorenz et al. 1991 ). However, the use of independent reports from parents for their education and marital problem solving strengthened our confidence in the results presented here. That is, the associations among these variables cannot be attributed to reliance on a single informant. We were also somewhat limited by having only three time points for assessing adolescent mastery and problem solving interactions. However, the ability to analyze these data by the age of each respondent at each measurement occasion increased our ability to examine the nature of mastery over the second decade of life (i.e., 10 to 21 years of age as opposed to three calendar years, 1990–1992). Finally, we must be cautious in generalizing these results due to the homogeneous sample; however, we note that other findings from this panel study have been replicated in more diverse ethnic and cultural groups (e.g., R. Conger et al. 2002 ; Parke et al. 2004 ; Solantus, Leinonen, and Punamaki 2004 ), which increases our confidence in the potential generalizability of these results as well.

Although these results examined the effects of family problem solving on the development of mastery, it is also likely that a developing sense of mastery may impact a person’s approach to problem solving. That is, the process may be reciprocal, such as the reciprocal relationship between negative life events and young adult mastery found by Shanahan and Bauer (2004) . One can imagine a scenario in which adolescents with higher mastery are more willing to engage in problem solving interactions, which in turn contribute to an increase in their mastery and self confidence (see Pulkinnen et al. 2002 ; Thoits 2006 ). This is consistent with the idea that mastery develops through personal experiences and social interactions ( Skinner 1996 ). Furthermore, adolescents who have more successful problem solving experiences may become better at selecting themselves out of situations where conflicts and negative events may occur (see Thoits 2006 ). Future research would benefit from an examination of the reciprocal effects of mastery and problem solving over multiple time points and in multiple settings across the life course.

The present findings could also have important implications for adolescents’ relationships with peers and romantic partners. In families with high levels of recurring or unresolved conflict, adolescents’ mastery may suffer from repeated failures in conflict resolution ( Forgatch 1989 ; Rueter and Conger 1995a ). These adolescents may feel less confident about resolving problems in close relationships when difficulties arise ( Rosenfield 1989 ; Rueter and Conger 1995b ). An important extension of the present study will be to examine how problem solving experiences in the family of origin and adolescent mastery combine to affect the ability to make a successful transition to adulthood by fostering better relationships with peers, co-workers, romantic partners and one’s own children. Although a small number of studies have begun to examine these issues (e.g. Lewis et al. 1999 ; Rosenfield 1989 ), a great deal of research remains to develop a richer understanding of how family processes and individual mastery affect a successful transition to adulthood.

As noted at the beginning of this article, a long history of empirical research has established the role of mastery in the maintenance of health and well-being during the adult years. The importance of mastery as an individual attribute of great significance is beyond question. With a few exceptions (e.g. Lewis et al. 1999 ), what has been lacking has been research that provides a clear understanding of how family social position and social dynamics foster the development of a strong sense of mastery. With such understanding, social services and policies can be advanced that will promote growth in mastery in subsequent generations of young people. If the results of this study are replicated and extended to more diverse populations, they suggest that social policies which increase educational quality and availability to all members of our society should promote individual mastery and social processes that foster the development of this attribute. The results also suggest specific, mastery-enhancing skills that might be taught to families with regard to the way they handle difficulties and disagreements. Simply put, while the present findings shed theoretical light on the issues investigated, they may also have applied significance of real social importance.

Biographies

Katherine Jewsbury Conger is an Associate Professor in Human Development and Family Studies at the University of California, Davis. Her research focuses on the impact of economic stress on family functioning with a special emphasis on sibling relationships and adolescent health and well being.

Shannon Tierney Williams is a research associate with the University of California, Davis and Zetetic Associates. Her research focuses on the importance of children's interpersonal relationships within the context of families, child care settings, and schools from infancy through adolescence.

Wendy M. Little is a doctoral candidate in Human Development at the University of California, Davis. Her research focuses on family processes, sibling relationships and individual adjustment during adolescence and emerging adulthood.

Katherine Masyn is an Assistant Professor at the University of California, Davis. Her research focuses on the development, refinement, and application of latent variable models for understanding population heterogeneity in longitudinal processes such as growth trajectory, event history, and latent transition analyses.

Barbara Shebloski is a postdoctoral scholar with the Family Research Group and lecturer in Human and Community Development at the University of California, Davis. Her research focuses on parenting and sibling relationship quality, and factors related to continuity and change in intergenerational educational achievement.

* Current support comes from the National Institute of Child Health and Human Development, the National Institute on Drug Abuse, and the National Institute of Mental Health ( {"type":"entrez-nucleotide","attrs":{"text":"HD047573","term_id":"300616611"}} HD047573 , {"type":"entrez-nucleotide","attrs":{"text":"HD051746","term_id":"300619503"}} HD051746 , and {"type":"entrez-nucleotide","attrs":{"text":"MH051361","term_id":"1394622422"}} MH051361 ). Support for earlier years of the study came from multiple sources, including NIMH (MH00567, MH19734, MH43270, MH59355, MH62989, and MH48165), NIDA (DA05347), NICHD (HD027724), the Bureau of Maternal and Child Health (MCJ-109572), and the MacArthur Foundation Research Network on Successful Adolescent Development Among Youth in High-Risk Settings. We thank Peggy Thoits, Eliza Pavalko and anonymous reviewers for constructive feedback on earlier versions of this paper.

1 Our study focuses specifically on mastery (personal control), and not self-efficacy. Although self efficacy, the belief that you can perform a specific behavior successfully or achieve a certain outcome falls under the larger umbrella of self-concept, as does mastery, it is a distinct concept and we do not address it in this study. We refer interested readers to the literatures on self-efficacy and self-concept (see Bandura 1997 and Harter 1999 respectively).

• Avison William R, Gotlib Ian H., editors. Stress and mental health: Contemporary issues and prospects for the future. New York, NY, US: Plenum Press; 1994. [ Google Scholar ]
• Bandura Albert. Self-efficacy: The exercise of control. New York: Freeman and Company; 1997. [ Google Scholar ]
• Barber Brian. Intrusive parenting: How psychological control affects children and adolescents. Washington, DC, US: American Psychological Association; 2002. [ Google Scholar ]
• Bradley RH, Corwyn RF. Socioeconomic status and child development. Annual Review of Psychology. 2002; 53 :371–399. [ PubMed ] [ Google Scholar ]
• Brooks-Gunn Jeanne, Warren Michelle P. The effects of delayed menarche in different contexts: Dance and nondance students. Journal of Youth and Adolescence. 1985; 14 :285–300. [ PubMed ] [ Google Scholar ]
• Brown Bradford B, Huang Bih-Hui. Examining parenting practices in different peer contexts: Implications for adolescent trajectories. In: Crockett LJ, Crouter AC, editors. Pathways through adolescence: Individual development in relation to social contexts. New Jersey: Erlbaum; 1995. pp. 151–174. [ Google Scholar ]
• Bryk Anthony S, Raudenbush Stephen W. Hierarchical linear models: Applications and data analysis methods. Thousand Oaks, CA: Sage Publications; 1992. [ Google Scholar ]
• Bulcroft Richard A, Carmody Dianne C, Bulcroft Kris A. Patterns of parental independence giving to adolescents: Variations by race, age, and gender of child. Journal of Marriage and the Family. 1996; 58 :866–883. [ Google Scholar ]
• Call Kathleen Thiede, Mortimer Jeylan T., editors. Arenas of Comfort in Adolescence: A Study of Adjustment in Context. New Jersey: Erlbaum; 2001. [ Google Scholar ]
• Caspi Avshalom. Social selection, social causation, and developmental pathways: Empirical strategies for better understanding how individuals and environments are linked across the life-course. In: Pulkinnen Lea, Avshalom Caspi., editors. Paths to successful development: Personality in the life course. Cambridge, UK: Cambridge University Press; 2002. pp. 281–301. [ Google Scholar ]
• Chubb Nancy H, Fertman Carl I, Ross Jennifer L. Adolescent self-esteem and locus of control: A longitudinal study of gender and age differences. Adolescence. 1997; 32 :113–129. [ PubMed ] [ Google Scholar ]
• Colten Mary Ellen, Gore Susan., editors. Adolescent Stress: Causes and Consequences. New York: Aldine de Gruyter; 1991. [ Google Scholar ]
• Conger Katherine J, Conger Rand D, Scaramella Laura V. Parents, siblings, psychological control, and adolescent adjustment. Journal of Adolescent Research. 1997; 12 :113–138. [ Google Scholar ]
• Conger Rand D. Problem solving scale. Ames: Iowa State University; 1989. Measure developed for the Iowa Youth and Families Project. [ Google Scholar ]
• Conger Rand D, Conger Katherine J. Resilience in Midwestern families: Selected findings from the first decade of a prospective, longitudinal study. Journal of Marriage and Family. 2002; 64 :361–373. [ Google Scholar ]
• Conger Rand D, Conger Katherine J. Sibling relationships. In: Simons RL, editor. Understanding differences between divorced and intact families: Stress, interaction, and child outcomes. Thousand Oaks, CA: Sage Publications, Inc.; 1996. pp. 104–121. [ Google Scholar ]
• Conger Rand D, Dogan Shannon J. Social class and socialization in families. In: Grusec Joan, Hastings Paul., editors. Handbook of Socialization. New York: Guildford Press; 2007. pp. 433–460. [ Google Scholar ]
• Conger Rand D, Donnellan M Brent. An interactionist perspective on the socioeconomic context of human development. Annual Review of Psychology. 2007; 58 :175–199. [ PubMed ] [ Google Scholar ]
• Conger Rand D, Elder Glen HJ, Lorenz Frederick O, Simons Ronald L, Whitbeck Leslie B., editors. Families in troubled times: Adapting to change in rural America. Hawthorne, NY: Aldine de Gruyter; 1994. [ Google Scholar ]
• Conger Rand D, Wallace Lora E, Sun Yumei, Simons Ronald L, McLoyd Vonnie C, Brody Gene H. Economic pressure in African American families: A replication and extension of the family stress model. Developmental Psychology. 2002; 38 :179–193. [ PubMed ] [ Google Scholar ]
• Conger Rand D, Cui Ming, Bryant Chalandra M, Elder Glen HJ. Competence in early adult romantic relationships: A developmental perspective on family influences. Journal of Personality and Social Psychology. 2000; 79 :224–237. [ PubMed ] [ Google Scholar ]
• Cox Martha J, Paley Blair. Families as systems. Annual Review of Psychology. 1997; 48 :243–267. [ PubMed ] [ Google Scholar ]
• Cummings EM, O'Reilly Anne Watson. Fathers in family context: Effects of marital quality on child adjustment. In: Lamb ME, editor. The role of the father in child development. 3rd ed. John Hoboken, NJ: Wiley & Sons, Inc.; 1997. pp. 49–65. [ Google Scholar ]
• Davies Patrick T, Cummings E Mark. Marital conflict and child adjustment: An emotional security hypothesis. Psychological Bulletin. 1994; 116 :387–411. [ PubMed ] [ Google Scholar ]
• Demo David H, Cox Martha J. Families with young children: A review of research in the 1990s. Journal of Marriage & the Family. 2000; 62 :876–895. [ Google Scholar ]
• Demo David H, Savin-Williams Ritch C. Early adolescent self-esteem as a function of social class: Rosenberg and Pearlin revisited. American Journal of Sociology. 1983; 88 :763–774. [ Google Scholar ]
• Eckenrode John, Gore Susan., editors. Stress between work and family. New York, NY, US: Plenum Press; 1990. [ Google Scholar ]
• Elder Glen H., Jr The life course as developmental theory. Child Development. 1998; 69 :1–12. [ PubMed ] [ Google Scholar ]
• Fauber Robert L, Long Nicholas. Children in context: The role of the family in child psychotherapy. Journal of Consulting and Clinical Psychology. 1991; 59 :813–820. [ PubMed ] [ Google Scholar ]
• Faust Kimberly A, McKibben Jerome N. Marital dissolution: Divorce, separation, annulment, and widowhood. In: Sussman M, Steinmetz SK, Peterson GW, editors. Handbook of Marriage and the Family. 2nd edition. New York: Plenum Press; 1999. pp. 475–499. [ Google Scholar ]
• Feldman S Shirley, Elliott Glen R., editors. At the threshold: The developing adolescent. Cambridge, MA: Harvard University Press; 1990. [ Google Scholar ]
• Forgatch Marion S. Patterns and outcomes in family problem solving: The disrupting effect of negative emotion. Journal of Marriage & the Family. 1989; 51 :115–124. [ Google Scholar ]
• Furman Wyndol, Lanthier Richard P. Personality and sibling relationships. In: Brody GH, editor. Sibling relationships: Their causes and consequences. Westport, CT: Ablex Publishing; 1996. pp. 127–146. [ Google Scholar ]
• Gecas Viktor. The influence of social class on socialization. In: Burr WR, Hill R, Nye FI, Reiss IL, editors. Contemporary theories about the family. Vol. 1. New York: Free Press; 1979. pp. 364–404. [ Google Scholar ]
• Gecas Viktor, Seff Monica A. Social class and self-esteem: Psychological centrality, compensation, and the relative effects of work and home. Social Psychology Quarterly. Special Issue: Social Structure and the Individual. 1990; 53 :165–173. [ Google Scholar ]
• Harold Gordon T, Fincham Frank D, Osborne Lori N, Conger Rand D. Mom and dad are at it again: Adolescent perceptions of marital conflict and adolescent psychological distress. Developmental Psychology. 1997; 33 :333–350. [ PubMed ] [ Google Scholar ]
• Harter Susan. The construction of the self: A developmental perspective. New York: The Guilford Press; 1999. [ Google Scholar ]
• Harter Susan. Self and identity development. In: Feldman SS, Elliott GR, editors. At the threshold: The developing adolescent. Cambridge, MA: Harvard University Press; 1990. pp. 352–387. [ Google Scholar ]
• Hauser Stuart T, Bowlds Mary K. Stress, coping, and adaptation. In: Feldman SS, Elliott GR, editors. At the threshold: The developing adolescent. Cambridge, MA: Harvard University Press; 1990. pp. 388–413. [ Google Scholar ]
• Hokoda Audrey, Fincham Frank D. Origins of children's helpless and mastery achievement patterns in the family. Journal of Educational Psychology. 1995; 87 :375–385. [ Google Scholar ]
• Keyes Corey LM, Ryff Carol D. Generativity in adult lives: Social structural contours and quality of life consequences. In: McAdams DP, de St. Aubin E, editors. Generativity and adult development: How and why we care for the next generation. Washington, DC: American Psychological Association; 1998. pp. 227–263. [ Google Scholar ]
• Khoo Siek-Toon, Muthén Bengt. Longitudinal data on families: Growth modeling alternatives. In: Rose JS, Chassin L, Presson CC, Sherman SJ, editors. Multivariate applications in substance use research: New methods for new questions. Mahwah, New Jersey: Erlbaum Associates; 2000. pp. 43–78. [ Google Scholar ]
• Kohn Melvin L, Schooler Carmi. Job conditions and personality: A longitudinal assessment of their reciprocal effects. American Journal of Sociology. 1982; 87 :1257–1286. [ Google Scholar ]
• Lewis Susan K, Ross Catherine E, Mirowsky John. Establishing a sense of personal control in the transition to adulthood. Social Forces. 1999; 77 :1573–1599. [ Google Scholar ]
• Lin Nan, Ensel Walter M. Life stress and health: Stressors and resources. American Sociological Review. 1989; 54 :382–399. [ Google Scholar ]
• Liprie Mary-Lou. Adolescents’ contributions to family decision making. Marriage and Family Review. 1993; 18 :241–253. [ Google Scholar ]
• Little Wendy, Conger Katherine J. Decision-Making Processes in Single-Parent Families and Adolescent Mastery. Boston, MA: Poster presented at biennial Meeting of Society for Research in Child Development; 2007. [ Google Scholar ]
• Lorenz Frederick O, Conger Rand D, Simon Ronald L, Whitbeck Leslie B, Elder Glen HJ. Economic pressure and marital quality: An illustration of the method variance problem in the causal modeling of family processes. Journal of Marriage and the Family. 1991; 53 :375–388. [ Google Scholar ]
• Masten Ann S, Coatsworth JD, Neemann Jennifer, Gest Scott D, Tellegen Auke, Garmezy Norman. The structure and coherence of competence from childhood through adolescence. Child Development. 1995; 66 :1635–1659. [ PubMed ] [ Google Scholar ]
• McGuire Shirley, Manke Beth, Eftekhari Afsoon, Dunn Judy. Children's perceptions of sibling conflict during middle childhood: Issues and sibling (dis)similarity. Social Development. 2000; 9 :173–190. [ Google Scholar ]
• Mirowsky John, Ross Catherine E. Education, personal control, lifestyle and health: A human capital hypothesis. Research on Aging. 1998; 20 :415–449. [ Google Scholar ]
• Mirowsky John, Ross Catherine E. Well-being across the life course. In: Horwitz AV, Scheid TL, editors. A Handbook for the study of mental health: Social contexts, theories, and systems. Cambridge, UK: Cambridge University Press; 1999. pp. 328–347. [ Google Scholar ]
• Mirowsky John, Ross Catherine E. Social Causes of Psychological Distress. 2nd ed. New York: Aldine de Gruyter; 2003. [ Google Scholar ]
• Mortimer Jeylan T, Larson Reed W., editors. The changing adolescent experience: Societal trends and the transition to adulthood. New York, NY, US: Cambridge University Press; 2002. [ Google Scholar ]
• Muthén Linda, Muthén Bengt. Mplus User’s Guide. 4th Edition. Los Angeles, CA: Muthén & Muthén; 2006. [ Google Scholar ]
• Noller Patricia, Feeney Judith A, Sheehan Grania, Peterson Candida. Marital conflict patterns: Links with family conflict and family members' perceptions of one another. Personal Relationships. 2000; 7 :79–94. [ Google Scholar ]
• Oakes John M, Rossi Paul H. The measurement of SES in health research: Current practice and steps toward a new approach. Social Science and Medicine. 2003; 56 :769–784. [ PubMed ] [ Google Scholar ]
• Parke Ross D, Coltrane Scott, Duffy Sharon, Buriel Raymond, Dennis Jessica, Powers Justina, French Sabine, Widaman Keith F. Economic Stress, Parenting, and Child Adjustment in Mexican American and European American Families. Child Development. 2004; 75 :1632–1656. [ PubMed ] [ Google Scholar ]
• Pearlin Leonard I, Menaghan Elizabeth G, Lieberman Morton A, Mullan Joseph T. The stress process. Journal of Health and Social Behavior. 1981; 22 :337–356. [ PubMed ] [ Google Scholar ]
• Pearlin Leonard I, Schooler Carmi. The structure of coping. Journal of Health and Social Behavior. 1978; 19 :2–21. [ PubMed ] [ Google Scholar ]
• Pulkinnen Lea, Nygren Hille, Kokko Katja. Successful development: Childhood antecedents of adaptive psychosocial functioning in adulthood. Journal of Adult Development. 2002; 9 :251–265. [ Google Scholar ]
• Reese-Weber Marla. Middle and late adolescents' conflict resolution skills and siblings: Associations with interparental and parent-adolescent conflict resolution. Journal of Youth and Adolescence. 2000; 29 :697–711. [ Google Scholar ]
• Rinaldi Christina M, Howe Nina. Perceptions of Constructive and Destructive Conflict Within and Across Family Subsystems. Infant and Child Development. 2003; 12 :441–459. [ Google Scholar ]
• Rosenfield Sarah. The effects of women's employment: Personal control and sex differences in mental health. Journal of Health and Social Behavior. 1989; 30 :77–91. [ PubMed ] [ Google Scholar ]
• Rueter Martha A, Conger Rand D. Antecedents of parent-adolescent disagreements. Journal of Marriage and the Family. 1995a; 57 :435–448. [ Google Scholar ]
• Rueter Martha A, Conger Rand D. Interaction style, problem-solving behavior, and family problem-solving effectiveness. Child Development. 1995b; 66 :98–115. [ PubMed ] [ Google Scholar ]
• Rueter Martha A, Conger Rand D. Reciprocal influences between parenting and adolescent problem-solving behavior. Developmental Psychology. 1998; 34 :1470–1482. [ PubMed ] [ Google Scholar ]
• Schulenberg John, Maggs Jennifer L, Hurrelmann Klaus., editors. Health risks and developmental transitions during adolescence. New York, NY, US: Cambridge University Press; 1997. [ Google Scholar ]
• Shanahan Michael J, Bauer Daniel J. Developmental properties of transactional models: The case of life events and mastery from adolescence to young adulthood. Development and Psychopathology. 2004; 16 :1095–1117. [ PubMed ] [ Google Scholar ]
• Shantz Carolyn U, Hobart Cathy J. Social conflict and development: Peers and siblings. In: Berndt TJ, Ladd GW, editors. Peer Relationships in Child Development. Oxford, England: John Wiley & Sons; 1989. pp. 71–94. [ Google Scholar ]
• Singer Judith D, Willett John B. Applied longitudinal data analysis: Modeling change and event occurrence. New York, NY: Oxford University Press; 2003. [ Google Scholar ]
• Skinner Ellen A. A guide to constructs of control. Journal of Personality and Social Psychology. 1996; 71 :549–570. [ PubMed ] [ Google Scholar ]
• Solantaus Tytti, Leinonen Jenni, Punamäki Raija-Leena. Children's Mental Health in Times of Economic Recession: Replication and Extension of the Family Economic Stress Model in Finland. Developmental Psychology. 2004; 40 :412–429. [ PubMed ] [ Google Scholar ]
• Steinberg Lawrence. Autonomy, conflict, and harmony in the family relationship. In: Feldman SS, Elliott GR, editors. At the threshold: The developing adolescent. Cambridge, MA: Harvard University Press; 1990. pp. 255–276. [ Google Scholar ]
• Thoits Peggy A. Stressors and problem-solving: The individual as psychological activist. Journal of Health and Social Behavior. 1994; 35 :143–160. [ PubMed ] [ Google Scholar ]
• Thoits Peggy A. Stress, Coping, and Social Support Processes: Where Are We? What Next? Journal of Health and Social Behavior. 1995:53–79. (Extra Issue) [ PubMed ] [ Google Scholar ]
• Thoits Peggy A. Personal Agency in the Stress Process. Journal of Health and Social Behavior. 2006; 47 :309–323. [ PubMed ] [ Google Scholar ]
• Tucker Corinna J, McHale Susan M, Crouter Ann C. Conflict resolution: Links with adolescents' family relationships and individual well-being. Journal of Family Issues. 2003; 24 :715–736. [ Google Scholar ]
• U. S. Bureau of the Census. Money income and poverty status in the United States: 1988. Washington, DC: Department US of Commerce; 1989. Current Population Reports, Consumer Income (Series P-60, No. 166). [ Google Scholar ]
• Wheaton Blair. Models for the Stress-Buffering Functions of Coping Resources. Journal of Health and Social Behavior. 1985; 26 :352–364. [ PubMed ] [ Google Scholar ]
• Whitbeck Les B, Simons Ronald L, Conger Rand D, Lorenz Frederick O. Family economic hardship, parental support, and adolescent self-esteem. Social Psychology Quarterly. 1991; 54 :353–363. [ Google Scholar ]
• Whitbeck Leslie B, Simons Ronald L, Conger Rand D, Wickrama KAS, Ackley Kevin A, Elder Glen H., Jr The effects of parents’ working conditions and family economic hardship on parenting behaviors and children’s self-efficacy. Social Psychology Quarterly. 1997; 60 :291–303. [ Google Scholar ]

Jump to navigation

• About Child Trauma
• Trauma Types
• Populations at Risk
• Trauma Treatments
• Screening and Assessment
• Psychological First Aid and SPR
• Core Curriculum on Childhood Trauma
• Trauma-Informed Systems
• Culture and Trauma
• Families and Trauma
• Family-Youth-Provider-Partnerships
• Secondary Traumatic Stress
• Trauma-Informed Organizational Assessment
• All NCTSN Resources
• Información en Español
• Public Awareness
• Structure and Governance
• Network Members
• Strategic Partnerships
• Policy Issues
• Position Statements
• About This Website

GET HELP NOW

Family Assessment Device

You are here, fad - family assessment device.

Based on the McMaster Model of Family Functioning (MMFF), the FAD measures structural, organizational, and transactional characteristics of families. It consists of 6 scales that assess the 6 dimensions of the MMFF - affective involvement, affective responsiveness, behavioral control, communication, problem solving, and roles - as well as a 7th scale measuring general family functioning. The measure is comprised of 60 statements about a family; respondents (typically, all family members ages 12+) are asked to rate how well each statement describes their own family. The FAD is scored by adding the responses (1-4) for each scale and dividing by the number of items in each scale (6-12). Higher scores indicate worse levels of family functioning.

The FAD has been widely used in both research and clinical practice. Uses include: (1) screening to identify families experiencing problems, (2) identifying specific domains in which families are experiencing problems, and (3) assessing change following treatment.

Epstein, N. B., Baldwin, L. M., Bishop, D. S. (1983). The McMaster family assessment device. Journal of Marital and Family Therapy. 9, (2), 171-180.

Available from the authors at the Family Research Program, Butler Hospital, 345 Blackstone Boulevard, Providence, RI 92906.

Administration

The FAD utilizes a 4 point Likert scale, with answer choices “strongly agree,” “agree,” “disagree,” and “strongly disagree.” Answers are coded 1 - 4 with higher numbers indicating more problematic functioning.

Parallel or Alternate Forms

The General Functioning Scale can be used as a brief stand-alone measure of family functioning (FAD-12). This scale/measure has solid psychometric properties.

Psychometrics

General family functioning: 2.00 Communication, affective responsiveness, problem solving: 2.20 Roles: 2.3 Behavior control: 1.9 Affective involvement: 2.10 (Epstein, Baldwin, & Bishop, 1983)

NOTES: In a study comparing mother vs. father scores, scores on 5 out of the 7 FAD scales were significantly correlated (Akister & Stevenson-Hinde, 1991). In a study exploring the use of the FAD with school age children, inter-rater reliability was calculated for 2 groups of mother-child dyads: those with a child aged 7 - 11 and those with a child aged 12 - 17. Young children’s FAD scores showed good agreement with maternal scores; scores on 6 out of the 7 FAD scales were significantly correlated. In contrast, older children’s scores on only 2 of the 7 scales significantly correlated with maternal scores. This suggests that, while some rater pairs demonstrate good inter-rater reliability, other raters have unique perspectives on family functioning. The authors suggest that this reflects the adolescent developmental stage as well as decreased physical proximity to the family as the child grows up (Bihun et al., 2002). Akister, J. & Stevenson-Hinde, J. (1991). Identifying families at risk: Exploring the potential of the McMaster Family Assessment Device. Journal of Family Therapy, 13, 411-421. Bihun, J.T., Wamboldt, M.Z., Gavin, L.A., & Wamboldt, F.S. (2002). Can the Family Assessment Device (FAD) be used with school aged children? Family Process, 41, 723-731. Epstein, N., Baldwin, L., & Bishop, D. (1983). The McMaster family assessment device. Journal of Marital and Family Therapy, 9, 19-31. Miller, I.W., Epstein, N.B., Bishop, D.S., & Keitner, G.I. (1985). The McMaster Family Assessment Device: Reliability and validity. Journal of Marital & Family Therapy, 11(4), 345-356.

Akister, J. & Stevenson-Hinde, J. (1991). identifying families at risk: Exploring the potential of the McMaster Family Assessment Device. Journal of Family Therapy, 13, 411-421. Clark, M.S., Rubenach, S., & Winsor, A. (2003). A randomized controlled trial of an education counseling intervention for families after stroke. Clinical Rehabilitation, 17, 703-712. Evans, R.L., Matlock, A.L., Bishop, D.S., Stranahan, S., & Pederson, C. (1988). Family intervention after stroke: Does counseling or education help? Fristad, M.A. (1989). A comparison of the McMaster and Circumplex family assessment instruments. Journal of Marital & Family Therapy, 15(3), 259-269. Kabacoff, R.I., Miller, I.W., Bishop, D.S., Epstein, N.B., & Keitner, G.I. (1990). A psychometric study of the McMaster Family Assessment Device in psychiatric, medical, and nonclinical samples. Journal of Family Psychology, 3(4), 431-439. Joffe, R., Offord, D., & Boyle, M. (1988). Ontario Child Health Study: Suicidal behavior in youth age 12-16 years. American Journal of Psychiatry, 145, 1420-1423. Maziade, M., Cote, R., Boutin, P., Bernier, H., & Thivierge, J. (1987). Temperament and intellectual development: A longitudinal study from infancy to four years. American Journal of Psychiatry, 144, 144-150. Miller, I.W., Epstein, N.B., Bishop, D.S., & Keitner, G.I. (1985). The McMaster Family Assessment Device: Reliability & validity. Journal of Marital & Family Therapy, 11(4), 345-356. Miller, I.W., Kabacoff, R.I., Epstein, N.B., Bishop, D.S., Keitner, G.I., Baldwin, L.M., et al. (1994). The development of a clinical rating scale for the McMaster Model of Family Functioning. Family Process, 33, 53-69. Tonge, B., Brereton, A., Kiomall, M., MacKinnon, A., King, N., & Rinehart, N. (2006). Effects on parental mental health of an education and skills training program for parents of young children with autism: A randomized controlled trial. Journal of the American Academy of Child and Adolescent Psychiatry, 45(5), 561-569.

Predictive Validity: clinical samples, nonclinical samples (Arpin, Fitch, Browne, & Corey, 1990; Bishop et al., 1987; Browne, Arpin, Corey, Fitch, & Cafni, 1990; Joffe et al., 1988; Maziade et al., 1985, 1987) Concurrent Validity: clinical samples, nonclinical samples (Miller, Epstein, Bishop, & Keitner, 1985; Miller et al., 1994) Sensitivity: Sensitivity and specificity were calculated based on clinician interview ratings matched with FAD assessments. FAD cutoffs have sensitivity rates of 57%-87%. The authors assert that these rates are similar to other assessments including some lab tests. (Miller, Epstein, Bishop, & Keitner, 1985) Specificity: Sensitivity and specificity were calculated based on clinician interview ratings matched with FAD assessments. FAD cutoffs have specificity rates of 64%-79%. The authors assert that these rates are similar to other assessments including some lab tests. (Miller, Epstein, Bishop, & Keitner, 1985) Arpin, K., Fitch, M., Browne, G., & Corey, C. (1990). Prevalence and correlates of family dysfunction and poor adjustment to chronic illness in speciality clinics. Journal of Clinical Epidemiology, 43, 373-383. Bishop, D., Evans, R., Minden, S., McGowan, M., Marlowe, S., Andreoli, N., Trotter, J., & Williams, C. (1987). Family functioning across different chronic illness/disability groups. Archives of Physical Medicine and Rehabilitation, 68, 79-87. Brown, G., Arpin, K., Corey, P., Fitch, M., & Cafni, A. (1990). Individual correlates of health service utilization and the cost of poor adjustment to chronic illness. Medical Care, 28, 43-58. Joffe, R., Offord, D., & Boyle, M. (1988). Ontario Child Health Study: Suicidal behavior in youth age 12-16 years. American Journal of Psychiatry, 145, 1420-1423. Maziade, M., Caperaa, P., Laplante, B., Boudreault, M., Thivierge, J., Cote, R., et al. (1985). Value of difficult temperament among 7-year-olds in the general population for predicting psychiatric diagnosis at age 12. American Journal of Psychiatry, 142, 943-946. Maziade, M., Cote, R., Boutin, P., Bernier, H., & Thivierge, J. (1987). Temperament and intellectual development: A longitudinal study from infancy to four years. American Journal of Psychiatry, 144, 144-150. Miller, I.W., Epstein, N.B., Bishop, D.S., & Keitner, G.I. (1985). The McMaster Family Assessment Device: Reliability and validity. Journal of Marital & Family Therapy, 11(4), 345-356. Miller, I.W., Kabacoff, R.I., Epstein, N.B., Bishop, D.S., Keitner, G.I., Baldwin, L.M., et al. (1994). The development of a clinical rating scale for the McMaster Model of Family Functioning. Family Process, 33(1), 53-69.

(1) Some studies have called the FAD’s 7 factor structure into question. One study suggests that a 2 factor model (comprised of connection and commitment factors), rather than the 7 factor McMaster Model, provides a better fit for the data used to develop the FAD (Ridenour, Daley, & Reich, 1999). Cross-cultural research has also challenged the 7 factor structure of the FAD. However, rather than reflecting poorly on the original FAD, it is possible that these cross-cultural differences in factor structure reflect different cultural norms and expectations for family functioning. (2) The McMaster Model proposes that the 7 dimensions of family functioning are subsumed by an underlying health-pathology dimension; it follows that the 7 dimensions will be intercorrelated. This has led to criticism that the scales are not adequately independent and should not be considered discrete dimensions. (3) Acceptable psychometric properties have been demonstrated using primarily white, middle class samples; however, additional psychometric studies with racially, ethnically, and socioeconomically diverse samples are warranted. (4) Additional psychometric research is essential to establish the reliability and validity of several of the FAD translations. (5) The FAD’s utility is limited by the lack of adequate standardization and norms.

Translations

Population information.

The FAD was developed using a sample of 503 individuals drawn from both the general U.S. population as well as various clinical populations. The sample included 209 undergraduate students, 9 advanced psychology students and their families, 6 families of patients in a stroke rehabilitation unit, 4 families with children in a psychiatric day hospital, and 93 families with an adult in a psychiatric hospital. The adult inpatient members represented a variety of diagnoses, including adjustment disorders, major depressive disorder, bipolar disorder, personality disorders, substance use disorders, schizophrenic disorders, somatoform disorders, and mental retardation. Demographic characteristics (i.e., race/ethnicity, gender, SES) of the development sample are not reported (Epstein, Baldwin, & Bishop, 1983).

Pros & Cons/References

(1) The FAD is based on the McMaster Model of Family Functioning, a multi-dimensional clinical model with constructs derived from clinical experience. (2) The FAD’s 6 domains (problem solving, communication, roles, affective responsiveness, affective involvement, and behavior control) and overall functioning domain provide a comprehensive picture of family functioning in multiple areas. (3) The FAD is a multi-informant assessment designed to be completed by all family members over age 12. This provides insight into multiple perspectives on family functioning. (4) The FAD has considerable clinical utility. The FAD and FAD-12 can be used to screen for families with problematic functioning; the FAD can also be used to identify specific areas of problematic family functioning and to assess changes post intervention.

(1) The FAD’s clinical utility is limited by the lack of a manual, adequate standardization, and instructions for interpreting multiple family member perspectives. (2) Historically, the FAD has been used primarily with white, middle-class families. Additional research with diverse racial/ethnic and socio-economic groups is needed to establish utility with these populations. (3) While the FAD has been translated into 14 languages, these translations have varying levels of reliability and validity and warrant further study. (4) The FAD scales are correlated with – rather than independent of – one another. Thus, families with problematic functioning in one area are likely to experience problems in other areas as well.

Akister, J. & Stevenson-Hinde, J. (1991). Identifying families at risk: Exploring the potential of the McMaster Family Assessment Device. Journal of Family Therapy, 13, 411-421. Arpin, K., Fitch, M., Browne, G., & Corey, C. (1990). Prevalence and correlates of family dysfunction and poor adjustment to chronic illness in speciality clinics. Journal of Clinical Epidemiology, 43, 373-383. Barroilhet, S., Cano-Prous, A., Cervera-Enguix, S., Forjax, M.J., & Guillen-Grima, F. (2009). A Spanish version of the Family Assessment Device. Social Psychiatry & Psychiatric Epidemiology, 44(12), 1051-1065. Bihun, J.T., Wamboldt, M.Z., Gavin, L.A., & Wamboldt, F.S. (2002). Can the Family Assessment Device (FAD) be used with school aged children? Family Process, 41, 723-731. Bishop, D., Evans, R., Minden, S., McGowan, M., Marlowe, S., Andreoli, N., Trotter, J., & Williams, C. (1987). Family functioning across different chronic illness/disability groups. Archives of Physical Medicine and Rehabilitation, 68, 79-87. Brown, G., Arpin, K., Corey, P., Fitch, M., & Cafni, A. (1990). Individual correlates of health service utilization and the cost of poor adjustment to chronic illness. Medical Care, 28, 43-58. Clark, M.S., Rubenach, S., & Winsor, A. (2003). A randomized controlled trial of an education counseling intervention for families after stroke. Clinical Rehabilitation, 17, 703-712. Epstein, N., Baldwin, L., & Bishop, D. (1983). The McMaster family assessment device. Journal of Marital and Family Therapy, 9, 19-31. Evans, R.L., Matlock, A.L., Bishop, D.S., Stranahan, S., & Pederson, C. (1988). Family intervention after stroke: Does counseling or education help? Fristad, M.A. (1989). A comparison of the McMaster and Circumplex family assessment instruments. Journal of Marital & Family Therapy, 15(3), 259-269. Joffe, R., Offord, D., & Boyle, M. (1988). Ontario Child Health Study: Suicidal behavior in youth age 12-16 years. American Journal of Psychiatry, 145, 1420-1423. Kabacoff, R.I., Miller, I.W., Bishop, D.S., Epstein, N.B., & Keitner, G.I. (1990). A psychometric study of the McMaster Family Assessment Device in psychiatric, medical, and nonclinical samples. Journal of Family Psychology, 3(4), 431-439. Maziade, M., Caperaa, P., Laplante, B., Boudreault, M., Thivierge, J., Cote, R., et al. (1985). Value of difficult temperament among 7-year-olds in the general population for predicting psychiatric diagnosis at age 12. American Journal of Psychiatry, 142, 943-946. Maziade, M., Cote, R., Boutin, P., Bernier, H., & Thivierge, J. (1987). Temperament and intellectual development: A longitudinal study from infancy to four years. American Journal of Psychiatry, 144, 144-150. Miller, I.W., Epstein, N.B., Bishop, D.S., & Keitner, G.I. (1985). The McMaster Family Assessment Device: Reliability and validity. Journal of Marital & Family Therapy, 11(4), 345-356. Miller, I.W., Ryan, C.E., Keitner, G.I., Bishop, D., & Epstein, N.B. (2000). The McMaster approach to families: Theory, assessment, treatment and research. Journal of Family Therapy, 22, 168-189. Ridenour, T.A., Daley, J., & Reich, W. (1999). Factor analyses of the Family Assessment Device. Family Process, 38(4), 497-510. Speranza, M., Guenole, F., Revah-Levy, A., Egler, P.J., Negadi, F., Falissard, B., et al. (2012). The French version of the Family Assessment Device. Canadian Journal of Psychiatry, 57(9), 570-577. Tonge, B., Brereton, A., Kiomall, M., MacKinnon, A., King, N., & Rinehart, N. (2006). Effects on parental mental health of an education and skills training program for parents of young children with autism: A randomized controlled trial. Journal of the American Academy of Child and Adolescent Psychiatry, 45(5), 561-569.

• Printer-friendly version

• Open access
• Published: 12 March 2019

A psychometric study of the Family Resilience Assessment Scale among families of children with autism spectrum disorder

• Emily Gardiner 1 , 2 ,
• Louise C. Mâsse 3 , 2 &
• Grace Iarocci 4  

Health and Quality of Life Outcomes volume  17 , Article number:  45 ( 2019 ) Cite this article

15k Accesses

22 Citations

9 Altmetric

Metrics details

The family system represents a critical context within which children develop. Although raising a child with a disability may represent a challenge to this dynamic system, research demonstrates that families have the capacity to demonstrate both maladaptation and resilience in the face of related stressors. In the current study, we examined the psychometric properties of the Family Resilience Assessment Scale (FRAS) among families of children with autism spectrum disorder (ASD). This tool is the only measure of family resilience that seeks to identify within-family protective factors, including the extent to which they rely on adaptive belief systems, organizational patterns, and communication processes. Identifying protective processes utilized by those who show resilience is critical within both clinical practice and research, as it aligns with a strength-based perspective that builds on what families are doing well.

Participants included 174 caregivers of individuals with ASD (84% mothers). Caregivers completed the FRAS, as well as the Beach Center Family Quality of Life Scale. The 54-item FRAS was submitted to an exploratory factor analysis, using the iterated principal factor method with a promax rotation.

Fifty-one items across 3 factors (Family Communication and Problem Solving, Utilizing Social and Economic Resources, Family Spirituality) were retained, explaining 52% of the total variance. The final scale demonstrated convergent validity with the Family Quality of Life assessment tool.

Conclusions

It is our hope that identifying the optimal scale structure will encourage other researchers to utilize this measure with families of children with ASD, thus continuing to advance the study of family resilience within this unique context.

The study of resilience emerged from early work that identified groups of children who demonstrated a remarkable ability to not only recover, but flourish despite early conditions characterized by extreme adversity (e.g., [ 1 , 2 , 3 ]). In such work, the family was often viewed as a primary mechanism of pathology, and something that resilient children were able to successfully overcome. Work to expand the construct of resilience has increasingly acknowledged that although the family environment may impose constraints that place a child’s optimal development at risk, it can also serve an important protective function [ 4 ]. In particular, strong inter-family relationships have been shown to play important buffering roles in the face of different circumstances of adversity, such as divorce [ 5 ], chronic illness [ 6 ], and disability [ 7 ]. This has contributed to the proliferation of a strength-based perspective in research and practice, in which the family is not viewed as disordered, but as capable [ 8 ]. The family is considered to be a dynamic system that holds the capacity to demonstrate both maladaptation and resilience in the face of encountered stressors [ 9 ]. By identifying critical protective processes utilized by those who show resilience, professionals working with at-risk families can encourage positive appraisal and coping patterns that promote continued adaptation [ 10 , 11 ].

Raising a child with autism spectrum disorder (ASD) represents one circumstance that may place the family at risk. Upon receiving a diagnosis, families must re-negotiate established roles and adjust their expectations for child rearing within this new and uncharted context [ 12 ]. Although this situation may present considerable challenge, with research indicating that many families experience heightened stress [ 13 ] and poor quality of life [ 14 ], qualitative research highlights families’ tremendous capacity for resilience [ 15 , 16 , 17 , 18 ].

Various models of family resilience have been put forth, primarily by McCubbin and colleagues (e.g., Double ABCX, Family Adjustment and Adaptation Response, T-Double ABCX, Resiliency Model of Family Adjustment and Adaptation; see Nichols [ 4 ] for a review), that theorize about the interactional nature of stressors, family perceptions, and protective factors in the facilitation of resilience. These theories, however, do not suggest what those protective factors might be. Walsh has written extensively about critical family processes, as outlined within her framework that conceptualizes resilience as more than surviving, but as emerging strengthened and more resourceful [ 19 , 20 , 21 , 22 ]. Walsh’s perspective aligns with findings from qualitative research with families of children with ASD, in which participants’ descriptions suggest that their experiences are more than a ‘bouncing back’ from adversity, but instead reflect a process of growth. For example, in Bayat’s [ 15 ] study, one parent described a strengthened marriage, with another stating: “Autism has made us stronger and more cohesive” (p. 709).

Walsh [ 20 , 21 ] suggests that family resilience involves three broad processes, each containing three sub-processes: Family Belief Systems that allow individuals to find meaning within adversity, maintain an optimistic outlook, and to have strong spiritual beliefs; Organizational Patterns that are flexible, connected, and include access to necessary social and economic resources; and Communication Processes characterized by clarity, open emotional expression, and collaborative problem solving. The Family Resilience Assessment Scale (FRAS) [ 23 ], which is based on Walsh’s model, includes 6 subscales, which are meant to reflect Walsh’s 9 processes. Table  1 presents the conceptual alignment between the 6 FRAS subscales and Walsh’s 9 processes.

The FRAS has received considerable attention, and has been used to examine family resilience in various contexts, including among international adoptees [ 24 ], vocational rehabilitation clients [ 25 ], as well as among families of children with developmental disabilities [ 26 ], including epilepsy [ 27 ], attention deficit hyperactivity disorder [ 28 ] and ASD [ 29 , 30 , 31 ]. Moreover, validation studies have been conducted with participants from Singapore [ 27 ], Turkey [ 32 ], and China [ 33 ], though none have included caregivers of children with disabilities.

Conceptually, the FRAS is particularly appropriate to study family life when a child is diagnosed with ASD, as the subscales reflect processes that have been identified within qualitative research as important for family adaptation. For example, research with families of children with ASD has shown that positive appraisal and humour are important protective factors [ 16 ], and highlights the importance of family environments characterized by mutually supportive relationships [ 15 , 17 ]. Others report on the centrality of family spirituality [ 15 ], as well as families’ reliance on social support networks [ 16 , 17 , 34 ]. As the English version of the scale has not been validated with caregivers beyond the original scale development, we do not know whether the FRAS’ proposed 6-factor model is appropriate for families of children with ASD.

In the present study, we examined the psychometric properties of the 6 scales included within the FRAS among families who care for a child with ASD to determine whether the structural properties of the scales hold. In addition, this study examined whether the FRAS was associated with the Beach Center Family Quality of Life Scale [ 35 ]. It is our hope that identifying the optimal scale structure will encourage other researchers in the field to utilize this measure with families of children with ASD, thus continuing to advance the study of family resilience within this unique context.

Participants

Data were examined from 174 caregivers of individuals with ASD who had participated in a larger study exploring family quality of life and resilience. Caregivers were mostly mothers (83.9%) and represented a range of family ethnicities. Most were married, had received some kind of formal education post high school, and had high family incomes, with the median reported income range being $80,000 - $109,999. Almost half the sample (42.5%) indicated their family had experienced a significant life event (e.g., divorce, death, move, job loss) in the previous 6 months, most frequently endorsing death of a family member or close friend (24.3%), move (23.0%), family breakup (21.6%) (i.e., divorce or separation), or family illness or injury (17.6%). Informed consent was obtained by all individual participants included in the study, and the study received approval from the University Research Ethics Board. See Table  2 for demographic characteristics of the sample.

Diagnostic confirmation

All individuals with ASD had received a standardized clinical diagnosis of ASD from a qualified paediatrician, registered doctoral-level psychologist, or psychiatrist associated with the provincial government-funded autism assessment network, or through a qualified private clinician. All diagnoses were based on the Diagnostic and Statistical Manual of Mental Disorders ( DSM ) [ 36 , 37 ] and confirmed using the Autism Diagnostic Interview–Revised (ADI-R) [ 38 ] and Autism Diagnostic Observation Schedule (ADOS) [ 39 ], both of which are gold standard tools of ASD diagnostic assessment. As the ASD diagnosis is tied directly to substantial provincial funding programs, British Columbia has instituted standardized diagnostic practices. All individuals are required to be diagnosed by ADOS- and ADI-R-trained clinicians who use these tools and clinical judgment to make the diagnosis. This also pertains to individuals who have been diagnosed in a different province or country, as they are required to be re-diagnosed upon their arrival to British Columbia using these assessment standards.

• Family resilience

The FRAS [ 23 ] includes 54 items and 6 scales (see Table 1 ), to which respondents are asked to rate the extent that each item describes their family based on a 4-point Likert-type scale ranging from ‘Strongly Disagree’ (1) to ‘Strongly Agree’ (4). Responses are summed, with higher scores indicating greater resilience. The instrument has been shown to demonstrate good internal consistency across the total and subscale scores (alpha = .70–.96) [ 23 ]. Subscales on the FRAS also demonstrate strong convergent validity with subscales of the Family Assessment Device (FAD) [ 40 ], a well-established tool assessing structural, organizational, and transactional aspects of family life [ 41 ].

• Family quality of life

The Beach Center Family Quality of Life (FQOL) Scale [ 35 ] assesses FQOL across five domains: Family Interaction, Parenting, Emotional Well-Being, Physical/Material Well-Being, and Disability-Related Support. This measure includes 25 questions with responses based on a 5-point rating scale ranging from ‘Very Dissatisfied’ (1) to ‘Very Satisfied’ (5). Domain scores are determined by calculating the mean rating of domain-relevant items (ranging from 1 to 5). An overall score can also be calculated by averaging all item ratings, with higher scores signifying greater quality of life satisfaction. The scale is internally consistent (alpha values ranged from 0.72 to 0.86, and was 0.92 for the overall score in this study), and has demonstrated concurrent validity with other family scales (specifically the Family APGAR [ 42 ] and Family Resources Scale [ 43 ]; see also [ 44 ]), and test-retest reliability for each subscale, as assessed 3 months apart, ranged from .60–.77 [ 35 ]. As the FQOL Scale is a well-established tool that has been used extensively among families of children with ASD [ 13 , 34 , 45 , 46 , 47 , 48 , 49 ], we expected that particular FQOL Scale subscales (e.g., Family Interaction) would be positively associated with FRAS subscales (e.g., Family Communication and Problem Solving). Thus, the FQOL Scale was chosen to establish convergent validity.

Statistical analyses

All data analyses were conducted using Stata, Version 14.2. To assess the structural properties of the FRAS among our sample of families of individuals with ASD, the 54-item measure was submitted to a Confirmatory Factor Analysis (CFA). A CFA was conducted on the proposed 6-factor model, and model fit was assessed with various indices, including Root Mean Squared Error of Approximation (RMSEA), Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), and Standardized Root Square Mean Residual (SRMR). Cut-off criteria as recommended by Hu and Bentler [ 50 ] were utilized, such that for the RMSEA, a value up to .06 with a 90% confidence interval (CI) less than .08 was considered acceptable [ 51 ]. Cut-off values of .95 were utilized for the CFI and TLI, and values up to .08 were considered acceptable for the SRMR. The maximum likelihood estimation procedure was used as the data did not severely violate the normality assumption. As the 6-factor structure did not hold as hypothesized and evaluation of the modification indices did not result in modification that significantly improved the fit of the solution, the 54 FRAS items were submitted to an exploratory factor analysis (EFA). The maximum likelihood method of extraction was first utilized; however, the solution did not converge due to the presence of a Heywood case. As such, the iterated principal factor method with a promax rotation was utilized to identify the number of latent variables represented within the FRAS. Both the eigenvalues rule of 1 and percentage of variance explained by each factor (explaining at least 5 to 10% of the total variance) were evaluated to determine the number of factors retained with the EFA procedure. In addition, the solution was conceptually evaluated.

Structural properties of the FRAS

The proposed 6-factor model, which includes 54 measured variables, was submitted to a CFA. The overall fit of the model was not optimal [χ 2 (df = 1362) = 2401.53; RMSEA = .07; 90% CI = .06–.07; CFI = .81; TLI = .80; SRMR = .08], as the CFI and TLI were both well below recommended levels. Moreover, examination of the modification indices highlighted significant issues with the existing factor structure, and significant cross-loading among many variables onto other factors, leading us to re-evaluate the fit of the overall scale. As any attempts to include some of these cross-loadings did not result in improved model fit, an EFA using the iterated principal factor method of extraction was run on the data. Evaluation of the eigenvalues and percentage of variance explained by each factor suggested that a 3-factor solution be retained. The initial solution identified one item that did not load on any factors (factor loadings less than .30 in absolute value), and this item was therefore dropped (“We feel taken for granted by family members”). When the EFA was rerun, two additional items did not load on any of the three factors and were also removed (“We keep our feelings to ourselves”; “We seldom listen to family members’ concerns or problems”). See Tables  3 and 4 for item and subscale descriptives. Item skewness ranged from −.76–.97. Item kurtosis ranged from − 1.00-3.15, and 78% of the items had kurtosis values less than |1.0|. The EFA on the remaining 51-items (FRAS-ASD) suggested that a 3-factor solution be retained and the total solution explained 52.2% of the total variance. For this solution, there were no cross-loadings. The three factors conceptually measured the following concepts: ‘Family Communication and Problem Solving’ (Factor 1), ‘Utilizing Social and Economic Resources’ (Factor 2), and ‘Family Spirituality’ (Factor 3). Factors 1 to 3 explained 31, 14.7, and 6.5% of the total variance, respectively. The EFA solution is shown in Table  4 . Although Sixbey [ 23 ] indicates that subscale scores are comprised of the total of relevant item responses (i.e., additive), we report average scores, as this is more amenable for comparing family’s perceived resilience across subscales. The correlations among the 3 factors are presented in Table  5 .

Convergent validity

The Family Communication and Problem Solving subscale showed a strong and significant association with the Beach Center FQOL Scale’s Family Interaction subscale ( r  = .70, p  < .001), which assesses how well respondents feel their family can solve problems, talk openly and show affection, as well as handle unexpected events. The Utilizing Social and Economic Resources subscale was strongly related to the FQOL Emotional Well-Being ( r  = .63, p  < .001) and Physical/Material Well-Being ( r  = .41, p  < .001) subscales. The former assesses the extent to which respondents feel they can rely on family members for support, as well as access help outside the family. The Physical/Material Well-Being subscale includes items about access to transportation and medical care, but also includes an item about feeling safe in the community. As expected, the FRAS-ASD Family Spirituality subscale was not significantly correlated with any of the FQOL subscales. With the exception of Family Spirituality, associations between the FRAS-ASD and FQOL subscales not noted here were in the medium-to-large range [ 52 ] (see Table  6 ).

The current study is the first to validate the FRAS among a sample of family caregivers of children with ASD. The CFA indicated that Sixbey’s 6-factor structure did not hold. As such, we conducted an EFA, which retained 51 items across 3 factors, all of which were internally consistent. In fact, the subscale alpha values were improved from those reported in the original study [ 23 ]. Conceptually, the three factors, Family Communication and Problem Solving (FCPS), Utilizing Social and Economic Resources (USER), and Family Spirituality (FS), broadly align with Walsh’s [ 20 ] 3-domain model of family communication processes, organizational patterns, and belief systems, respectively (refer to Table 1 ).

Although the sub-processes proposed by Walsh did not emerge as distinct factors within our analyses, they are reflected within the retained factors. For example, family communication processes are said to include clarity and consistency, open emotional expression, and collaborative problem solving. Indeed, the majority of items within the FCPS subscale focus on collaborative problem solving, though both clarity of communication (e.g., “We understand communication from other family members”) and open emotional expression (e.g., “We feel free to express our opinions”) are certainly represented. In the current study all items from the original scale’s Maintaining a Positive Outlook and Ability to Make Meaning Within Adversity subscales loaded on factor 1 (FCPS) (standardized factor loadings ranged from .44–.74). This finding is likely related to the fact that all Maintaining a Positive Outlook items refer to optimism with regard to the family’s ability to solve problems, and are therefore very difficult to distinguish conceptually from those included within the original FCPS subscale (e.g., “We can work through difficulties as a family” versus “We can solve major problems”). This was also the case for two of the items from the original Ability to Make Meaning Within Adversity subscale, which refer to accepting that problems happen. The other item (“The things we do for each other make us feel a part of the family”) was likely retained within factor 1, as one would expect that giving time and energy to the family, openly demonstrating love and affection, collaborating in problem solving, and being attuned to others’ emotions would make members feel closely connected.

Family organizational patterns refer to a family’s use of social and economic resources, sense of connectedness, and flexibility in response to change. The FRAS-ASD USER subscale aligns with this process, as most items refer to whether families can rely on people in their community for assistance, and at a broad level, reflects one’s sense of community connectedness. The original FRAS contained a Family Connectedness subscale; however, in the current analyses, three of the items were not retained in the final solution and three were distributed across other subscales. One such item (“We show love and affection for family members”) loaded strongly (.51) on the FCPS subscale, and the other two, which referenced involvement in the community and feeling valued by friends, loaded on factor 2 (standardized factor loadings of .62 and .46, respectively). The final organizational sub-process identified by Walsh, flexibility, is only assessed in relation to problem solving (e.g., “We try new ways of working with problems” in the FCPS subscale), and is not represented within the FRAS-ASD as a distinct construct.

Finally, Walsh described family belief systems as involving transcendence and spirituality, positivity, and finding meaning from adversity. Within the FRAS-ASD, only family spirituality is assessed as a distinct sub-process (i.e., FS emerged as its own factor), with the structure being identical to that of the original tool. A limitation of this subscale, however, is that the included items portray a fairly narrow conceptualization of spirituality, with three of the four items inquiring directly about formal religion. Walsh, in contrast, highlights that families may practice spirituality in other ways, such as through participation in cultural traditions (e.g., through nature, music, or the arts). As such, it may be fruitful for future research to evaluate whether the inclusion of items broadening this conceptualization (e.g., We can express ourselves through music; Our family enjoys being outside together; We feel connected to the natural elements around us) produce a tool that better captures Walsh’s original intent. As described, the other two sub-processes (positivity and making meaning of adversity) are reflected within the FCPS subscale.

Overall, the content of the FRAS-ASD subscales align closely with research examining protective factors for families of children with ASD. Indeed, within qualitative research, caregivers highlight the importance of processes included within the FCPS subscale, specifically describing how flexibility, positivity, open communication and working together has helped them to adapt successfully to raising a child with ASD [ 15 , 17 ]. Similarly, quantitative research identifies that positivity, optimism, acceptance, and confidence with regard to confronting future adversities act as buffers for maladaptive outcomes, such as parental stress, depression, and poor life satisfaction (as reviewed by Bekhet and colleagues [ 16 ]). The USER subscale content is also consistent, as families identify social support from friends, family, and the community as valued [ 16 , 17 , 23 ]. Although spirituality has received less attention in the literature, themes around gaining strength through faith, and from spiritual and religious practices do emerge [ 15 , 16 , 17 ]. Although this suggests that the FRAS-ASD is well-suited to evaluating resilience within families of children with ASD, the tool does not include items specific to this circumstance. Research identifies that families place great value on access to professionals as well as to other families of children with ASD [ 17 ], from whom they can gain knowledge and seek advice, and items related to the availability of these kinds of supports may improve the tool for this context.

The fact that the 9 sub-processes described by Walsh [ 20 , 21 ] did not emerge as distinct factors is not surprising, given that one would expect at least some degree of overlap across these components. In reality, family resilience processes are not mutually exclusive, but are instead mutually influencing and reinforcing. For example, successful family problem solving involves sub-processes that cross domains, such as flexible organizational patterns in trying new approaches when existing ones are unsuccessful and open communication in order to reach a solution that is acceptable to all family members. The success of utilizing these approaches, in turn, contribute to a family’s optimistic beliefs about how well prepared they are to face both ongoing and future adversities. In fact, Walsh’s descriptions reflect this overlap, as making meaning from adversity, a belief system sub-process, is described in relation to the framing of family crises. This points to the inherent challenges associated with operationalizing conceptual models, as well as with measuring complex and dynamic inter-family processes. We suggest that the FRAS-ASD provides a psychometrically-sound assessment of resilience amongst families of children with ASD that is guided by Walsh’s 3-domain model.

Although this is the first study to validate this tool with a sample of caregivers of children with ASD, other validation studies have been conducted. The structure of the FRAS-ASD differs somewhat from those identified using CFA with Turkish [ 32 ] and Chinese [ 33 ] undergraduate students. Kaya and Arici [ 32 ] reported a 4-factor structure, consisting of FCPS, USER, Maintaining a Positive Outlook, and Ability to Make Meaning from Adversity. Li et al.’s [ 33 ] FRAS-C, in contrast, retained 32 items across 3 factors: FCPS, Utilizing Social Resources, and Maintaining a Positive Outlook. The observed differences are not surprising when taking into account the significant ‘family life’ differences one would expect between young adults attending university and family caregivers of children with ASD, as well as the inextricable influence of culture on resilience [ 53 , 54 ]. Chew and Haase [ 27 ] conducted an EFA with adolescents (mean age = 15 years) with epilepsy in Singapore, and identified 7 factors (Meaning-Making and Positive Outlook, Transcendence and Spirituality, Flexibility and Connectedness, Resources-Community, Resources-Neighbours, Clarity and Open Emotional Expression, Collaborative Problem-Solving). We suspect that the observed structural differences also relate to cultural contexts, particularly around spirituality, connectedness, open communication, and reliance on social resources. However, their findings highlight the possibility that this tool may be utilized to attain multiple family members’ perspectives, representing an exciting extension for a field that has been dominated by maternal report.

The correlations across the FRAS-ASD and Beach Center FQOL Scale subscales provide support for convergent validity. Although FQOL and ‘family resilience’ represent distinct constructs, the Beach Center FQOL Scale was chosen given both the lack of research with this population and of established measures. It is not surprising that the strongest association was between the FCPS and FQOL Family Interaction subscale, as both refer to a family environment that supports open communication, collaborative problem solving, and demonstrations of affection. The other subscales, though related, are more conceptually distinct, and this is reflected in the observed associations. Future research may seek to examine how these two FRAS-ASD subscales converge with specific measures that purport to assess analogous constructs. Though family resilience can certainly be construed as an indicator of overall family well-being, the FRAS-ASD may best provide a measure of how well a family is utilizing various processes (i.e., positive communication patterns, reliance on social support, and religious practices) that can act as protective factors, with the potential to moderate relationships between established risk factors, such as high child behavior problems, and FQOL.

Limitations, strengths, and future directions

The current findings should be considered within the context of the study limitations. First, the conclusions may be limited by our relatively small sample size, and may not generalize. Though the Kaiser-Meyer-Olkin measure verified the sampling adequacy for the analysis (KMO = .90) and Bartlett’s test of sphericity, χ 2 (1431) = 6069.71, p  < .001, indicated that correlations between items was sufficiently large for EFA, others have reported that samples of at least 300 participants are necessary [ 55 ]. Second, our sample represented a fairly ‘resource-rich’ group, as most respondents reported that they were married, well educated, and had high family incomes. The structure of the FRAS may look somewhat different for families who face more socioeconomic challenges, and future research should make efforts to include a more heterogeneous sample. The generalizability of our findings, however, are strengthened by our sample diversity with respect to reported family ethnicities, as well as the inclusion of families with children spanning a wide age range (2–35 years). Studies of family adaptation when there is a child with a disability have typically focused on the early and middle childhood years, thus neglecting the distinct challenges facing families of young adults. This developmental period can represent a significant transition for family life [ 56 ], as families are adapting to decreases in service availability, and may be negotiating changes around the structure of caregiving (i.e., responsibility may shift from parents to siblings) [ 57 ]. The fact that the FRAS-ASD has been validated with families of children with ASD ranging from early childhood to adulthood suggests that this tool may facilitate our understanding of how resilience develops and evolves across the family life cycle. Of course, however, there are myriad within-family individual factors that will influence how families experience and perceive their resilience, including those related to the vast diversity in presentation amongst those with ASD. Presence of child internalizing and externalizing difficulties, functional status, and age of diagnosis are but a few examples. Future longitudinal research would help to disentangle how resilience processes unfold in response to adversities across the life cycle of the family.

Finally, although the FRAS purports to measure ‘family’ resilience, it is important to recognize that any insights gained are representative of only one member’s perspective. This issue plagues the family adaptation literature, which is dominated by maternal report (see DeHaan et al. [ 11 ] for a discussion of methodological strategies to ameliorate this challenge). Chew and Haase’s [ 27 ] findings, however, suggest that the FRAS may be appropriate for use with multiple members, including children, which at the very least provides a more holistic picture as to how different individuals perceive inter-family dynamics. Research designs that incorporate multiple family members will best capture the “moving pictures” of family life ([ 58 ], p. 7).

We have demonstrated that the FRAS-ASD provides a valid assessment of resilience amongst families of individuals with ASD, and maps to Walsh’s 3-domain theoretical model. We suggest that this tool facilitates our understanding of within-family processes that help to protect against the impact of adversity. From a clinical perspective, there is much to be learned from families who thrive despite facing considerable challenge. By encouraging family members to listen to one another, freely express their views and needs, seek out community support, and utilize constructive problem solving strategies, family resilience can be strengthened, and we can continue to move the family adaptation field away from a focus on dysfunction.

Abbreviations

Autism Spectrum Disorder

Confirmatory Factor Analysis

Comparative Fit Index

Confidence Interval

Disability-Related Support

Exploratory Factor Analysis

Emotional Well-Being

Family Assessment Device

Family Communication and Problem Solving

Family Interaction

Family Quality of Life Total Scale

Family Quality of Life

Family Resilience Assessment Scale (Sixbey, 2005 version)

Family Resilience Assessment Scale–ASD Version

Family Spirituality

Physical/Material Well-Being

Root Mean Squared Error of Approximation

Standardized Root Square Mean Residual

Tucker-Lewis Index

Utilizing Social and Economic Resources.

Cicchetti D, Rogosch FA. The role of self-organization in the promotion of resilience in maltreated children. Dev Psychopathol. 1997;9:799–817.

Article   Google Scholar  

Garmezy N. Resiliency and vulnerability to adverse developmental outcomes associated with poverty. Am Behav Sci. 1991;34:416–30.

Werner EE, Smith RS. Journeys from childhood to midlife: risk, resilience, and recovery. Ithaca: Cornell University Press; 2001.

Google Scholar  

Nichols WC. Roads to understanding family resilience: 1920s to the twenty-first century. In: Bevcar DS, editor. Handbook of family resilience. New York: Springer; 2013. p. 3–16.

Chapter   Google Scholar  

Greeff AP, Van Der Merwe S. Variables associated with resilience in divorced families. Soc Indic Res. 2004;68:59–75.

Buchbinder M, Longhofer J, McCue K. Family routines and rituals when a parent has cancer. Fam Syst Health. 2009;27:213–27.

Greeff AP, Nolting C. Resilience in families of children with developmental disabilities. Fam Syst Health. 2013;31:396–405.

Windle G. What is resilience? A review and concept analysis. Rev Clin Gerontol. 2011;21:152–69.

Walsh F. The concept of family resilience: crisis and challenge. Fam Process. 1996;35:261–81.

Article   CAS   Google Scholar  

Black K, Lobo M. A conceptual review of family resilience factors. J Fam Nurs. 2008;14:33–55.

DeHaan LG, Hawley DR, Deal JE. Operationalizing family resilience as a process: proposed methodological strategies. In: Bevcar DS, editor. Handbook of family resilience. New York: Springer; 2013. p. 17–29.

Gardiner E, Iarocci G. Unhappy (and happy) in their own way: a developmental psychopathology perspective on quality of life for families living with developmental disability with and without autism. Res Dev Disabil. 2012;33:2177–92.

McStay RL, Dissanayake C, Trembath D. Maternal stress and family quality of life in response to raising a child with autism: from preschool to adolescence. Res Dev Disabil. 2014;35:3119–30.

Brown RI, MacAdam-Crisp J, Wang M, Iarocci G. Family quality of life when there is a child with a developmental disability. J Policy Pract Intel. 2006;3:238–45.

Bayat M. Evidence of resilience in families of children with autism. J Intellect Disabil Res. 2007;51:702–14.

Bekhet AK, Johnson NL, Zauszniewski JA. Resilience in family members of persons with autism spectrum disorder: a review of the literature. Issues Ment Health Nurs. 2012;33:650–6.

Greeff AP, van der Walt K. Resilience in families with an autistic child. Educ Train Autism De. 2010;45:347–55.

Kapp L, Brown O. Resilience in families adapting to autism spectrum disorder. J Psychol Africa. 2011;21:459–63.

Walsh F. A family resilience framework: innovative practice applications. Fam Relat. 2002;51:130–7.

Walsh F. Strengthening family resilience. 2nd ed. New York: Guilford; 2006.

Walsh F. Family resilience: strengths forged through adversity. In: Walsh F, editor. Normal family processes. 4th ed. New York: Guilford; 2012. p. 399–427.

Walsh F. Family resilience: a developmental systems framework. Eur J Dev Psychol. 2016;13:313–24.

Sixbey MT. Development of the family resilience assessment Scale to identify family resilience constructs: PhD thesis. Gainesville: University of Florida; 2005.

Buchanan T. Family resilience as a predictor of better adjustment among international adoptees: PhD Thesis. Arlington: University of Texas; 2008.

Openshaw KP. The relationship between family functioning, family resilience, and quality of life among vocational rehabilitation clients: PhD Thesis. Logan: Utah State University; 2011.

Leone E, Dorstyn D, Ward L. Defining resilience in families living with neurodevelopmental disorder: a preliminary examination of Walsh’s framework. J Dev Phys Disabil. 2016;28:595–608.

Chew J, Haase AM. Psychometric properties of the family resilience assessment Scale: a Singaporean perspective. Epilepsy Behav. 2016;61:112–9.

Holman J. Parenting stress and parenting behavior in families with a child diagnosed with ADHD: the moderating effect of family resilience. PhD Thesis. Santa Barbara: Fielding Graduate University; 2014.

Gardiner E. Quality of life in families of children with autism spectrum disorder: considerations of risk and resilience. PhD thesis. Burnaby: Simon Fraser University; 2014.

La Madrid Carpena G. Predictors of family resilience in families of children with autism spectrum disorder. MA thesis. Vancouver: Adler University; 2016.

Plumb JC. The impact of social support and family resilience on parental stress in families with a child diagnosed with an autism spectrum disorder. PhD thesis. Philadelphia: University of Pennsylvania; 2011.

Kaya M, Arici N. Turkish version of shortened family resilience assessment Scale (FRAS): the study of validity and reliability. Procd Soc Behv. 2012;55:512–20.

Li Y, Zhao Y, Zhang J, Lou F, Cao F. (2016) psychometric properties of the shortened Chinese version of the family resilience assessment Scale. J Child Fam Stud. 2016;25:2710–7.

Pozo P, Sarriá E, Brioso A. Family quality of life and psychological well-being in parents of children with autism spectrum disorders: a double ABCX model. J Intellect Disabil Res. 2014;58:442–58.

Hoffman L, Marquis J, Poston D, Summer JA, Turnbull A. Assessing family outcomes: psychometric evaluation of the beach center family quality of life Scale. J Marriage Fam. 2006;68:1069–83.

American Psychiatric Association (APA). Diagnostic and statistical manual of mental disorders. 4th ed., text rev. Washington, DC: Author; 2000.

American Psychiatric Association (APA). Diagnostic and statistical manual of mental disorders. 5th ed. Washington, DC: Author; 2013.

Book   Google Scholar  

Rutter M, Le Couteur A, ADI-R LC. Autism diagnostic interview – revised. Los Angeles: Western Psychological Services; 2008.

Lord C, Rutter M, DiLavore PC, Risi S. Autism diagnostic observation Scale–WPS (ADOS-WPS). Los Angeles: Western Psychological Services; 1999.

Epstein NB, Baldwin LM, Bishop DS. The McMaster family assessment device. J Marital Fam Ther. 1983;9:171–80.

Hamilton E, Carr A. Systematic review of self-report family assessment measures. Fam Process. 2016;55:16–30.

Smilkstein G, Ashworth C, Montano D. Validity and reliability of the family APGAR as a test of family function. J Fam Pract. 1982;15:303–11.

CAS   PubMed   Google Scholar  

Dunst CJ, Leet HE. Family Resource Scale: Reliability and validity. Asheville: Winterberry; 1985.

Hu X, Summers JA, Turnbull A, Zuna N. The quantitative measurement of family quality of life: a review of available instruments. J Intellect Disabil Res. 2011;55:1098–114.

Eskow KG, Chasson GS, Summers JA. A cross-sectional cohort study of a large, statewide Medicaid home and community-based services autism waiver program. J Autism Dev Disord. 2015;45:626–35.

Eskow K, Pineles L, Summers JA. Exploring the effect of autism waiver services on family outcomes. J Policy Pract Intel. 2011;8:28–35.

Gardiner E, Iarocci G. Family quality of life and ASD: the role of child adaptive functioning and behaviour problems. Autism Res. 2015;8:199–213.

Gardiner E, Iarocci G. Examining family quality of life within the context of a participant-directed ASD funding program in British Columbia, Canada. J Policy Pract Intel. 2018;15:110-123.

McStay RL, Trembath D, Dissanayake C. Stress and family quality of life in parents of children with autism spectrum disorder: parent gender and the double ABCX model. J Autism Dev Disord. 2014;44:3101–18.

Hu L, Bentler PM. Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. Struct Equ Modeling. 1999;6:1–55.

Acock AC. Discovering structural equation modeling using Stata, Revised edition. College Station: StataCorp; 2013.

Cohen J. A power primer. Psychol Bull. 1992;112:155–9.

Ungar M. Resilience across cultures. Brit J Soc Work. 2008;38:218–35.

Ungar M. What is resilience across cultures and contexts? Advances to the theory of positive development among individuals and families under stress. J Fam Psychother. 2010;21:1–16.

Tabachnick BG, Fidell LS. Using multivariate statistics. 6th ed. Boston: Pearson; 2013.

McGoldrick M, Carter B, Garcia-Preto N, editors. The expanded family life cycle: individual, family, and social perspectives. 4th ed. Boston: Allyn & Bacon; 2011.

Orsmond GI, Seltzer MM. Siblings of individuals with autism spectrum disorders across the life course. Ment Retard Dev D R. 2007;13:313–20.

Cowan PA, Cowan CP, Schulz MS. Thinking about risk and resilience in families. In: Hetherington EM, Blechman EA, editors. Stress, coping, and resiliency in children and families. Mahwah: Lawrence Erlbaum; 1996. p. 1–38.

Download references

Acknowledgements

Not applicable.

This research was supported by scholarships from the Social Sciences and Humanities Research Council of Canada (SSHRC) [Grant Number: 767–2011-2317] and the Autism Research Training (ART) program funded by the Canadian Institutes of Health Research (CIHR) to the first author [Grant Number: STN 63728], a Michael Smith Foundation for Health Research (MSFHR) Scholar Award to the last author, and a grant from the Laurel Foundation to the first and last authors [Grant Number: 869431]. The first author also received support from a BC Children’s Hospital / Kids Brain Health Network postdoctoral fellowship, as well as from the Sunny Hill Foundation for Children. The second author received salary support from the BC Children’s Hospital Research Institute.

Availability of data and materials

The datasets generated and/or analysed during the current study are not publicly available due to consent for this purpose not being obtained, but may be available from the corresponding author on reasonable request and with permission of the associated university Research Ethics Board.

Author information

Authors and affiliations.

BC Children’s Hospital Research Institute, 950 West 28th Avenue, Vancouver, BC, V5Z 4H4, Canada

Emily Gardiner

Department of Pediatrics, University of British Columbia, 4480 Oak Street, Vancouver, BC, V6H 3V4, Canada

Emily Gardiner & Louise C. Mâsse

School of Population and Public Health, University of British Columbia, 2206 East Mall, Vancouver, BC, V6T 1Z3, Canada

Louise C. Mâsse

Department of Psychology, Simon Fraser University, 8888 University Drive, Burnaby, BC, V5A 1S6, Canada

Grace Iarocci

You can also search for this author in PubMed   Google Scholar

Contributions

EG collected and analyzed participant data and was a major contributor in writing the manuscript. LM made substantial contributions to analysis and interpretation of data and was involved in critically revising the manuscript. GI made substantial contributions to conception and design of the study, and was a major contributor to drafting of the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Grace Iarocci .

Ethics declarations

Ethics approval and consent to participate.

Informed consent was obtained by all individual participants included in the study, and the study received approval from the associated University Research Ethics Board (2013 s0026).

Consent for publication

Competing interests.

The authors declare that they have no competing interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License ( http://creativecommons.org/licenses/by/4.0/ ), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated.

Reprints and permissions

About this article

Cite this article.

Gardiner, E., Mâsse, L.C. & Iarocci, G. A psychometric study of the Family Resilience Assessment Scale among families of children with autism spectrum disorder. Health Qual Life Outcomes 17 , 45 (2019). https://doi.org/10.1186/s12955-019-1117-x

Download citation

Received : 09 February 2018

Accepted : 07 March 2019

Published : 12 March 2019

DOI : https://doi.org/10.1186/s12955-019-1117-x

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Autism spectrum disorder
• Factor analysis

Health and Quality of Life Outcomes

ISSN: 1477-7525

• Submission enquiries: [email protected]

ORIGINAL RESEARCH article

Family resilience scale short form (frs16): validation in the us and chinese samples.

• 1 Department of Counselling and Psychology, Hong Kong Shue Yan University, North Point, Hong Kong SAR, China
• 2 Wan Chow Yuk Fan Centre for Interdisciplinary Evidence-based Practice and Research, Hong Kong Shue Yan University, North Point, Hong Kong SAR, China

Family resilience, which refers to the processes through which a family adapts to and thrives from adversities, has growing importance in recent years. In response to the need for further research on family resilience, the present research aims to abbreviate and validate Sixbey's Family Resilience Assessment Scale (FRAS) into a 16-item version Family Resilience Scale Short Form in the US (FRS16) and Chinese (FRS16_C) samples. The samples included 1,236 (Study 1) and 1,135 (Study 2) participants from the US and China, respectively. Results of confirmatory factor analysis (CFA) supported the proposed three-factor structure of FRS16: Family Communication and Connectedness, Positive Framing, and External Support across two samples. Overall, the reliability and validity of full and subscales of FRS16 and FRS16_C were satisfactory. Multi-group CFA revealed that both configural and metric invariance are supported, suggesting that participants in the US and Chinese samples assign comparable meaning to the latent factors of FRS16. Results suggested that FRS16 and FRS16_C are valid instruments for family resilience in the US and Chinese samples.

Introduction

Since 1990s, the term “family resilience” has emerged in family sciences to signify family processes that create perseverance, accommodation, and growth in responses to crises and challenges ( 1 – 3 ). The conceptualization of family resilience entails two important features. First, family resilience is more than the average of individual members' resilience, but rather encompasses dynamic, structural features of the family system which could only be understood through a broader, relational framework ( 4 ). Second, it involves more than the capacity to survive a crisis, but also to realize the potential for growth out of difficulties ( 3 ). Overcoming a challenge together, a family can emerge as more resourceful, connected, and loving in facing future crises.

Walsh's Family Resilience Model

Building upon earlier strength-based family paradigms which emphasize the interaction between stressors, family perceptions and protective factors in developing resilience (e.g., The Double ABCX model; ( 5 ), Walsh ( 3 ) proposed a family resilience model that specifies three overarching processes that promote family resilience, each of which consists of several subprocesses. The three overarching processes are Family Belief Systems , Organizational Patterns , and Communication Processes . Family Belief Systems refer to the shared reality that might facilitate normalizing and overcoming a crisis via meaning-making, maintaining positivity, and a transcendent, spiritual grounding. Organizational Patterns refer to the ways families operate during a crisis. Lastly, Communication Processes involve clear communication, open emotional expression, and collaborative problem-solving. In this framework, family resilience is a dynamic process that varies across families, contexts, and time.

This family resilience framework has tremendous implications on how to successfully cope with abrupt and persistent crises including caregiving for a member with chronic illnesses ( 6 ), traumatic societal events such as natural disaster ( 7 ) and pandemic ( 8 ), divorce ( 9 ), and the death of a family member ( 10 ). Importantly, family resilience influences children's positive development ( 11 , 12 ). Thus, prevention and intervention programs have been developed to identify and foster family resilience, in the hope of preserving family functioning and well-being during difficult situations [e.g., ( 13 – 15 )].

The Family Resilience Assessment Scale (FRAS)

In view of the theoretical importance of the family resilience construct and the lack of quantitative measurement of it, Sixbey ( 16 ) developed the Family Resilience Assessment Scale (FRAS). The 54-item FRAS was developed upon Walsh ( 3 )'s family resilience model. It includes six subscales, which reflect Walsh's three overarching family resilience processes (Family Belief Systems, Organization Patterns, and Communication Processes). Family Belief Systems are associated with the subscales Maintaining a Positive Outlook, Ability to Make Meaning of Adversity , and Family Spirituality . Organization Patterns are associated with two subscales, Utilizing Social and Economic Resources and Family Connectedness . Finally, Communication Processes are associated with the subscale Family Communication and Problem Solving .

Ever since its birth, researchers have observed the predictive value of the FRAS on mental health measures. For instance, higher score on the FRAS was associated with lower depression, anxiety, stress ( 17 ), and higher individual resilience ( 18 , 19 ). It also served as a protective factor for cancer patients and their spouses ( 20 ) as well as young people with severe epilepsy ( 21 ). Relevant to the current context, it also mitigated the negative effect of pandemic-related stressors on stress severity ( 17 ). Thus, far, the FRAS has been translated to different languages such as Chinese ( 22 – 24 ), Turkish ( 25 ), and Polish ( 26 ).

The Need of a Brief Measure of Family Resilience

The original FRAS consists of 54 items. Its length makes it cumbersome as a process or an outcome measure especially in studies that involve clinical trials, or longitudinal mutli-wave investigation, as researchers are faced with limited assessment time. A brief measure of family resilience is needed to advance the research of family resilience. To our best understanding, the shortest form of FRAS was the Chinese version which still contains 32 items ( 24 ), while most other versions have more than 40 items (e.g., ( 22 )). Furthermore, recent studies reveal the possibility of a more parsimonious factor structure of the FRAS. While Kaya and Arici ( 25 ) found a 4-factor solution, a 3-factor model showed adequate fit in Li et al. ( 24 )'s study.

As suggested by Burisch ( 27 ), acceptable validities and reliabilities can be achieved by a fairly small number of items. The present study proposed a 3-factor model which is built upon Li et al. ( 24 )'s findings. The three proposed factors are ( 1 ) Communication and Connectedness ; ( 2 ) Positive Framing ; and ( 3 ) External Resources . The first factor combines the Family Communication and Family Connectedness factors in the original scale as effective family communication can only be achieved with an optimal balance of mutual support and respect of individuals' autonomy ( 28 ). These two elements are theoretically intertwined. The second factor, Positive framing , refers to the attitudes or beliefs that help a family to maintain positivity during adversities. It combines the subcomponents Maintaining a Positive Outlook and Making Meaning of Adversities in the original FRAS which are two crucial components across different family resilience conceptualizations ( 29 , 30 ). External Resources are important for families to cope with internal and external crises. Across the literature, social [e.g., neighbor; ( 31 )], economic [e.g., welfare system; ( 32 )], and spiritual [e.g., church or other religious institutions; ( 33 )] resources are important external support for family to function well during crises. Thus, this factor captures the subcomponents Utilizing Social and Economic Resources and Family Spirituality in the original scale. Therefore, the present studies aim to derive a short scale of family resilience with a more parsimonious 3-factor structure to make family resilience research more feasible and less time-consuming. We started with selecting essential items which have the highest factor loadings from the original FRAS and then examined its psychometric properties in a US and then in a Chinese sample. Since past studies found that family resilience was associated with relationship functioning and well-being, the FRAS is hypothesized to be associated with family cohesion ( 34 ), relationship satisfaction ( 35 ), general health ( 36 ), quality of life ( 37 ), and perceived community support ( 38 ). Overall, the current research aims to abbreviate and validate Sixbey's Family Resilience Assessment Scale (FRAS) into a shorter form and cross-validated it in both the US (study 1) and Chinese (study 2) samples.

Study 1 aims to derive a reliable and valid short scale to measure family resilience based on FRAS in the US sample. We will first identify the most essential items of FRAS according to factor loadings and item-to-subscale correlation. Then, a factorial structure will be identified with CFA. Furthermore, its associations with related variables and the reliability of the composite and subscales will also be examined.

We recruited participants from the US via Amazon's Mechanical Turk (MTurk), a commonly used online crowdsourcing platform that has a diverse and stable subject pool ( 39 ). Past studies suggested that participants recruited from MTurk is more representative of the US sample in comparison to face-to-face convenience samples ( 40 ). Consent form was presented on the first page of the online survey, participants indicated understanding of their rights and granted consent by pressing “continue to next page.” Participants' confidentiality was guaranteed that personal identities are not traceable, and the data will only be used for research purposes. After data exportation from the platform, data will be stored in a local drive and is solely accessible to members of the research team. Participants must be US citizens aged above 18. They were given 1 week to complete the survey and were compensated with 0.5 USD through the built-in payment system in MTurk upon completion. This study was approved by the Human Research Ethics Committee of the affiliated university of the research team.

Characteristics of Participants

After excluding respondents who provided statistical improbable answers (e.g., worked for 500 h on average per week in the past 12 months) and did not provide answers to the key variables (i.e., family resilience and validation instruments), 1,236 participants were available for the CFA analysis. The average age of participants was 36.17 ( SD = 10.83, range = 18 to 75), and 63.7% were male. The majority of them were highly educated (92.1 % obtained a bachelor's degree or above), married (79.4%), and with an annual household income of $40,000 or above (71.1%) (see Table 1 ).

Table 1 . Descriptive statistics of demographics of US sample.

Family Resilience Scale Short Form (FRS16)

First, we selected essential items for the current proposed Family Resilience Scale Short Form from Family Resilience Assessment Scale (FRAS) ( 16 ). Considering the original scale has disproportionately large numbers of items in the Family Communication and Problem Solving (FCPS) subscale, the items of our proposed scale were selected with two criteria based on Sixbey's findings in 2005: ( 1 ) Two items with the highest factor loadings for each subscale (except for FCPS) ( 2 ) For FCPS, 6 items with both factor loadings and item-to-subscale correlation higher than 0.7. As a result, 16 items were selected for the short form (see Appendix A ). Thus, the abbreviation for Family Resilience Scale Short Form will be FRS16. Participants indicated their agreement with 16 statements on a 4-point scale (1 = strongly disagree to 4 = strongly agree ).

Validation Instruments

Quality of life was accessed with the 8-item EUROHIS-QOL scale which has been validated across different cultures ( 41 ). Participants indicated their satisfaction with different aspects of their lives on a 5-point scale (1 = not at all , 5 = completely ).

General health was accessed with the General Health Questionnaire-12 [GHQ-12; ( 42 )] which is intended to tap on respondents' recent functioning such as mood, sleeping quality and concentration compared to their normal state. The items are rated on a 4-point scale (1 = less than usual , 4 = much more than usual ), and scores were summed across items with a higher score reflecting worse health.

Relationship satisfaction was accessed with the 7-item Relationship Assessment Scale [RAS; ( 43 )]. Participants indicated how satisfied they were with their partner and relationship based on a 5-point scale (1 = low satisfaction , 5 = high satisfaction ).

Family cohesion was accessed with the 9-item family cohesion subscale of the Family Environment Scale ( 44 ). Participants indicated their agreement with the statements that described their family (0 = disagree , 1 = agree ), with a higher summation score reflecting higher family cohesion.

Perceived community support was assessed with the 2-item integration and need fulfillment subscale of the Brief Sense of Community Scale ( 45 ) with a 5-point scale (1 = strongly disagree , 5 = strongly agree ) to reflect the degree to which participants felt supported by their community.

The descriptive statistics for demographic and main variables are summarized in Tables 1 , 2 , respectively.

Table 2 . Descriptive statistics of main variables of US sample.

Confirmatory Factor Analysis (CFA) of FRS16

The factor structure of the proposed three-factor model was assessed with CFA. CFA was performed with the maximum likelihood estimation in Mplus 8.3 ( 46 ). As suggested by the CFA results, the proposed three-factor model has reasonable model fit: χ 2 (99) = 699.15, p <0.001; SRMR = 0.046; RMSEA = 0.070, CI = [0.065, 0.075]; TLI = 0.86; CFI = 0.88. The factor solution and standardized factor-loadings of all items are shown in Figure 1 . Additionally, four theoretically meaningful alternative models were also examined. In particular, the single-factor model assumes only one underlying latent factor of family resilience. The two-factor model distinguishes items into either external or internal resilience factors. The five-factor model involves problem solving, social and economic resources, positivity, family spiritually, and communication and connectedness. While the six-factor model is the one proposed in the original FRAS. As shown in Table 3 , the fit indices improved as the number of factors increased in the model, but this trend slowed down when the number of factors reached to three. Since there were no significant differences in fit indices, the three-factor model seems to have the best balance of parsimony and interpretability of item clustering.

Figure 1 . Factor structure and standardized factor loadings of FRS16.

Table 3 . Model fit indices of models with different number of factors.

Validity and Reliability

Content validity was examined with the item-to-scale and item-to-subscale correlations. As shown in Table 4 , all items were significantly and moderately correlated to both its subscale and full scale ( r s ≥ 0.36).

Table 4 . Item-to-scale and item-to-subscale correlations.

We also examined whether FRS16 is associated with quality of life, general life and family cohesion, relationship satisfaction and perceived community support. As shown in Table 2 , all of the subscales and the full scale of FRS16 were correlated with the proposed variables significantly the expected directions. Whereas, reliabilities of each subscale were above 0.6 and the reliability for the overall FRS16 scale was satisfactory.

Next, we aimed to cross-validate FRS16 with a Chinese sample which was abbreviated as FRS16_C. We first examined the construct validity of the three-factor structure in a Chinese sample with CFA, and conducted content validity tests and examination of the associations between FRS16_C and important outcomes. Additionally, the correlation patterns of the original and two other Chinese versions of FRAS and FRS16_C were examined to evaluate whether they are comparable instruments. Furthermore, test-retest reliability was also examined in a 2-week follow-up. Lastly, measurement invariance analysis was followed to obtain a stronger support that FRS16 operates similarly for both the US and Chinese samples, and participants from the two regions share parallel understandings of the construct.

Procedure and Participants

Procedures were similar to that of Study 1, we recruited participants from mainland China via an online survey distribution platform Credamo ( www.credamo.com ) which is a professional research data platform with survey distribution and data modeling services. Datasets collected from this platform has been published in well-respected international journals such as Psychological Science and Journal of Consumer Research 1 . Credamo has over 2 million registered users cover all provincial regions in China which provides researchers a wider reach to participants in different regions within China. Upon completion, participants received 16 CNY in their Credamo account which can be transferred to his or her online payment platform WeChat Pay.

After excluding those who were not residing in mainland China, the final sample consisted of 1,135 participants for the CFA test. On average, participants aged 30.01 ( SD = 5.99, range = 18–59) with slightly higher proportion of female (57.4%). The majority of them were married (68.8%), highly educated (85.6% with a bachelor's degree or above), and with a high annual household income within mainland China 2 (61%) (see Table 5 ).

Table 5 . Descriptive statistics of demographics (Chinese sample).

This study included available Chinese translations of measures in Study 1. A list of validation studies on these Chinese versions of the measurement scales is available from the corresponding author upon request. In particular, the FRS16_C used the Chinese translation of FRAS by Chiu et al. ( 22 ) and refined some wordings to fit the context of mainland China.

All participants completed the full set of questionnaires, including the FRS16_C. Among them, 213 participants filled in the full version of FRAS to test the associations between FRS16_C with the original and two Chinese versions of FRAS and compare their correlations with different demographic and convergent variables as composite scores. Another subsample of 63 participants filled in the same set of measurements 2 weeks later for the temporal consistency test of FRS16_C.

All descriptive statistics for demographic and main variables are summarized in Tables 5 , 6 , respectively.

Table 6 . Descriptive statistics of main variables (Chinese sample).

Confirmatory Factor Analysis (CFA) for FRS16_C

Similar to Study 1, CFA was estimated with the maximum likelihood with the aid of Mplus 8.3. CFA results showed that the three-factor model demonstrated excellent model fit: χ 2 (99) = 287.84, p <0.001; SRMR = 0.033; RMSEA = 0.041, CI = [0.036, 0.047]; TLI = 0.94; CFI = 0.95 (See Figure 2 for standardized factor loadings). The same set of alternative models were examined, as shown in Table 7 , the fit indices of the one- and two-factor models were poor, whereas the five- and six-factor models showed parallel model fits to that of the proposed three-factor model. Considering the three-factor model has both more interpretable and parsimonious factor structure and five- and six-factor models did not show significant improvements in model fit, a holistic evaluation suggests a three-factor model should be adopted.

Figure 2 . Factor structure and standardized factor loadings of FRS16_C.

Table 7 . Model fit indices of models with different number of factors.

Validity and Reliability of FRS16_C

Content validity.

As shown in Table 8 , all items were significantly correlated to both its subscale and the full scale with a small to moderate strength ( r s ≥ 0.31), demonstrating acceptable content validity.

Table 8 . Item-to-scale and item-to-subscale correlations.

Associations With Important Outcomes

As shown in Table 6 , the full and subscales of FRS16_C were all significantly associated with proposed outcome variables in the expected directions.

Part-Whole Association

To our best knowledge, there are two Chinese versions of FRAS with three- ( 23 ) and five-factor models ( 24 ) which were validated with family caregivers and university students samples, respectively. Therefore, we would like to compare the performance of FRS16_C in comparison with other existing FRASs by examining the correlation coefficients of FRS16_C with the original FRAS, three-factor FRAS, and five-factor FRAS and compare their correlational patterns with demographic and related variables as a composite. As shown in Table 9 , FRS16_C performed similarly to different versions of FRAS. Importantly, we observed that the shortened FRS16_C retained at least 70% of the predictive value of the 54-item scale. This pattern indicated that, given its brevity, the 16-item scale developed in the present study can stand in for the full 54-item scale in terms of predictive value.

Table 9 . Correlations of FRASs and FRS16_C with demographic and outcome variables.

Reliability

The Cronbach's alphas of Communication and Connectedness, Positive Framing, and External Resources were 0.67, 0.51, and 0.68, respectively. The reliability of overall FRS16_C was 0.78 and the test-retest reliability of FRS16_C in 2-weeks interval was acceptable (ICC = 0.87).

Measurement Invariance of FRS16

Since the group-level CFA test results were found satisfactory with both the US and Chinese samples separately, configural invariance was established. Thus, the three-factor structure is valid for both groups ( 47 ). A multi-group three-factor model was still tested with no parameter constraints, as shown in Table 10 , the fit indices of this baseline model were satisfactory, which showed converging support for configural variance. Next, we tried to establish metric variance by running a more stringent multi-group model with the factor loadings coefficients constrained to be equal across groups on top of the baseline model. The multiple-group model showed satisfactory fit (see Table 10 ), both ΔCFI and ΔRMSEA were below the recommended cut-offs [ΔCFI ≤ 0.010 and ΔRMSEA ≤ 0.015; ( 48 )], suggesting that respondents attributed the same meaning to the latent factor ( 47 ).

Table 10 . Evaluations of measurement invariance.

General Discussion

To provide a measure of family resilience for contexts in which assessment time is relatively limited, we abbreviated the Family Resilience Assessment Scale [FRAS ( 16 )] to a 16-item version, the FRS16. The proposed 3-factor model was supported across the US and Chinese samples. The 3-factor model consists of easily interpretable factors that cover the most essential components of family resilience, namely, Family Communication and Connectedness , Positive Framing , and External Support ( 4 , 28 , 49 , 50 ). These three factors also cover all of the subprocesses in Walsh ( 28 )'s theory and correspond to the different family adaptive subsystems (maintenance system, meaning system, and ecosystem) discussed in Henry et al. ( 34 ). Family communication and connectedness signifies the way through which the family members organize and communicate in response to external stressors. It captures the core elements of the maintenance system. Meanwhile, the positive framing factor involves the family world view and the ability to maintain positive strength-based outlook. This factor represents the adaptability of the family meaning system. Finally, the external resources capture positive features and resources in the external ecosystem in which crises and opportunities occur.

We also noticed that the factor loadings of the items related to family spirituality (e.g., item 13: we attend church/synagogue/mosque services) were lower (<0.30) than the two other items about social support (>0.60) in the Chinese sample. Comparatively, the factor loadings of these four items were more homogenous in the US sample. Li et al. ( 24 ) suggested that religious support is less significant in Chinese societies. In Chinese culture, the practice of ancestor worship is considered as the most common form of spiritual activity. In a study conducted in Singapore, Chinese participants identified ancestors as their spiritual sources ( 51 ), yet prominence and the format of these spiritual activities are diverse even within the Chinese culture. This might explain the more heterogeneous factor loadings of this subscale in the Chinese sample. Since the factor loadings were still statistically significant, and the item-to-subscale and item-to-total correlations remain high, we decided to keep these two items in the External Resources subscale. Nevertheless, future research should further explore the value of religious and spiritual activities in Chinese family resilience.

To assess the psychometric equivalence of the FRS16 across the US and Chinese samples, measurement invariance analyses were conducted. The configural invariance suggested that the basic organization (the 3-factor structure) is supported in both cultures while the metric invariance suggested that the contribution of each item to the latent factor to a similar degree across the two samples. These findings suggested that the proposed FRS16 includes the most essential items that measure the common, fundamental processes of family resilience across cultures. Thus, it is a viable tool for assessing family resilience in cross-cultural studies especially when assessment time is limited. Nevertheless, if cultural sensitivity is the major concern of the research, researchers can consider other longer versions which were developed specifically for a single culture [Chinese: ( 24 )]. We also examined the performance of the FRS16 in comparison with three other versions of FRAS including the full 54-item version. The part-whole correlation between the FRS16 with the original FRAS reaches 0.87. We also observed strong correlations between the FRS16 with other adapted versions of the scale ( rs >0.77). Additionally, the FRS16 shares comparable associations with validation instruments with FRAS in three different lengths.

In both samples, the FRS16 demonstrated good internal consistency for the total scale (αs > 0.80). The reliabilities of all subscales were satisfactory except Positive Framing in the Chinese sample. Nevertheless, despite the relatively low Cronbach's α (0.51), the item-to-subscale, and item-to-total correlations were both high and significant. The four items in the Positive Framing subscale were also strongly loaded on the expected latent factor. Given that this subscale consists of only four items that were related to two different types of positive framing—showing high efficacy in problem solving (e.g., Item 10: We can survive if another problem comes up) and normalizing stressors and crises (e.g., Item 15: We accept stressful events as a part of life), the relatively lower alpha was considered acceptable.

The present studies indicated that the FRS16 and its three subscales were all significantly associated with proposed outcome variables (i.e., quality of life, general health, family cohesion, relationship satisfaction, perceived community support) in the expected directions for both the US and Chinese samples. It is noted that the External Resources subscale showed relatively lower correlations with relationship satisfaction and family cohesion in both samples. It might suggest that the ability to utilize external and internal resources to cope with crisis were comparatively less crucial to the cohesion and relationship satisfaction of a family compared to Communication and Connectedness and Positive Framing. However, it does not in any way imply that External Resources was a less important component of family resilience, as it demonstrated correlation with other validation instruments in similar sizes as other components, such as its associations with general health and perceived community support.

The present study has a few limitations. First, the study was conducted on online crowdsourcing samples. Although crowdsourcing data enjoys benefits such as the more diversified demographic background of participants ( 52 ), it still cannot represent the general population. This concern has also been reflected in our US sample with higher proportion of male, and in Chinese sample with majority coming from high-annual-household-income family. Efforts should be made in validating the FRS16 with more representative and diverse samples. Additionally, some researchers argued that crowdsourcing data has the problem of cross-contamination and selective dropouts, but these issues are more relevant to experimental studies ( 53 ). Afterall, precaution is still needed in interpreting the results obtained from these crowdsourcing platforms. Second, the present study did not focus on populations that are facing acute or chronic stressors and traumatic events (e.g., caregiver of a family member who has chronic illness, recent death of a family member, economic hardship) or populations with different family characteristics (e.g., the number of family members cohabiting). The importance and meaning of family resilience could be different for people who are encountering crises or with different family characteristics. Future research could test the reliability and validity of the FRS16 in diverse populations. Third, we used quality of life, general health, relationship satisfaction, family cohesion, and perceived community support for validating the FRS16. These variables might not fully reflect the three proposed processes in family resilience. Future research might include more theoretically relevant outcome measures such as family and community support [Social Support Index; ( 54 )]. Forth, given that the cross-sectional nature of current studies, the directions of associations between FRS16 and validation instruments remain unknown, future studies are needed to provide support for predictive validations with a longitudinal or experimental design. Fifth, as the measurements were self-reported, responses might be contaminated by social desirability. However, respondents' anonymity is underscored in the beginning of the survey to minimize the influence of social desirability. Lastly, only individual family member's perception of the family dynamics and resilience was captured in this study. Other family members' perceptions may vary. In other words, it might not reflect family resilience comprehensively, as it fails to capture this process positioning families as a unit. Nevertheless, validating a short version of family resilience scale is a starting point to “study resilience in the family as a functional unity” ( 55 ) by making the measurement of each family members' perceptions less time-consuming.

Overall, the present study contributes to the literature by developing a short form of family resilience scale and cross-validated it in Western and Eastern cultures. The short version captures the essential items and the most fundamental processes underlying family resilience. Given its brevity, the FRS16 possesses acceptable psychometric properties in both cultures. It is suitable for cross-cultural study and research in which the testing time is limited such as therapy outcome assessment, intensive longitudinal research, and telephone survey.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics Statement

The studies involving human participants were reviewed and approved by Human Research Ethics Committee of the Hong Kong Shue Yan University. The patients/participants provided their written informed consent to participate in this study.

Author Contributions

TC contributed to study conceptualization, survey design, leading statistical analyses and interpretation of results, and writing the original text. CT contributed to study conceptualization, survey design, results interpretation, administrating and supervising the project, providing comments to revise the manuscript, and refine the data analysis. TS facilitated the data analysis and contributed to reviewing and editing the manuscript. HK provided study measures and contributed to the survey development and refinement. All authors contributed to the article and approved the submitted version.

This study was funded by a matching research grant awarded to the Wan Chow Yuk Fan Centre for Interdisciplinary Evidence-based Practice & Research at the Hong Kong Shue Yan University UL/RS/2021/001.

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's Note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary Material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpsyt.2022.845803/full#supplementary-material

1. ^ See https://help.credamo.com/web/#/4?page_id=118 for a list of publications in 2020 using Credamo to collect data.

2. ^ The categorization of annual household income in 2019 set by National Bureau of Statistics of China was as follow: the lowest 20% cutoff was 7380.4 CNY, following by 15777.0 CNY (mid low), 25034.7 CNY (mid), 39230.5 CNY (upper mid) and 76400.7 CNY (high). See http://www.stats.gov.cn/tjsj/ndsj/2021/indexeh.ht .

1. Ahlert IA, Greeff AP. Resilience factors associated with adaptation in families with deaf and hard of hearing children. Am Ann Deaf. (2012) 157:391–404. doi: 10.1353/aad.2012.1629

PubMed Abstract | CrossRef Full Text | Google Scholar

2. McCubbin MA, McCubbin HI. Families coping with illness: the resiliency model of family stress, adjustment and adaptation. In: Danielson CB, Hamel-Bissell B, Winstead-Fry P, editors. Families, Health and Illness: Perspectives on Coping and Intervention. St. Louis, MO: Mosby (1993). pp. 21–63.

Google Scholar

3. Walsh F. A Family resilience framework: Innovative practice application. Fam Relat. (2002) 51:130–7. doi: 10.1111/j.1741-3729.2002.00130.x

CrossRef Full Text | Google Scholar

4. Patterson JM. Integrating family resilience and family stress theory. J Marr Fam. (2002) 64:349–60. doi: 10.1111/j.1741-3737.2002.00349.x

5. McCubbin HI, Patterson JM. The family stress process: the Double ABCX Model of adjustment and adaptation. Marr Fam Rev. (1983) 6:7–37. doi: 10.1300/J002v06n01_02

6. Hamall KM, Heard TR, Inder KJ, McGill KM, Kay-Lambkin F. The Child Illness and Resilience Program (CHiRP): a study protocol of a stepped care intervention to improve the resilience and wellbeing of families living with childhood chronic illness. BMC Psychol. (2014) 2:5. doi: 10.1186/2050-7283-2-5

7. Hackbarth M, Pavkov T, Wetchler J, Flannery M. Natural disasters: an assessment of family resiliency following Hurricane Katrina. J Marital Fam Ther. (2012) 38:340–51. doi: 10.1111/j.1752-0606.2011.00227.x

8. Walsh F. Loss and resilience in the time of COVID-19: meaning making, hope, and transcendence. Fam Process. (2020) 59:898–911. doi: 10.1111/famp.12588

9. Shin SH, Choi H, Kim MJ, Kim YH. Comparing adolescents' adjustment and family resilience in divorced families depending on the types of primary caregiver. J Clin Nurs. (2010) 19:1695–706. doi: 10.1111/j.1365-2702.2009.03081.x

10. Hooghe A, Neimeyer RA. Family resilience in the wake of loss: a meaning-oriented contribution. In: Becvar DS, editor. Handbook of Family Resilience . New York, NY: Springer (1993). pp. 269–84. doi: 10.1007/978-1-4614-3917-2_16

11. Masten AS, Monn AR. Child and family resilience: a call for integrated science, practice, professional training. Fam Relat Interdisc J Appl Fam Stud. (2015) 64:5–21. doi: 10.1111/fare.12103

12. Uddin J, Alharbi N, Uddin H, Hossain MB, Hatipoglu SS, Long DL, et al. Parenting stress and family resilience affect the association of adverse childhood experiences with children's mental health and attention-deficit/hyperactivity disorder. J Affect Disord. (2020) 272:104–9. doi: 10.1016/j.jad.2020.03.132

13. Feinberg ME, Boring J, Le Y, Hostetler ML, Karre J, Irvin J, et al. Supporting military family resilience at the transition to parenthood: a randomized pilot trial of an online version of family foundations. Fam Relat. (2020) 69:109–24. doi: 10.1111/fare.12415

14. Lim H, Han K. Effects of the family resilience enhancement program for families of patients with chronic schizophrenia. J Korean Acad Nurs. (2013) 43:133–42. doi: 10.4040/jkan.2013.43.1.133

15. Saltzman WR. The FOCUS family resilience program: an innovative family intervention for trauma and loss. Fam Process. (2016) 55:647–59. doi: 10.1111/famp.12250

16. Sixbey MT,. Development of the Family Resilience Assessment Scale to identify family resilience constructs [Doctorate dissertation, University of Florida]. The University of Florida Digital Collections. Available online at: http://ufdc.ufl.edu/UFE0012882/00001 (2005).

17. Chan ACY, Piehler TF, Ho GWK. Resilience and mental health during the COVID-19 pandemic: findings from Minnesota and Hong Kong. J Affect Disord. (2021) 295:771–80. doi: 10.1016/j.jad.2021.08.144

18. Dong C, Xu R, Xu L. Relationship of childhood trauma, psychological resilience, and family resilience among undergraduate nursing students: a cross-sectional study. Perspect Psychiatr Care. (2021) 57:852–9. doi: 10.1111/ppc.12626

19. Finklestein M, Pagorek-Eshel S, Laufer A. Adolescents' individual resilience and its association with security threats, anxiety and family resilience. J Fam Stud. (2020) 1–17. doi: 10.1080/13229400.2020.1778504 [Epub ahead of print].

20. Chen JJ, Wang QL, Li HP, Zhang T, Zhang SS, Zhou MK. Family resilience, perceived social support, and individual resilience in cancer couples: analysis using the actor-partner interdependence mediation model. Eur J Oncol Nurs. (2021) 52:101932. doi: 10.1016/j.ejon.2021.101932

21. Chew J, Haase AM, Carpenter J. Individual and family factors associated with self-esteem in young people with epilepsy: a multiple mediation analysis. Epilepsy Behav. (2017) 66:19–26. doi: 10.1016/j.yebeh.2016.09.039

22. Chiu SJ, Chou YT, Chen PT, Chien LY. Psychometric properties of the Mandarin version of the family resilience assessment scale. J Child Fam Stud. (2019) 28:354–69. doi: 10.1007/s10826-018-1292-0

23. Chu AMY, Tsang JTY, Tiwari A, Yuk H, So MKP. Measuring family resilience of Chinese family caregivers: psychometric evaluation of the Family Resilience Assessment Scale. Fam Relat . (2021) 70:1297–302. doi: 10.1111/fare.12601

24. Li Y, Zhao Y, Zhang J, Lou F, Cao F. Psychometric properties of the shortened Chinese version of the Family Resilience Assessment Scale. J Child Fam Stud. (2016) 25:2710–17. doi: 10.1007/s10826-016-0432-7

25. Kaya M, Arici N. Turkish version of shortened Family Resiliency Scale (FRAS): the study of validity and reliability. Procedia Soc Behav Sci. (2012) 55:512–20. doi: 10.1016/j.sbspro.2012.09.531

26. Nadrowska N, Błazek M, Lewandowska-Walter A. Polish adaptation of the Family Resilience Assessment Scale (FRAS). Commun Ment Health J. (2021) 57:153–60. doi: 10.1007/s10597-020-00626-3

27. Burisch M. Test length and validity revisited. Eur J Pers. (1997) 11:303–15.

PubMed Abstract | Google Scholar

28. Walsh F. Family resilience: a framework for clinical practice. Fam Process. (2003) 42:1–18. doi: 10.1111/j.1545-5300.2003.00001.x

29. Benzies K, Mychasiuk R. Fostering family resiliency: a review of the key protective factors. Child Fam Soc Work. (2009) 14:103–14. doi: 10.1111/j.1365-2206.2008.00586.x

30. Black K, Lobo M. A conceptual review of family resilience factors. J Fam Nurs. (2008) 14:33–55. doi: 10.1177/1074840707312237

31. VanderWeele TJ. Love of neighbor during a pandemic: Navigating the competing goods of religious gatherings and physical health. Journal of Religion and Health. (2020) 59:2196–202. http://dx.doi.org/10.1007/s10943-020-01031-6

32. Meyer BD, Sullivan JX. The effects of welfare and tax reform: the material well-being of single mothers in the 1980s and 1990s. J Public Econ. (2004) 88:1387–420. doi: 10.1016/S0047-2727(02)00219-0

33. Stone HW, Cross DR, Purvis KB, Young MJ. A study of church nembers during times of crisis. Pastoral Psychol. (2004) 52:405–21. doi: 10.1023/B:PASP.0000020688.71454.3d

34. Henry CS, Sheffield Morris A, Harrist AW. Family resilience: moving into the third wave. Fam Relat . (2015) 64:22–43. doi: 10.1111/fare.12106

35. Izumi M, Gullón-Rivera AL. Family resilience among sojourning Japanese mothers: links to marital satisfaction and children's behavioral adjustment. Fam Consumer Sci Res J. (2018) 46:282–96. doi: 10.1111/fcsr.12247

36. Lester P, Stein JA, Saltzman W, Woodward K, MacDermid SW, Milburn N, et al. Psychological health of military children: longitudinal evaluation of a family-centered prevention program to enhance family resilience. Mil Med. (2013) 178:838–45. doi: 10.7205/MILMED-D-12-00502

37. Huang CY, Mu PF, Jou ST. Relationships between uncertainty, family resilience and quality of life in mothers of children with cancer. Int J Evid Based Healthc. (2011) 9:289. doi: 10.1097/01258363-201109000-00022

38. Garrard ED, Fennell KM, Wilson C. ‘We're completely back to normal, but I'd say it's a new normal': a qualitative exploration of adaptive functioning in rural families following a parental cancer diagnosis. Support Care Cancer . (2017) 25:3561–8. doi: 10.1007/s00520-017-3785-6

39. Mason W, Suri S. Conducting behavioral research on Amazon's Mechanical Turk. Behav Res Methods. (2012) 44:1–23. doi: 10.3758/s13428-011-0124-6

40. Berinsky AJ, Huber GA, Lenz GS. Evaluating online labor markets for experimental research: Amazon.com's Mechanical Turk. Political Anal. (2012) 20:351–68. doi: 10.1093/pan/mpr057

41. Schmidt S, Mühlan H, Power M. The EUROHIS-QOL 8-item index: psychometric results of a cross-cultural field study. Eur J Public Health. (2006) 16:420–8. doi: 10.1093/eurpub/cki155

42. Makowska Z, Merecz D, Moscicka A, Kolasa W. The validity of general health questionnaires, GHQ-12 and GHQ-28, in mental health studies of working people. Int J Occup Med Environ Health. (2002) 15:353–62.

43. Hendrick SS. A generic measure of relationship satisfaction. J Marr Fam. (1988) 50:93–8. doi: 10.2307/352430

44. Moos RR, Moss BS. Family Environment Scale Manual . Palo Alto, CA: Consulting Psychologist Press (1981).

45. Peterson NA, Speer PW, McMillan DW. Validation of A brief sense of community scale: confirmation of the principal theory of sense of community. J Commun Psychol. (2008) 36:61–73. doi: 10.1002/jcop.20217

46. Muthén LK, Muthén BO. Mplus User's Guide . 8th ed. Los Angeles, CA: Muthén and Muthén (1998–2017).

47. van de Schoot R, Lugtig P, Hox J. A checklist for testing measurement invariance. Eur J Dev Psychol. (2012) 9:486–92. doi: 10.1080/17405629.2012.686740

48. Chen FF. Sensitivity of goodness of fit indexes to lack of measurement invariance. Struct Equat Model Multidisc J. (2007) 14:464–504. doi: 10.1080/10705510701301834

49. Masten AS, Coatsworth JD. The development of competence in favorable and unfavorable environments: lessons from research on successful children. Am Psychol. (1998) 53:205–20. doi: 10.1037/0003-066X.53.2.205

50. Whitchurch GG, Constantine LL. Systems theory. In: Boss PG, Doherty WJ, LaRossa R, Schumm WR, Steinmetz KS, editors. Sourcebook of Family Theories and Methods: A Contextual Approach. New York, NY: Plenum Press (1993). pp. 325–55.

51. Chang W, Neo A, Fung D. In search of family resilience. Psychology. (2015) 6:1594–607. doi: 10.4236/psych.2015.613157

52. Hauser DJ, Schwarz N. Attentive Turkers: MTurk participants perform better on online attention checks than do subject pool participants. Behav Res Methods. (2016) 48:400–7. doi: 10.3758/s13428-015-0578-z

53. Zhou H, Fishbach A. The pitfall of experimenting on the web: how unattended selective attrition leads to surprising (yet false) research conclusions. J Pers Soc Psychol. (2016) 111:493–504. doi: 10.1037/pspa0000056

54. McCubbin HI, Thompson AI, McCubbin MA editors. Family Assessment: Resiliency, Coping, and Adaptation: Inventories for Research and Practice. 1st ed. Madison, MI: University of Wisconsin Publishers (1996).

55. Walsh F. Strengthening Family Resilience. 3rd ed. New York, NY: Guilford Press (2016).

Keywords: family resilience, short scale, Chinese, US, measurement validation

Citation: Chow TS, Tang CSK, Siu TSU and Kwok HSH (2022) Family Resilience Scale Short Form (FRS16): Validation in the US and Chinese Samples. Front. Psychiatry 13:845803. doi: 10.3389/fpsyt.2022.845803

Received: 30 December 2021; Accepted: 23 March 2022; Published: 13 May 2022.

Reviewed by:

Copyright © 2022 Chow, Tang, Siu and Kwok. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Tak Sang Chow, tschow@hksyu.edu

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

For more audio journalism and storytelling, download New York Times Audio , a new iOS app available for news subscribers.

• March 29, 2024   •   48:42 Hamas Took Her, and Still Has Her Husband
• March 28, 2024   •   33:40 The Newest Tech Start-Up Billionaire? Donald Trump.
• March 27, 2024   •   28:06 Democrats’ Plan to Save the Republican House Speaker
• March 26, 2024   •   29:13 The United States vs. the iPhone
• March 25, 2024   •   25:59 A Terrorist Attack in Russia
• March 24, 2024   •   21:39 The Sunday Read: ‘My Goldendoodle Spent a Week at Some Luxury Dog ‘Hotels.’ I Tagged Along.’
• March 22, 2024   •   35:30 Chuck Schumer on His Campaign to Oust Israel’s Leader
• March 21, 2024   •   27:18 The Caitlin Clark Phenomenon
• March 20, 2024   •   25:58 The Bombshell Case That Will Transform the Housing Market
• March 19, 2024   •   27:29 Trump’s Plan to Take Away Biden’s Biggest Advantage
• March 18, 2024   •   23:18 Your Car May Be Spying on You
• March 17, 2024 The Sunday Read: ‘Sure, It Won an Oscar. But Is It Criterion?’

Hamas Took Her, and Still Has Her Husband

The story of one family at the center of the war in gaza..

• Share full article

Hosted by Sabrina Tavernise

Produced by Lynsea Garrison and Mooj Zadie

With Rikki Novetsky and Shannon Lin

Edited by Michael Benoist

Original music by Marion Lozano ,  Dan Powell ,  Diane Wong and Elisheba Ittoop

Engineered by Alyssa Moxley

Listen and follow The Daily Apple Podcasts | Spotify | Amazon Music

Warning: this episode contains descriptions of violence.

It’s been nearly six months since the Hamas-led attacks on Israel, when militants took more than 200 hostages into Gaza.

In a village called Nir Oz, near the border, one quarter of residents were either killed or taken hostage. Yocheved Lifshitz and her husband, Oded Lifshitz, were among those taken.

Today, Yocheved and her daughter Sharone tell their story.

On today’s episode

Yocheved Lifshitz, a former hostage.

Sharone Lifschitz, daughter of Yocheved and Oded Lifshitz.

Background reading

Yocheved Lifshitz was beaten and held in tunnels built by Hamas for 17 days.

There are a lot of ways to listen to The Daily. Here’s how.

We aim to make transcripts available the next workday after an episode’s publication. You can find them at the top of the page.

Fact-checking by Susan Lee .

Additional music by Oded Lifshitz.

Translations by Gabby Sobelman .

Special thanks to Menachem Rosenberg, Gershom Gorenberg , Gabby Sobelman , Yotam Shabtie, and Patrick Kingsley .

The Daily is made by Rachel Quester, Lynsea Garrison, Clare Toeniskoetter, Paige Cowett, Michael Simon Johnson, Brad Fisher, Chris Wood, Jessica Cheung, Stella Tan, Alexandra Leigh Young, Lisa Chow, Eric Krupke, Marc Georges, Luke Vander Ploeg, M.J. Davis Lin, Dan Powell, Sydney Harper, Mike Benoist, Liz O. Baylen, Asthaa Chaturvedi, Rachelle Bonja, Diana Nguyen, Marion Lozano, Corey Schreppel, Rob Szypko, Elisheba Ittoop, Mooj Zadie, Patricia Willens, Rowan Niemisto, Jody Becker, Rikki Novetsky, John Ketchum, Nina Feldman, Will Reid, Carlos Prieto, Ben Calhoun, Susan Lee, Lexie Diao, Mary Wilson, Alex Stern, Dan Farrell, Sophia Lanman, Shannon Lin, Diane Wong, Devon Taylor, Alyssa Moxley, Summer Thomad, Olivia Natt, Daniel Ramirez and Brendan Klinkenberg.

Our theme music is by Jim Brunberg and Ben Landsverk of Wonderly. Special thanks to Sam Dolnick, Paula Szuchman, Lisa Tobin, Larissa Anderson, Julia Simon, Sofia Milan, Mahima Chablani, Elizabeth Davis-Moorer, Jeffrey Miranda, Renan Borelli, Maddy Masiello, Isabella Anderson and Nina Lassam.

Advertisement

Semiconductors at scale: New processor achieves remarkable speedup in problem solving

A nnealing processors are designed specifically for addressing combinatorial optimization problems, where the task is to find the best solution from a finite set of possibilities. This holds implications for practical applications in logistics, resource allocation, and the discovery of drugs and materials.

In the context of CMOS (a type of semiconductor technology), it is necessary for the components of annealing processors to be fully "coupled." However, the complexity of this coupling directly affects the scalability of the processors.

In a new IEEE Access study led by Professor Takayuki Kawahara from Tokyo University of Science, researchers have developed and successfully tested a scalable processor that divides the calculation into multiple LSI chips. The innovation was also presented in IEEE 22nd World Symposium on Applied Machine Intelligence and Informatics (SAMI 2024) on 25 January 2024.

According to Prof. Kawahara, "We want to achieve advanced information processing directly at the edge, rather than in the cloud or performing preprocessing at the edge for the cloud. Using the unique processing architecture announced by the Tokyo University of Science in 2020, we have realized a fully coupled LSI (Large Scale Integration) on one chip using 28nm CMOS technology. Furthermore, we devised a scalable method with parallel-operating chips and demonstrated its feasibility using FPGAs (Field-Programmable Gate Arrays) in 2022."

The team created a scalable annealing processor. It used 36 22nm CMOS calculation LSI (Large Scale Integration) chips and one control FPGA. This technology enables the construction of large-scale, fully coupled semiconductor systems following the Ising model (a mathematical model of magnetic systems) with 4096 spins.

The processor incorporates two distinct technologies developed at the Tokyo University of Science. This includes a "spin thread method" that enables 8 parallel solution searches, coupled with a technique that reduces chip requirements by about half compared to conventional methods. Its power needs are also modest, operating at 10MHz with a power consumption of 2.9W (1.3W for the core part). This was practically confirmed using a vertex cover problem with 4096 vertices.

In terms of power performance ratio, the processor outperformed simulating a fully coupled Ising system on a PC (i7, 3.6GHz) using annealing emulation by 2,306 times. Additionally, it surpassed the core CPU and arithmetic chip by 2,186 times.

The successful machine verification of this processor suggests the possibility of enhanced capacity. According to Prof. Kawahara, who holds a vision for the social implementation of this technology (such as initiating a business, joint research, and technology transfer), "In the future, we will develop this technology for a joint research effort targeting an LSI system with the computing power of a 2050-level quantum computer for solving combinatorial optimization problems."

"The goal is to achieve this without the need for air conditioning, large equipment, or cloud infrastructure using current semiconductor processes. Specifically, we would like to achieve 2M (million) spins by 2030 and explore the creation of new digital industries using this."

In summary, researchers have developed a scalable, fully coupled annealing processor incorporating 4096 spins on a single board with 36 CMOS chips. Key innovations, including chip reduction and parallel operations for simultaneous solution searches, played a crucial role in this development.

More information: Taichi Megumi et al, Scalable Fully-Coupled Annealing Processing System Implementing 4096 Spins Using 22nm CMOS LSI, IEEE Access (2024). DOI: 10.1109/ACCESS.2024.3360034

Provided by Tokyo University of Science