Multilevel Models in Family Research: Some Conceptual and Methodological Issues

Examining the impact of context on individual‐level outcomes has become an increasingly common undertaking in the social sciences. The growth in concern for identifying the effects of macrolevel characteristics has generated both theoretical and methodological advancements. In this issue of Journal of Marriage and Family, Butler (2002) researches whether the effect of welfare benefit levels on premarital childbearing varies by context, Hoffmann (2002) researches the effect of context on adolescent drug use, and Simons et al. (2002) examine how the relationship between parenting and child conduct varies by context. These articles are used as a background to discuss important theoretical and methodological issues surrounding the analysis of multilevel data. The authors present a simple analysis of data pertaining to age at first marriage taken from the Panel Study of Income Dynamics and merged with census data to measure contextual effects as a pedagogical device for introducing readers to the benefits of multilevel modeling.     

Generalized Linear Models in Family Studies

Generalized linear models (GLMs), as defined by J. A. Nelder and R. W. M. Wedderburn (1972) , unify a class of regression models for categorical, discrete, and continuous response variables. As an extension of classical linear models, GLMs provide a common body of theory and methodology for some seemingly unrelated models and procedures, such as the logistic, Poisson, and probit models, that are increasingly used in family studies. This article provides an overview of the principle and the key components of GLMs, such as the exponential family of distributions, the linear predictor, and the link function. To illustrate the application of GLMs, this article uses Canadian national survey data to build an example focusing on the number of close friends among older adults. The article concludes with a discussion of the strengths and weaknesses of GLMs.     

Longitudinal Associations Between Husbands' and Wives' Depressive Symptoms

Although concordance between husbands' and wives' mental health problems is often reported, questions remain about the nature of these relations. Extending research in this area, this study examined dynamic-longitudinal pathways among husbands' and wives' depressive symptoms and marital satisfaction as a moderator of associations. Participants were 296 heterosexual couples. Husbands and wives reported on their depressive symptoms and marital satisfaction. Results from dynamic bivariate latent difference score analyses indicated unidirectional longitudinal coupling such that higher levels of husbands' depressive symptoms predicted subsequent elevations in wives' depressive symptoms over time. This relation was stronger among couples reporting marital distress as compared to couples reporting higher marital satisfaction. The findings underscore the importance of considering one's spouses' depressive symptoms in treatment efforts.     

Secondary Data Analysis in Family Research

This article first provides an overview of the part that secondary data analysis plays in the field of family studies in the early 21st century. It addresses changes over time in the use of existing omnibus data sets and discusses their advantages and disadvantages. The second part of the article focuses on the elements that make a study a “family” study. It describes sources of existing data and the process of selecting a data set. The third part presents special issues and concerns in using existing data, and the fourth part addresses the future of secondary data analysis in the field of family research.     

Effects of Family Structure Type and Stability on Children's Academic Performance Trajectories

Using five waves of panel data from 8,008 children in the ECLS‐K, the current study compared children's academic performance growth curves from kindergarten through fifth grade among three types of nondisrupted and three types of disrupted families. The analyses found that children in nondisrupted two‐biological‐parent and nondisrupted stepparent households consistently made greater progress in their math and reading performances over time than their peers in nondisrupted single‐parent, disrupted two‐biological‐parent, and disrupted alternative families with multiple transitions. These trajectory differences were either partially or completely accounted for by family resources in the kindergarten year (Time 1). Overall, our findings provided strong support for the resource‐deprivation perspective and partial support for the instability‐stress perspective.     

Relationship of Dyadic Closeness With Work‐Related Stress: A Daily Diary Study

We examined the association between work‐related stress of both spouses and daily fluctuations in their affective states and dyadic closeness. Daily diary data from 169 Israeli dual‐earner couples were analyzed using multilevel modeling. The findings indicate that work stress has no direct effect on dyadic closeness but rather is mediated by the spouses’ negative mood. Evidence was found for spillover of stress from work to mood at home, as well as negative crossover among couples with higher marital quality, resulting in greater distance on stressful days. Such increased distance may reflect either a deleterious effect of work stress on marital relationships or a protective mechanism used by couples in times of stress.     

The Relationship Between Work-to-Family Conflict and Family-to-Work Conflict: A Longitudinal Study

The relationship between work-to-family conflict and family-to-work conflict was tested using a longitudinal research design with 234 dual-earner couples caring for both children and aging parents. Two waves of mailed survey data were collected. The mediating effects of role-related satisfaction were hypothesized to link these two forms of work-family conflict. The analytical steps for determining mediating effects followed suggestions by Baron and Kenny (1986). The results failed to show significant mediating effects of role-related satisfaction between the two forms of conflict over time for both husbands and wives, contrary to suggestions by Frone, Yardley, and Markel (1997). Instead, positive, direct relationships between the two types of work-family conflict were found. Limitations and implications are discussed.     

Families Under the Microscope: Repeated Sampling of Perceptions,Experiences, Biology,and Behavior

Families are not frozen dioramas; they are alive, active, and changing. Models of real‐life marital and family processes require many moving pieces. An intensive repeated measures approach reaches beyond static representations of the family toward more dynamic models that depict “life as it is lived.” The appearance of studies that use diaries and biological sampling in everyday life has burgeoned in the family research literature. These methods are part of a larger class of naturalistic methods that assess families in action and that includes direct video and audio recordings of families in their everyday environments. This article summarizes research that uses diaries, observations, and biological data collected over time in natural settings to study families. It provides an overview of the major research questions, designs, methods of data collection, and statistical models used in those literatures. Theoretical contributions and next steps in naturalistic research on families are discussed.     

Becoming a Parent and Relationship Satisfaction: A Longitudinal Dyadic Perspective

This study investigated the extent to which women's and men's relationship satisfaction within couples is similarly or differently affected by becoming a parent and the extent to which changes in work hours and hours spent on household labor affect a person's own and his or her spouse's relationship satisfaction across the transition to parenthood. The authors conducted longitudinal dyadic analyses, based on 12 waves of the British Household Panel Study (BHPS). They selected 689 couples who remained together during the period of observation and who were employed, childless, and living with their partner (of which 28% married) at the first moment of observation. The results revealed that relationship satisfaction of both members in a couple changed in tandem. Although work hours and household labor had some effect on people's own and their spouse's relationship satisfaction, these factors did not account for the U‐shaped relationship satisfaction pattern associated with the transition to parenthood.     

Handling Missing Values in Longitudinal Panel Data With Multiple Imputation

This article offers an applied review of key issues and methods for the analysis of longitudinal panel data in the presence of missing values. The authors consider the unique challenges associated with attrition (survey dropout), incomplete repeated measures, and unknown observations of time. Using simulated data based on 4 waves of the Marital Instability Over the Life Course Study (n = 2,034), they applied a fixed effect regression model and an event‐history analysis with time‐varying covariates. They then compared results for analyses with nonimputed missing data and with imputed data both in long and in wide structures. Imputation produced improved estimates in the event‐history analysis but only modest improvements in the estimates and standard errors of the fixed effects analysis. Factors responsible for differences in the value of imputation are examined, and recommendations for handling missing values in panel data are presented.     

When She Earns More Than He Does: A Longitudinal Study of Dual‐Earner Couples

In a random sample of 286 full‐time‐employed dual‐earner couples, we tested 3 competing hypotheses: when wives earn more than their husbands, (a) each partner's marital‐role quality (MRQ) decreases; (b) his MRQ increases, whereas effects on her MRQ are mixed; and (c) relationships vary with gender‐role beliefs (i.e., gender‐role ideology and subjective rewards of salary). We conceptualized salary as a couple‐level predictor with 4 components, 2 time varying and 2 time invariant, and estimated the relationship between 2 time‐varying components and MRQ. Women's MRQ was not significantly related to change in relative earnings. However, among men, the relationship varied by salary rewards.     

A Longitudinal Examination of Family Processes,Demographic Variables,and Adolescent Weight

Nationally representative studies estimate that almost one in five adolescents in the United States is overweight. This is a major concern for individuals' physical and psychological health and the overall economy in terms of health care costs and loss of productivity. A 12- to 14-year-old subsample of 4,688 adolescents from the NLSY97 cohort was used to address the association between family processes, demographic variables, and adolescent body mass index (BMI) percentile over 4 years. The final model indicated that frequency of family meals, gender, race, control, and mothers' BMI were important predictors of adolescent BMI percentile over time. Mothers' BMI was the strongest predictor of adolescent BMI percentile. More frequent family meals led to decreases in BMI percentile over time, whereas males, African Americans, and Latinos had higher average BMI percentiles than other groups.     

Differences in Family Norms for Men and Women Across Europe

The authors examined double standards regarding divorce among parents with young children and regarding having children within nonmarital cohabitation across Europe. Whether people disapprove more of men or of women engaging in these behaviors is not a priori clear. The authors formulated arguments in both directions and expected double standards to vary cross‐nationally by the level of socioeconomic gender equality in a country. They conducted multilevel analyses on a sample of about 44,000 individuals nested in 25 countries, obtained from the European Social Survey (2006). Double standards were measured with a split‐ballot design. The findings indicated that men were generally more disapproved of than women when displaying the family behaviors under study. Overall, women endorsed both double standards, whereas men endorsed only the double standard regarding divorce. However, substantial cross‐national differences in the double standards exist. The higher the level of socioeconomic gender equality, the larger these double standards in favor of women.     

Longitudinal Mediators of Relations Between Family Violence and Adolescent Dating Aggression Perpetration

Few longitudinal studies have examined the pathways through which family violence leads to dating aggression. In the current study the authors used 3 waves of data obtained from 8th‐ and 9th‐grade adolescents (N = 1,965) to examine the hypotheses that the prospective relationship between witnessing family violence and directly experiencing violence and physical dating aggression perpetration is mediated by 3 constructs: (a) normative beliefs about dating aggression (norms), (b) anger dysregulation, and (c) depression. Results from cross‐lagged regression models suggest that the relationship between having been hit by an adult and dating aggression is mediated by changes in norms and anger dysregulation, but not depression. No evidence of indirect effects from witnessing family violence to dating aggression was found through any of the proposed mediators. Taken together, the findings suggest that anger dysregulation and normative beliefs are potential targets for dating abuse prevention efforts aimed at youth who have directly experienced violence.     

Marital Conflict Behaviors and Implications for Divorce Over 16 Years

This study examined self‐reported marital conflict behaviors and their implications for divorce. Husbands and wives (N = 373 couples; 47% White American, 53% Black American) reported conflict behaviors in Years 1, 3, 7, and 16 of their marriages. Individual behaviors (e.g., destructive behaviors) and patterns of behaviors between partners (e.g., withdrawal‐constructive) in Year 1 predicted higher divorce rates. Wives' destructive and withdrawal behaviors decreased over time, whereas husbands' conflict behaviors remained stable. Husbands reported more constructive and less destructive behaviors than wives, and Black American couples reported more withdrawal than White American couples. Findings support behavioral theories of marriage demonstrating that conflict behaviors predict divorce and accommodation theories indicating that conflict behaviors become less negative over time.     

Ten Research Questions for Family Therapy

A growing evidence‐base shows that family therapy works, but many gaps in our knowledge remain about the conditions under which family therapy is effective and how it works. In this paper, ten critical research questions about family therapy that need to be addressed are considered. In short these are:

• 1 Is family therapy as effective in community settings as it is in specialist clinics?
• 2 For what problems is family therapy cost‐effective?
• 3 Does family therapy work for under‐researched problems and populations?
• 4 Do social‐constructionist and narrative approaches to family therapy work?
• 5 Can family therapy protocols be enhanced for non‐responders?
• 6 Can family therapy be combined with other psychotherapies to effectively treat specific problems?
• 7 Can family therapy be combined with pharmacotherapy to effectively treat specific problems?
• 8 What specific factors contribute to the effectiveness of family therapy with particular problems?
• 9 What common factors contribute to the effectiveness of family therapy?
• 10 What therapist and client factors contribute to the effectiveness of family therapy?

     

Predicting a Partner's End‐of‐Life Preferences,or Substituting One's Own?

Numerous studies comparing patients' end‐of‐life health care treatment preferences with their surrogates' reports of those preferences indicate that partners do not know one another's treatment preferences: Random guesses are just as likely as surrogate choices to match the patients' preferences. The present study uses the empathic accuracy model and the assumed similarity model to shed further light on accuracy and bias in surrogate reports. The data come from the National Center for Family and Marriage Research/Knowledge Networks Pilot Study 2010, a national sample of 1,075 heterosexual American couples age 18 to 64. Surrogate reports were simultaneously biased (i.e., correlated with the surrogate's own preference) and accurate (i.e., correlated with the partner's actual preference). Higher relationship quality and legal marriage (vs. cohabitation) were associated with increased assumed similarity bias. These findings suggest that practitioners working in end‐of‐life care need to be familiar with their partnered patients' relationship dynamics.