Economic hardship, family relationships, and adolescent distress: an evaluation of a stress-distress mediation model in mother-daughter and mother-son dyads.
Much of this process-oriented research has focused on how economic hardship is appraised and on intra-family relationships as important intervening variables in the link between economic hardship and physical and psychological distress. Lazarus and his associates (Lazarus, 1991, 1993; Lazarus & Folkman, 1984) have theoretically elaborated and empirically documented the role of appraisal processes in the stress, coping, and distress relationship. With respect to economic hardship specifically, several recent studies point out that differences in how objective economic conditions are perceived account for variations in resulting distress (Conger et al., 1992, 1993; Lempers et al., 1989).
The two intra-family relationships that have received the most attention from researchers interested in documenting the effects of economic hardship on families are the parent-child relationship and the marital relationship. An abundance of research has documented the overall importance of the parent-child relationship for the child's development (Baumrind, 1971; Belsky, 1984; 1990; Maccoby & Martin, 1983).
Recent studies on the impact of economic hardship on children have demonstrated that some of the negative effects are mediated through changes in the parent-child relationship. Elder, Van Nguyen, and Caspi (1985) found that economic hardship increased children's socioemotional distress by increasing punitive and arbitrary parenting behaviors, especially of the father. Harold-Goldsmith, Radin, and Eccles (1988) found that, although unemployed fathers had more time for child care, they displayed fewer nurturing behaviors than did other fathers. Lempers et al. (1989) observed that under economic hardship, parental nurturance decreased and inconsistent discipline increased. Conger et al. (1992, 1993) showed that economic pressure had an effect on adolescent adjustment by increasing parents' depression, which was associated with less involved parenting. Findings by Galambmos and Silbereisen (1987), Flanagan, 1988), and Larson (1984) all have indicated that parents facing economic hardship felt more depressed about the future of their children, felt less competent in helping their children choose future careers, tended to lower their expectations for their children's education, and were less likely to encourage them to finish college. These lowered parental expectations were associated with decreased academic aspirations in the children, who expected to undergo vocational training instead of attending a four-year college (Isralowitz & Singer, 1986; McLoyd, 1990). Flanagan (1990) showed that adolescents in families experiencing job loss reported more conflict with their parents.
In the current study, depression and loneliness were selected as indicators of adolescent distress because of the previously documented effects of economic hardship on the parents and the parent-child relationship. Parents who, because of financial pressure, become more depressed, more irritable, and/or more self-preoccupied might be less nurturant and supportive in their daily interactions with their children, and perhaps more distant, uninvolved, and rejecting on a daily basis (Maccoby & Martin, 1983; Patterson, 1982). These stress-induced parental moods and behaviors, and the resulting negative parent-child interactions, may be related to their children feeling depressed, less wanted, and more lonely. Further, as Jones (1988) reported, children from economically deprived families tended to feel isolated, to have had conflictual relationships with peers, and to be suspicious of children from social classes different from their own. Hook (1990) found that adolescents lacked identification with peers as a result of their inability to discuss with 'anyone in their communities their families' financial status, for which they felt partly responsible.
Research also documents the negative effects of economic hardship on marital quality and stability (Voydanoff & Majka, 1988). Studies have noted the effects of the quality of the marital relationship on child outcomes (Emery, 1988; Reid & Crisafulli, 1990). Most studies have tended to support the generalization that a decrease in the quality of the marital relationship is correlated with an increase in child disturbances (Dadds, 1987). Several studies have indicated that the marital relationship mediates the effect of economic hardship on child outcomes (Conger et al., 1992, 1993; Ge et al., 1992). According to McLoyd (1989, 1990), financially pressed parents might be less likely to be affectionate with and respectful of each other and even be more hostile.
Children exposed to more frequent instances of marital conflict may not be able to deal with such conflict and may react emotionally to the stress of the marital conflicts with feelings of distress. Both theoretical and empirical work have emphasized the importance of parental financial strain, the marital relationship, and the parent-child relationship as process variables in the relationship between economic hardship and child distress. Much theorizing has focused on how the marital and the parent-child relationship influence each other in this process. Several authors (Baumrind, 1967; Fauber, Forehand, McCombs Thomas, & Wierson, 1990; Margolin, 1981; Tschann, Johnston, Kline & Wallestein, 1989) have argued that the marital relationship mediates between family economic hardship and child outcomes by disturbing the parent-child relationship. The strain resulting from economic hardship might result in a more conflictual marital relationship which, in turn, could lead to less involved parenting. Some findings have provided evidence to support that hypothesis (Simons, Lorenz, Conger, & Wu, 1992).
Most of the studies investigating the relationship among economic hardship, family relationships, and child behaviors have used composite variables that combine information either across mothers and fathers and/or across daughters and sons. Several studies indicate that, during early adolescence, the dynamics of the parent-child interaction might be different for mother-son, mother-daughter, father-son, and father-daughter dyads and that the effects of economic hardship on children as mediated by the marital and the parent-child relationships might vary for these different dyads (Conger et al., 1992, 1993; Youniss & Smollar, 1985). This study focused on the relationships among economic hardship, maternal financial strain, maternal marital happiness, the mother-child relationship, and distress in female and male adolescents. The hypotheses evaluated are those contained in what has been called a family mediation model (Coyne & Downey, 1991; Elder et al., 1985). Within this model, stressful family conditions are assumed to result in child distress primarily through the disrupting effects of these conditions on the quality of the marital and parent-child relationships within the stressed families. Consistent with the family mediation model, it was predicted that increased economic hardship is likely to lead to more maternal financial strain, which may indirectly result in more adolescent distress by decreasing the quality of the marital and of the parent-child relationships. Because of the financial strain mothers are experiencing, their marital relationship may become less happy and satisfying to them; they may become more argumentative with their spouses and have more quarrels and conflicts. A second consequence of the increased maternal strain might be that her relationship with her child might become less nurturant and supportive. These changes in the marital and the parent-child relationships have the potential to contribute to more adolescent distress.
A sample of 188 sixth and 210 eighth graders and their families participated in this study. Of the 188 sixth graders, 96 were female and 92 male; of the 210 eighth graders, 104 were female and 106 were male. This group was recruited for a three-wave longitudinal study on stress, support, and distress. The parents of all sixth and eighth graders of 27 school districts with a K through 6 (elementary school), 7 through 8 (junior high), and 9 through 12 (senior high) school structure in a midwestern state were contacted by mail through the schools. The 27 selected districts represented a random sample, stratified by size (large versus small) and location (northeast, northwest, southeast, southwest), of all the state's public school districts with the previously described structure, the most common school organization in the state.
The letter sent to the parents described the general nature of the longitudinal research project, which involved three family contacts over three years, requested their voluntary participation as a family, and informed them that they would be reimbursed for their participation in the study in the amount of $75.00 for each family visit.
Together with the letter describing the study was a short questionnaire to ascertain if the family met the criteria for participation, a form on which the family was to indicate willingness or refusal to participate, and a stamped return envelope. Criteria for participation in the study were that the target child was a sixth or eighth grader, there were two parents in the home, and the target child had a sibling within three years of age. Families who returned the envelope, were willing to participate, and seemed eligible were contacted by phone, and an appointment was made to visit each family. During that home visit, the interviewer provided and reviewed a written description of the project with all participating family members so each of them had a clear understanding of the requirements of the project. If the family was still willing to participate, family members and the interviewer signed a dated statement of informed consent.
Of the 726 families screened, 464 were eligible for the study. Of those, 398 families agreed to participate. In each family, only one child could be a target child.
The sixth graders ranged in age from 11 through 13 years with a mean of 11.4 years; the eighth graders varied in age from 12 through 14 with a mean of 13.4 years. Of the 398 families, 48% had been farming since at least 1983; and 52% were nonfarm families. They were from 22 agriculture-dependent counties typical of the many counties that experienced economic stress during the 1980s (Jolly, 1986). Based on the parents' current occupation and highest grade completed in school, socioeconomic status was determined by using Hollingshead's (1957) procedure, which showed that most were middle- and working-class families.
The interviewers for the project were recruited by the Survey Section of the Statistical Laboratory at Iowa State University. All 12 interviewers involved had experience working on similar projects. The Survey Section conducted training workshops with the interviewers for this particular project to ensure uniformity of interview administration procedures.
During the first home visit, after the interviewer had reviewed the project and the participating families had signed the informed consent, the interviewer instructed the mother, the father, the target child, and the sibling on how to proceed in responding to their questionnaires. Each was asked to carefully read the instructions accompanying the questionnaire and, if needed, to request clarification from the interviewer.
Each participant in the family was asked to go to a separate room, if possible, to complete his or her questionnaire. After each participant finished, the interviewer checked to make sure all questions had been answered and, if not, requested the respondent to provide his or her best answer. The interviewer then thanked the family for their participation, provided them with the reimbursement, and informed them that the Survey Section would contact them for two more visits. Total time needed for the visit varied an hour to an hour and a half.
This study presents some of the findings of the first wave of data collection for the 200 mother-daughter and the 198 mother-son dyads. Each mother-child dyad was independent because no family provided more than one target child.
Economic hardship. Mothers responded to questions dealing with total family income, change in family income, and change in family employment. For total family income, they were asked to include all sources (e.g., earned income, investments, social security) before taxes for all members of the family. The numbers provided were adjusted for the number of family members. Change in family income was assessed by having the mothers provide information on if, and how much, their total family income had changed over the past year and the past three years. The set of six response alternatives to both questions varied from 1 = increased more than 25% to 5 = decreased more than 25%. Change in family employment was assessed by three questions. The mothers were asked whether during the year preceding the interview (1) they or a family member had taken additional employment to help meet expenses, (2) they or a family member had taken a second or third job, and (3) a family member who was not working had taken a job, all to help meet family expenses. Response alternatives to these three questions consisted of a simple yes or no, with a higher score indicating more additional employment.
The total family income measure adjusted for number of family members, the change in family income measure, and the change in family employment measure served as manifest indicators of the latent construct of economic hardship in the latent variable path analysis.
Financial strain was assessed by two measures. On the first, the mothers were asked how much difficulty they had in paying their bills (the 4 response alternatives were 1 = no difficulty, 2 = a little difficulty, 3 = some difficulty, and 4 = a great deal of difficulty), and how much money they had left over at the end of the month (the 3 response alternatives were 1 = some money left over, 2 = just enough money to make it, and 3 = not enough money to make it).
For the second measure of financial strain, the mothers answered eight questions that asked if they agreed of disagreed, on a 5-point scale ranging from 1 = strongly agree through 5 = strongly disagree, that their family could afford a suitable house; was able to afford furniture or household equipment that needed replacement; could buy the kind of car needed; had enough money for the kind of food, clothing, medical care, and leisure activities, respectively, the family should have; and if their money seemed to be enough for what they wanted. These items were a subset of those of the Economic Strain Questionnaire (Pearlin, Lieberman, Menaghan, & Mullan, 1981). Coefficient alpha for this questionnaire was 0.88. The mothers' responses to the two measures served as manifest indicators of the latent variable of financial strain in the latent variable path analysis.
Mothers' marital relationship was assessed by seven questions taken from the Spanier (1976) Dyadic Adjustment Scale and the Booth, Johnson, and Edwards (1983) Marital Instability Index. The first two questions dealt with the mothers' degree of happiness with the marital relationship and with their perception of their partners' degree of happiness with the marital relationship as assessed on a 7-point scale from 0 = extremely unhappy through 6 = extremely happy.
The remaining five questions asked the mothers whether (1) they ever thought their marriage to be in trouble, (2) the thought of getting a separation or divorce had crossed their minds, (3) they had discussed separation or divorce with a close friend, (4) they or their spouses ever seriously had suggested the idea of divorce, and (5) they and their spouses had talked about consulting an attorney about a possible divorce or separation. Each mother responded using the following 4-point scale: 1 = never, 2 = yes, prior to the last 3 years, 3 = yes, within the last 3 years, or 4 = yes, within the last 3 months. Coefficient alpha for the marital relationship questionnaire equalled 0.87.
A principal axis factor analysis with varimax rotation of the responses of the mothers to these seven marital relationships items indicated the existence of two factors, one on which the two happiness items had high loadings, the other on which the five separation and divorce items had high loadings. The two factors accounted for 65% of the variance in the mothers' responses. The responses of the mothers to the two happiness items and to the five separation and divorce items served as two manifest indicators of the latent construct of marital happiness in the path analysis.
The mother-child relationship was assessed by the sixth- and eighth-graders' responses to the Furman and Buhrmester's (1985) Network of Relationship Inventory (NRI). Each male and female adolescent indicated (on a 5-point scale ranging from 1 = not at all, 2 = a little, 3 = some, 4 = a lot, to 5 = very, very much) how much each of nine attributes occurred in their relationships with their mothers. The nine attributes are: being admired by the parent, feeling loved, having a companion, frequency of conflict, being intimate, receiving tangible aid, being nurturing toward parent, having an enduring relationship, and degree of satisfaction with the relationship with the parent.
Each attribute was assessed by three items, for a total of 27 items. Coefficient alpha for this mother-child relationship questionnaire was 0.94. The children's responses to these 27 questions were subjected to a principal-axis factor-analysis with varimax rotation. The results indicated the presence of two factors, accounting for 46% of the variance. The first was composed of the items assessing being admired and loved by the mother and having a satisfying and enduring relationship with her, and seemed to tap affectionate aspects of the relationship. The second factor on which the tangible aid and the companionship items loaded, seemed to tap instrumental aspects of the relationship. The responses of the male and female adolescents to those items having high (0.45 or higher) and unambiguous (no high loadings on other factors) loadings on these two factors functioned as two manifest indicators for the parent-child relationship construct in the latent variable path analysis.
Adolescent distress was measured by administering a loneliness and a depression scale. For loneliness, 12 items of the Asher, Hymel, and Renshaw (1984) loneliness scale were used. Coefficient alpha for this measure equalled 0.91. For depression, a modification of the Beck Depression Inventory (1967) was administered in which the response format of the inventory was modified to be identical to the format of the other questionnaires used. Cronbach alpha for this modified inventory equalled 0.75. The loneliness and depression measures served as two manifest indicators of the latent construct of adolescent distress in the path analysis.
Latent variable path analysis with partial least-squares estimation procedures (LVPLS) was used to analyze the data (Falk & Miller, 1992). LVPLS is a structural equation modelling technique that can be used when the data do not meet the more stringent assumptions of such structural equation techniques as LISREL (Falk & Miller, 1992; Fornell & Bookstein, 1982).
Falk and Miller (1992) recommend several LVPLS indices to evaluate how well the model fits the data. To evaluate the adequacy of the outer or measurement model, which specifies the relationships between the manifest variables and their respective latent constructs, they recommend that the factor loading of each individual manifest variable on its respective latent construct should be 0.55 or higher, and that a value of 0.30 or less for the mean communality coefficient, the average of all squared factor loadings of all manifest variables on their respective latent constructs, would be too low to be acceptable; the higher the latter value, the better the measurement model.
To assess the inner model, which specifies the relationships among the exogenous and endogenous latent constructs, it is recommended that each path coefficient which is a standardized regression coefficient or beta weight, should equal at least 0.10 (Falk, 1987; Falk & Miller, 1992). Falk and Miller (1992) also suggest calculating the arithmetic average of the multiple [R.sup.2] for all endogenous constructs. A multiple [R.sup.2] can vary from zero to one; the larger the average multiple [R.sup.2], the better the prediction of the endogenous constructs in the model.
To evaluate the overall model, Falk and Miller (1992) recommend the coefficient RMS COV (E, U) which stands for the root mean square of the covariance between the residuals of the manifest variables and the latent constructs. ARMS COV (E, U) coefficient of 0 would indicate a perfect fit between the overall model and the raw data; a coefficient of .20 or higher would be indicative of an inadequate fit, and a coefficient of .02 would be indicative of a superior fitting overall model. These indices were used in the present study to describe how the hypothesized model as presented in Figure 1 fits the data.
Figure 1 shows the results for the mother-daughter and the mother-son dyads. All individual factor loadings are 0.64 or higher, well surpassing the minimum 0.55 value recommended by Falk and Miller (1992). The mean communality coefficient, which can vary between 0 and 1, equals 0.73 for the mother-daughter dyads and 0.72 for the mother-son dyads. Both the individual factor loadings and the mean communality coefficient indicate that the outer or measurement model is adequate and acceptable.
The values of the path coefficients are also specified in Figure 1. The coefficient of the path from economic hardship to financial strain equals 0.75 in the mother-daughter dyads and 0.59 in the mother-son dyads. The coefficients of the path from financial strain to marital happiness and from financial strain to the mother-child relationship equal, respectively, -0.31 and -0.14 in the mother-daughter dyads, and -0.26 and -0.14 in the mother-son dyads. It can be seen that the coefficients of the path from mothers' marital happiness to adolescent distress for the mother-daughter and mother-son dyads, -0.08 and -0.07, respectively, are too low to be acceptable. The coefficients of the path from the parent-child relationship to adolescent distress equal, respectively, -0.42 and -0.35 in the mother-daughter and mother-son dyads.
A total of 18% of the variance in daughters' distress and of 14% of the variance in sons' distress was accounted for in the model that specified direct paths from mothers' marital happiness and from the mother-child relationship to adolescent distress. The mean [R.sup.2] for the four endogenous constructs equals 0.22 for the mother-daughter dyads and 0.15 for the mother-son dyads. The root mean square of the covariance between the residuals of the manifest variables and the residuals of the latent variables, RMS, COV (E, U), equals 0.03 for the mother-daughter dyads and 0.04 for the mother-son dyads, well below the 0.20 value indicating an inadequate model and close to the 0.02 value indicating a superior model (Falk & Miller, 1992).
Three alternative models were evaluated. All three are less parsimonious than the original model because they have either one or two more paths than the original model. The first alternative model added a path from financial strain to adolescent distress. The model thus includes the hypothesis that there is a direct effect of maternal financial strain on children's distress. The addition of this path, which had a coefficient of 0.02 in both mother-daughter and mother-son dyads, did not make this first alternative a better model than the original: there was a one percentage point increase in the variance accounted for in daughters' distress but no change in the percentage of variance accounted for in sons' distress. The average [R.sup.2] for the endogenous constructs did change 0.25 for the mother-daughter dyads as a result of the one percentage point increase in variance accounted for in daughters' distress; the average [R.sup.2] did not change for the mother-son dyads. There was no decrease in the value of the RMS (E, U) coefficient for either dyad.
The second alternative model added a path from mothers' marital happiness to the parent-child relationship. The coefficient of this path was -0.03 for the mother-daughter dyads and did not change the average [R.sup.2] or the RMS (E, U) coefficient. The coefficient of this path was 0.16 for the mother-son dyads; the variance accounted for in the mother-son relationship construct increased by two percentage points, increasing the average [R.sup.2] by half a percentage point. The RMS (E, U) coefficient decreased from 0.31 in the original model to 0.30 in this alternative model. These changes do not indicate that this alternative model is substantially better than the original.
The third alternative model added two paths: the path from financial stress to adolescent distress, as in the first alternative model, and the path from mothers' marital happiness to the mother-child relationship, as in the second alternative model. For the mother-daughter dyads, the coefficients of these two paths were 0.02 and -0.03, respectively. The addition of these two paths did not increase for the mother-daughter dyads, the average [R.sup.2] of the original hypothesized model nor did it decrease the RMS (E, U) coefficient found in the hypothesized model. For the mother-son dyads, the coefficient of the path from financial stress to adolescent male distress was 0.02, and the coefficient of the path from mothers' marital happiness to the mother-son relationship was 0.16. The variance accounted for in the mother-son relationship variable increased by two percentage points from its value in the original model, and the average [R.sup.2] for all endogenous constructs in this model increased by just half a percentage point. The RMS (E, U) coefficient did not change. Again, these results indicate that the third alternative model is not better than the original.
It was predicted that increasing economic hardship would lead to more maternal financial strain, which would ultimately and indirectly result in more adolescent distress. The effect of financial strain on adolescent distress was expected to be indirect and to come about through strain-induced changes in the marital and parent-child relationship. Because of the financial strain the mothers were experiencing, the marital relationship might have become less happy and satisfying to them; they could have become more argumentative with their spouses, possibly resulting in more quarrels and conflicts. A second consequence of the increased strain could have been that the mothers might have become less nurturant and supportive with their children. Both changes could have contributed to more adolescent distress. Changes in the marital relationship could have exposed the child to more instances of marital quarrels and conflicts, which might have been a source of stress for the child and lead to more feelings of distress. Changes in the mother-child relationship might have made the child feel less wanted, less loved and, therefore, more distressed.
The evidence in the present study provides support for some aspects of the hypothesized model. A strong association between economic hardship and maternal financial strain was found. Similar findings have been reported in other studies (Conger et al., 1991, 1992, 1993; Ge et al., 1992; Kessler, Turner, & House, 1989). This result of a strong association between economic hardship, more objectively measured, and perceived financial strain is consistent with the argument by Lazarus and Folkman (1984) that stressors affect people to the extent that they are experienced as strain; that is to say, are "appraised [by families] as taxing or exceeding [the family's] resources and endangering [the family's] well-being" (p. 19).
The path-analytic results showed that the effect of maternal financial strain on adolescent distress was mediated through changes in the mother-adolescent child relationship for both the mother-son and mother-daughter dyads. More financial strain in the mothers led to a decrease in the quality of the mother-child relationship, as perceived by the children, and this resulted in more distress for both sons and daughters. These results for the mother-daughter and the mother-son dyads are consistent with findings of several recent studies (Clark-Lempers et al., 1990; Conger et al., 1992; 1993; Elder et al., 1985; Lempers et al., 1989). Economic and other stressors increasing the strains on parents are likely to disrupt the parent-child relationship; it is these disruptions that seem to cause distress in children.
Of interest in the present study were issues concerning the marital and the mother-child relationships. One issue concerns the nature of the association between mothers' marital happiness and the mother-child relationship. There is a diversity of theoretical viewpoints on this question (Hinde & Stevenson-Hinde, 1988). One viewpoint, primarily found in the empirical literature, argues for a positive association between these two relationships: a good marital relationship fosters a good parent-child relationship (Belsky, 1984, 1990; Engfer, 1988). Another viewpoint, more prevalent in the clinical literature, argues for a negative association between the marital and the parent-child relationship: less satisfying marital bonds are compensated for by more involved parent-child bonds (Christensen & Margolin, 1988; Engfer, 1988). Most studies tend to support the first hypothesis, but it is unclear whether the path of influence is unidirectional, from marital to parent-child relationship, or bidirectional and reciprocal. The evaluation of the second alternative model showed that in the present study the association between mothers' marital happiness and the mother-child relationship was positive but weak in the mother-son dyads and nonexistent in the mother-daughter dyads.
Another issue concerns whether the marital relationship affects child outcomes directly, indirectly, or both. The assumption of an indirect effect presupposes the existence of an association between the marital and the parent-child relationships, which goes back to the first issue. The indirect effect of the marital relationship on child outcomes would come about through the parent-child relationship. Several studies provide evidence for an indirect effect model. Few studies, however, explicitly evaluate direct effect only, indirect effect only, and combined direct and indirect effects models in a nested-sequence model comparison. Ge and colleagues (1992), comparing a combined direct and indirect effects model with an indirect effect only model, found that the combined model did not provide a significant improvement over the indirect effect model. Fauber and colleagues (1990) found that most of the impact of marital conflict on child outcomes could be explained by the effect of the marital conflict on the quality of the parent-child relationship, supporting an indirect or mediational model. In one of their samples, however, they also found evidence for a direct path from marital conflict to children's externalizing problems. Comparing direct, indirect, and combined effects is important for the theoretical explication of the etiological mechanisms through which the marital relationship may affect children. Indirect or mediational and direct effect hypotheses point to different mechanisms. For example, Patterson (1982) in formulating a mediational hypothesis has suggested that marital conflict might lead to lax parenting that might then result in antisocial behavior by the child. Direct effect hypotheses emphasize that exposure to marital conflict is itself a stressor for children (Emery, 1988) or that this exposure shows children destructive models of conflict resolution and teaches them aggression (Porter & O'Leary, 1980).
The expectation in the present study was to find both direct and indirect effects of the marital relationship, but the mechanisms involved were thought to be different. For that reason we evaluated the original model, in which there was a path from marital happiness to child distress but no path from marital happiness to the mother-child relationship, against the alternative model, in which both paths were present. Mothers faced with financial problems and with the challenge these problems pose for their marital relationship might pay less attention to their children. This might make the children feel less wanted and, therefore, more distressed. The strain on the marital relationship would thus change the mother-child relationship for the worse, which might result in more child distress, an indirect effect. However, the exposure to increased instances of conflict between their mothers and fathers might. also have a direct effect; children might feel more distressed because they do not know how to deal with the conflicts they witness.
The results of the present study showed no direct association between mothers' marital happiness and their children's distress. Therefore, mothers' marital happiness did not mediate between mothers' financial strain and their children's distress directly. However, the data did show a positive association between mothers' marital happiness and their relationship with their sons, but not with their daughters. These data indicate that, in addition to the indirect path of influence from mothers' financial strain to the mother-son relationship to the sons' distress, there also is an indirect path of influence from mothers' financial strain to mothers' marital happiness to the mother-son relationship to the sons' distress. This latter finding, together with the finding of no direct association between mothers' marital happiness and sons' distress, seems consistent with the viewpoint of those who argue that the marital relationship influences child outcomes through the effects of the marital relationship on the parent-child relationship. However, as already indicated, this finding was obtained only in the mother-son dyads.
This research was made possible by a grant from the National Institute of Mental Health (grant 1 R01 MH42858) to both authors, and by grants from the Research Institute of the College of Family and Consumer Sciences and the Iowa Agriculture and Home Economics Experiment Station, Iowa State University, to the first author. The authors thank the families who volunteered for this project.
Asher, S. R., Hymel, S., & Renshaw, P. D. (1984). Loneliness in children. Child Development, 55, 1456-1464.
Baumrind, D. (1967). Child care practices anteceding three patterns of preschool behavior. Genetic Psychology Monographs, 75, 43-88.
Baumrind, D. (1971). Current patterns of parental authority. Developmental Psychology Monographs, 4(1, Pt. 2).
Beck, A. T. (1967). Depression: Causes and treatment. Philadelphia: University of Pennsylvania Press.
Belsky, J. (1984). The determinants of parenting: A process model. Child Development, 55, 83-96.
Belsky, J. (1990). Parental and nonparental child care and children's socioemotional development: A decade in review. Journal of Marriage and the Family, 52, 885-903.
Booth, A., Johnson, D., & Edwards, J. N. (1983). Measuring marital stability. Journal of Marriage and the Family, 38, 387-394.
Christensen, A., & Margolin, G. (1988). Conflict and alliance in distressed and nondistressed families. In R. A. Hinde, & J. Stevenson-Hinde (Eds.), Relationship within families: Mutual influences (pp. 263-282). New York: Oxford University Press.
Clark-Lempers, D. S., Lempers, J. D., & Netusil, A. J. (1990). Family financial stress, parental support, and young adolescents' academic achievement and depressive symptoms. Journal of Early Adolescence, 10, 21-36.
Conger, R. D., Conger, K. J., Elder, G. H., Lorenz, F. O., Simons, F. L., & Whitebeck, L. B. (1992). A family process model of economic hardship and adjustment of early adolescent boys. Child Development, 63, 526-541.
Conger, R. D., Conger, K. J., Elder, G. H., Lorenz, F. O., Simons, R. L., & Whitebeck, L. B. (1993). Family economic stress and adjustment of early adolescent girls. Developmental Psychology, 29, 206-219.
Conger, R. D., Lorenz, F. O., Elder, G. H., Melby, J. N., Simons, R. L., & Conger, K. J. (1991). A process model of family economic pressure and early adolescent alcohol use. Journal of Early Adolescence, 11, 430-449.
Coyne, J. C., & Downey, C. (1991). Social factors and psychopathology: Stress, social support, and coping processes. In M. R. Rosenzweig, & L. W. Porter (Eds.), Annual Review of Psychology (Vol. 42, pp. 401-425). Palo Alto, CA: Annual Reviews.
Dadds, M. R. (1987). Families and the origins of child behavior problems. Family Process, 26, 341-357.
Elder, G., Van Nguyen, T., & Caspi, A. (1985). Linking family hardship to children's lives. Child Development, 56, 361-375.
Emery, R. E. (1988). Marriage, divorce, and children's adjustment. Newbury Park, CA: Sage.
Engfer, A. (1988). The interrelatedness of marriage and the mother-child relationship. In R. A. Hinde, & J. Stevenson-Hinde (Eds.), Relationships within families: Mutual influences (pp. 104-118). Oxford: Clarendon Press.
Falk, R. F. (1987). A primer for soft modeling. Berkeley: University of Berkeley Press.
Falk, R. F., & Miller, N. B. (1992). A primer for soft modeling. Akron, OH: The University of Akron Press.
Fauber, R., Forehand, R., McCombs Thomas, A., & Wierson, M. (1990). A mediational model of the impact of marital conflict on adolescent adjustment in intact and divorced families: The role of disrupted parenting. Child Development, 61, 1112-1123.
Flanagan, C. A. (1988). The effects of a changing economy on the socialization of children's academic and vocational aspirations. Paper presented at the meetings of the American Educational Research Association, New Orleans.
Flanagan, C. A. (1990). Change in family work status: Effects on parent-adolescent decision making: Child Development, 61, 163-177.
Fornell, C., & Bookstein, F. L. (1982). A comparative analysis of two structural equation models: LISREL and PLS applied to market data. In C. Fornell (Ed.), A second generation of multivariate analysis (pp. 289-394). New York: Praeger.
Furman, W., & Buhrmester, D. (1985). Children's perceptions of the personal relationships in their social networks. Developmental Psychology, 21, 1016-1024.
Galambos, N. L., & Silbereisen, R. K. (1987). Income change, parental life outlook, and adolescent expectations for job success. Journal of Marriage and the Family, 48, 309-318.
Ge, X., Conger, R. D., Lorenz, F. O., Elder, G. H., Montague, R. B., & Simons, R. L. (1992). Linking family economic hardship to adolescent distress. Journal of Research on Adolescence, 2, 351-378.
Harold-Goldsmith, R., Radin, N., & Eccles, J. S. (1988). Objective and subjective reality: The effects of job loss and financial stress on fathering behaviors. Family Perspective, 22, 309-326.
Hinde, R. A., & Stevenson-Hinde, J. (Eds.). (1988). Relationships within families. Oxford: Clarendon Press.
Hollingshead, A. B. (1957). Four factor index of social status. Unpublished manuscript, Department of Sociology, Yale University, New Haven, Connecticut.
Hook, Van, M. (1990). The Iowa farm crisis: Perceptions, interpretations, and family patterns. In V. C. McLoyd, & C. A. Flanagan (Eds.), Economic stress: Effects on family life and child development. New directions for child development (Vol. 46, pp. 71-86). San Francisco, CA: Jossey-Bass.
Isralowitz, R., & Singer, M. (1986). Unemployment and its impact on adolescent work values. Adolescence, 21, 145-158.
Jolly, B. (1986). Survey of Iowa farm operators. Ames, Iowa: Iowa State University.
Jones, L. D. (1988). The effect of unemployment on children and adolescents. Children and Youth Services Review, 10, 199-218.
Kessler, R. C., Turner, J. B., & House, J. S. (1989). Unemployment, reemployment, and emotional functioning in a community sample. American Sociological Review, 54, 648-657.
Larson, J. (1984). The effect of husband's unemployment on marital and family relations in blue-collar families. Family Relations, 33, 503-511.
Lazarus, R. S. (1991). Emotion and adaptation. New York: Oxford University Press.
Lazarus, R. S. (1993). Psychological stress and the emotions. In L. W. Porter, M. R. Rosenzweig (Eds.), Annual Review of Psychology (Vol. 44, pp. 1-21). Palo Alto, CA: Annual Reviews.
Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer.
Lempers, J. D., & Clark-Lempers, D. (1990). Family economic stress, maternal and paternal support and adolescent distress. Journal of Adolescence, 13, 217-229.
Lempers, J. D., Clark-Lempers, D., & Simons, R. L. (1989). Economic hardship, parenting, and distress in adolescence. Child Development, 60, 25-39.
Maccoby, E. E., & Martin, J. A. (1983). Socialization in the context of the family: Parent-child interaction. In E. M. Hetherington (Ed.), P. H. Mussen (Series Ed.), Handbook of child psychology: Vol. 4, Socialization, personality, and social development (pp. 1-101). New York: Wiley.
Margolin, G. (1981). The reciprocal relationship between marital and child problems. In J. P. Vincent (Ed.), Advances in family intervention, assessment, and theory (Vol. 2, pp. 131-182). Greenwich, CT: JAI.
McLoyd, V. C. (1989). Socialization and development in a changing economy: The effects of paternal job and income loss on children. American Psychologist, 44, 293-302.
McLoyd, V. C. (1990). The impact of economic hardship on black families and children: Psychological distress, parenting, and socioemotional development. Child Development, 61, 311-346.
McLoyd, V. C., & Flanagan, C. A. (1990). Economic stress: Effects on family life and child development. New Directions for Child Development, 46, 71-86.
Patterson, G. R. (1982). Coercive family process. Eugene, Or: Castalia Press.
Pearlin, L. I., Lieberman, M. A., Menaghan, E. G., & Mullan, J. T. (1981). The stress process. Journal of Health and Social Behavior, 22, 337-356.
Porter, B., & O'Leary, D. (1980). Marital discord and child behavior problems. Journal of Abnormal Child Psychology, 8, 287-295.
Reid, W. J., & Crisafulli, A. (1990). Marital discord and child behavior problems: A meta-analysis. Journal of Abnormal Child Psychology, 18, 105-117.
Simons, R. L., Lorenz, F. O., Conger, R. D., & Wu, C. (1992). Support from spouse as mediator and moderator of the disruptive influence of economic strain on parenting. Child Development, 63, 1282-1301.
Spanier, G. B. (1976). Measuring dyadic adjustment: New scales for assessing the quality of marriage and similar dyads. Journal of Marriage and the Family, 38, 15-28.
Tschann, J. M., Johnston, M. K., Kline, M., & Wallerstein, J. S. (1989). Family process and children's functioning during divorce. Journal of Marriage and the Family, 51, 431-444.
Voydanoff, P., & Majka, L. C. (1988). Families and economic distress: Coping strategies and social policy. Beverly Hills, CA: Sage.
Youniss, J., & Smollar, J. (1985). Adolescent relations with mothers, fathers and friends. Chicago: University of Chicago Press.
Dania S. Clark-Lempers, Head Start Coordinator, Drake University.