Chronic Family Adversity and Early Child Behavior Problems:
A Longitudinal Study of Low Income Families
Introduction
An initial step in the prevention of psychopathology in children is the identification of conditions and events associated with a disproportionately high incidence of behavior problems. Research efforts in this area have frequently focused on familial stressors. In one of the most comprehensive investigations of this topic, Rutter, Cox, Tupling, Berger, and Yule (1975a) and Rutter et al. (1975b) reported a dramatic increase in the probability of children exhibiting a behavior disorder as a function of multiple family stressors. Rutter and colleagues compiled a Family Adversity Index (FAI) of chronic stressors in two diverse communities, an inner London borough and the Isle of Wight. Family stressors included overcrowding in the home or large family size, the mother suffering from depression or a neurotic disorder, the father having been convicted of any offense against the law, marital discord, and the father having an unskilled/semiskilled job (Rutter, 1978). In isolation any one of these family stressors was not associated with an increased likelihood of child behavior problems; however, when two or more stressors were present, the risk of child behavior problems was found to increase two- to four-fold in both communities.
More recent investigations of Rutter's cumulative stressor hypothesis have confirmed that the likelihood of child behavior problems increases with the number of family stressors. This finding has been replicated across cultures and over time (Blanz, Schmidt, & Esser, 1991; Sanson, Oberklaid, Pedlow, & Prior, 1991; Shaw & Emery, 1988). Results of the Shaw and Emery (1988) study are typical of more recent investigations in that they confirm the need to go beyond considering correlations between individual stressors and child outcome by examining the cumulative impact of stressor variables on child functioning. As in the Rutter et al. (1975a) study, Shaw and Emery found that stressor groupings were significantly related to child behavior problems, though in the latter study the results were more consistent with an additive, rather than a multiplicative model. In recent investigations of the cumulative stressor hypothesis, conducted in Germany and Australia, respectively, family stressors were predictive of child adjustment across time (Blanz et al., 1991; Sanson et al., 1991). The Sanson et al. project was notable because it was one of the first to test the association between chronic family adversity and child outcome beginning during infancy. Given the developmental status of the children, Sanson et al. (1991) added several stressors to the FAI (e.g., infant temperament, prematurity, perinatal stress). Of these, difficult infant temperament, in combination with one or more risk factors (e.g., sex of the child, perinatal stress, prematurity), was a particularly strong predictor of both externalizing and internalizing problem behavior at ages 4-5. The child being male also appeared to increase the risk of early externalizing problems in conjunction with other risk factors being present.
The present investigation seeks to extend our understanding of children's early behavioral adjustment in facing stressful familial events and conditions by examining the relationship between family adversity across infancy and into the toddler period using a low income, urban American sample. Assessments of family adversity were conducted when children were ages 1 and 2, and behavior problems were assessed at age 3.
Method
Subjects
Subjects originally included 100 mother-child dyads (59 males and 41 females) recruited from the Women, Infants, and Children (WIC) Nutritional Supplement Program of Allegheny County as part of a larger longitudinal study of child development. WIC provides financial support to purchase nutritionally-sound food items for low income families. At the time of the infant's birth, mothers ranged in age from 17 to 36 years, with a mean age of 25. Sampling of marital status was not restricted due to the considerable relationship instability within the sample. Forty-six percent were either married or living together, whereas 54% were either divorced (8%) or separated (9%), or single (37%). The majority of families were Caucasian (61%) and the remainder were African-American (39%). The mean family income in the sample was between $500-1,000 per month, with 72.5% of the families having yearly earnings equal to or less than $12,000. Mothers' average level of education was 12 years, with 73.5% having a high-school degree or less.
Mothers of infants 6-11 months of age were recruited in two WIC waiting rooms by one of the principal investigators on days in which a high number of potential subjects were scheduled for appointments (i.e., four-five versus one-two mothers with infants 6-11 months old were scheduled). Mothers were informed that the study was a project examining child development and mother-child interaction patterns, and that they would be paid $15.00 for each lab visit, plus an additional $10.00 for transportation costs. Upon agreeing to participate, informed consent was sought, mothers completed three questionnaires, and the first lab visit was scheduled within two weeks of the infant's first birthday. Mothers were contacted two weeks before the assessment by mail, and one week before by phone, to confirm the appointment. Of the 144 women who were asked to take part in the study, 129 (89.6% of the 144) agreed to participate, but only 100 completed the 12-month assessment. Of those 100 subjects seen when infants were 12 months old, 89 participated in the 18- and 24-month laboratory assessments months). At age 3, 82 mothers returned completed questionnaire reports on their children's behavior problems. No significant or appreciable differences were found when demographic characteristics of families who completed all assessments (82 of the 100) and those who did not (18 of the 100) were compared. A similar comparison was made between families who completed two of the three laboratory assessments (e.g., 12- and 18-month, but not 24-month assessments) versus those who completed all three, with no appreciable differences found between groups on demographic variables.
Procedures
When mothers were originally recruited for participation in the study, they also were administered a demographic questionnaire, the Beck Depression Inventory (BDI), and the Bates' Infant Characteristics Questionnaire (ICQ). Three laboratory assessments were conducted at ages 12, 18, and 24 months, a home visit was conducted at 15 months, and mothers were sent the Achenbach Child Behavior Checklist (CBCL) when children were 36 months old. Different laboratories were used at adjacent assessments (i.e., at 12 and 18 months, and at 18 and 24 months), both of which were equipped with a one-way mirror through which the assessment was videotaped.
The instruments for the present study were completed at the recruitment session when infants were between 6-11 months, at three laboratory visits when infants were 12, 18, and 24 months, and by mail when children were age 3.
Measures
Background Information Questionnaire. This questionnaire, administered at the recruitment and all laboratory assessments, included items regarding demographic information and data on the home environment. Information pertaining to income, parental criminality, and overcrowding were derived from the questionnaire.
Beck Depression Inventory (BDI). Mothers completed the BDI, a well-established and widely used measure of depressive states, at recruitment, the 12-, 18-, and 24-month assessments (Beck,, Ward, Mendelon, Mock, & Erbaugh, 1961; Beck & Beamesderfer, 1974; Reynolds & Gould, 1981). Split-half reliability of the scale has been found to be high (.86 to .93). Research has identified maternal depressive symptomatology as an influential factor in predicting attachment quality as well as later child behavioral adjustment (Lyons-Ruth, Connell, Zoll, & Stahl, 1987; Radke-Yarrow, Cummings, Kuczynski, & Chapman, 1985; Shaw & Emery, 1988). In order to make the instrument more trait-like, directions for the BDI were altered so mothers reported about depressive symptoms experienced in the past six months rather than the past week. Though depressive symptoms are more likely to be episodic than trait-like, among families from impoverished backgrounds it also is likely to be recurrent because of chronic stress. Thus, it appeared justifiable to have mothers rate symptomatology during the past six months. Because the BDI was administered four times during the course of the study, it was decided to composite adjacent assessments that were theoretically meaningful. Empirically, this was justifiable given that correlations between composited BDI scores were equal to or greater than .53 for composited scores, p<.0001. Scores for the recruitment and 12-month administrations were summed and averaged, reflecting maternal depressive symptoms in the first year of the infant's life (termed "Year 1" below), as were scores for the 18- and 24-month administrations (termed "Year 2" below), reflecting BDI scores during the infant's second year.
Personality Research Form (PRF). Items from three factors of the PRF, aggression, defendence, and social desirability, were administered at the recruitment and 18-month laboratory visit (Jackson, 1989). Defendence assesses the degree to which individuals suspect that people mean them harm or are against them. Maternal personality risk attributes were included as part of the FAI because of prior research showing relationships with school-age children's poor functioning above and beyond variance accounted for by affective disorders (Hammen, 1991). During the infancy period, these specific PRF factors have been found to differentiate attachment security among two samples of low income subjects (Egeland & Farber, 1984; Shaw & Vondra, in press). Each factor is represented by 16 items, for a total of 48 true-false questions. Internal consistency reliabilities for the three factors range from .72 to .87 (Stumpf, Wieck, & Jackson, 1976), and test-retest reliabilities range from .84 to .87 for the three scales (Jackson & Morf, 1973). Over a 7-12-month period in the present sample, test-retest reliabilities ranged from .46 for defendence to .62 for desirability (p < .0001 for all three factors). Ratings from the recruitment administration are termed "Year 1 Personality Risk" scores, and 18-month assessment ratings are referred to as "Year 2 Personality Risk" scores.
Infant Characteristics Questionnaire (ICQ). At the recruitment, 12-, 18-, and 24-month assessments, parents rated their infant on the ICQ, a series of seven-point scales from which factor scores have been constructed indexing persistent, unstoppable, difficult behavior, and negative adaptation (Bates, Freeland, & Lounsbury, 1979). For the present investigation, the Difficulty factor was used given its relation to preschool behavior problems and comparability of items from 12 to 24 months (i.e., unstopability is not included as a factor prior to 13 months). Since the item structure of the Difficulty factor changes at 6, 13, and 24 months, the seven-item 6-month factor was used at the screening, the nine-item 13-month factor was used at the 12-month assessment, and the 7-item 24-month factor was used at the 18- and 24-month assessments. The instrument meets customary psychometric standards for maternal reports on infant temperament, has the advantage of brevity, and has shown longitudinal relations with preschool behavior problems (Bates et al., 1985). Scores for the ICQ Difficulty factor also were composited in the same manner as the BDI, based on theoretical and empirical grounds (r was equal to or greater than .51 for composited scores, p<.0001). "Year 1" scores refer to the summed and averaged recruitment and 12-month assessment ratings and "Year 2" scores refer to the summed and averaged 18- and 24-month ratings.
Marital Adjustment Test. Maternal satisfaction with her marital or significant other relationship were assessed using the short form of the Marital Adjustment Test at the 12-month laboratory visit (Locke & Wallace, 1959). This measure has proven successful in discriminating harmonious and disturbed marriages (Hershorn & Rosenbaum, 1983; Locke & Wallace, 1959; Rosenbaum & O'Leary, 1981), and predicting children's behavior problems (Emery & O'Leary, 1982). In the event of a recent separation, mothers were instructed to report on that period within the last year when she and her partner were living together. In those instances where mothers were not married, they were asked to rate their closest intimate relationship, including a live-in boyfriend, girlfriend, relative, or current dating partner, and the word "relationship" or "close relationship" was substituted for "marriage." When the relationship was non-sexual in nature, the single item concerned with sex relations was omitted. This strategy is responsive to the fact that many of the study's mothers were single (54%), and allowed for the provision of important information on whichever close relationship mothers considered to have primacy. Such a measurement strategy recognizes the significance of maternal perceptions above and beyond objective circumstances.
Child-Rearing Disagreements Scale (CRD). The CRD is a 21-item measure of common topics concerning child-rearing disagreements, based on interviews with parents over the course of several years (Jouriles, Murphy, Farris, Smith, Richters, & Waters, 1991). It was administered at the 24-month laboratory assessment. In comparison with general marital satisfaction scales, the CRD has been found to be correlated with a greater variety of child behavior problems and account for significant variance in behavior problems after accounting for non-child disagreements and child exposure to marital conflict. The response format is a 6-point Likert scale with a Cronbach alpha of .86. The CRD was administered in a similar manner to the MAT to account for contextual variability of childrearing arrangements among study participants. In those instances where mothers were not married, they were asked to complete the questionnaire if there was a co-parent who was responsible for rearing the child with them, including live-in boyfriends, ex-husbands, mothers, or other live-in relatives and friends.
Child Behavior Checklist for Ages 2-3 (CBCL) (Achenbach, Edelbrock & Howell, 1987). The CBCL is a one hundred item questionnaire designed to assess behavioral and emotional problems in children ages 2-3. The questionnaire generates two broad band factors, Externalizing and Internalizing Problems, which were used for analyses in this study. Unlike the CBCL for older children, there are no sex specific scales on the 2-3 year old version, so that the Externalizing and Internalizing factors consist of the same items for boys and girls. The mean test-retest reliability is reported by the authors to be .87. Discriminative validity between nonreferred children and children referred to mental health services is strong, and divergent validity has been demonstrated by a lack of significant correlations between the CBCL and standard cognitive measures.
Multiple Stressor Groupings. Modifications in the familial stressors used in the Isle of Wight Study (Rutter et al., 1975a) were made so that stressors were more specific to early child outcome and to the chronic stressors typical of a low income sample. In most cases, this meant a more stringent criteria for a factor to be considered a stressor, based on the high base rates of impoverished living conditions, and maternal and marital dysfunction in the sample. All family stressor variables were dichotomized into present or not-present categories. Since all subjects needed to meet financial criterion for low income status to take part in the WIC Program, criterion for the present study was more stringent; "low income" was defined as family earnings of $6,000 or less per year. Overcrowding was dichotomized based on criterion developed by Rutter (1978) in the Isle of Wight Study (Rutter & Quinton, personal communication, February 11, 1985). If families had more than four children living at home or if there was less than one room per person in the home, it was considered to be overcrowded. Parental criminality was based on maternal report of criminal behavior committed by either of the infant's parents living at home. For personality risk (PRF), criteria were met by scores above the sample median on the aggression and defendence factors, and below the sample median on the desirability factor. For depressive symptomatology (BDI), criterion was met by scores at or above 15 at Year 1 or Year 2 (one standard deviation above the sample mean), or scores at or above 9 at both administrations from which Year 1 and Year 2 scores were formed. BDI scores at or above 9 are considered to indicate mild depressive symptomatology while scores of 15 or above are generally considered in the moderate-to-severe range of depressive symptomatology (Beck & Beamesderfer, 1974). The Marital Adjustment Test (MAT) was used as an index of marital distress at Year 1, while the Child-Rearing Disagreements Scale (CRD) was used at Year 2 to assess conflict specifically related to child-rearing areas. For the MAT, mothers who scored one standard deviation below the sample mean were considered to be dissatisfied with their relationship. For the CRD, mothers with scores one standard deviation above the sample mean were considered to be conflictual. For the Difficulty factor of the Infant Characteristics Questionnaire, infants with scores one standard deviation above the Year 1 and Year 2 composites were considered to be difficult.
Once each of the seven family stressors was dichotomized, the sum of the number of stressors present was calculated. Subjects were then divided into four stressor groups (Group 1 = 0 stressors present, Group 2 = 1 stressor present, Group 3 = 2 stressors present, Group 4 = 3, 4, or 5 stressors present). Four groups were chosen based on the fact that no subjects had more than 5 stressors in Year 1 or 2 and only three and two subjects received scores of 5 stressors present in Years 1 and 2, respectively.
Results
Results are presented in four stages: (1) descriptive statistics among continuous independent and dependent variables; (2) correlations between individual family stressors and age 3 CBCL Externalizing and Internalizing problem behaviors; (3) group differences between stressor groups and age 3 behavior problems; and (4) multivariate relations between family stressors and age 3 behavior problems.
Descriptive Statistics
First, means, standard deviations, range of scores, and stressor cutoff scores for all continuous variables are presented in Table 1. The number of cases for variables differ due to the inappropriateness of the Locke and Wallace and Childrearing Disagreement inventories for all subjects, and to attrition at the follow-up assessments. ICQ scores are comparable to those of published norms when they are interpolated to account for summing and averaging across different age versions of the Difficulty factor (Bates, Freeland, & Lounsbury, 1979). BDI scores also are comparable to others samples of low-income mothers (Christopoulos, Cohn, Shaw, Joyce, Sullivan-Hanson, Kraft, & Emery, 1987), and CBCL t-scores are extremely close to those of published norms (Achenbach et al., 1987).
Individual Relations among Stressors and Child Behavior Problems
In Table 2, Pearson correlation coefficients among individual stressors and age 3 CBCL Externalizing and Internalizing factors are presented for all subjects and separately by gender. In general, few stressors at ages 1 and 2 were consistent predictors of age 3 Internalizing or Externalizing problems, and even fewer were consistent across sex. The one exception to the latter was parental conflict over childrearing disputes at age 2, which predicted age 3 Externalizing and Internalizing problems across sex. For boys only, maternal depressive symptoms and maternal personality risk factors at ages 1 and 2 were related to age 3 behavior problems, particularly Externalizing problem behavior. At age 2 only for boys, low family income and infant difficulty were associated with CBCL Internalizing problems. For girls only, low marital satisfaction at age 1 and infant difficulty at ages 1 and 2 were related to age 3 behavior problems.
Stressor Group Differences Analyses
In order to test Rutter's cumulative stressor hypothesis, procedures described in the Measures section were used to form stressor groups at age 1 and 2, and used as independent variables, with CBCL Externalizing and Internalizing factors used as dependent variables. Given that sex differences were found in the correlational analysis, MANCOVAs initially were computed with stressor group serving as the independent variable and sex as the covariate. No sex or sex by stressor group interactions were found; however, significant multivariate effects for stressor group were found at both ages 1 (F(3, 77) = 2.91, p < .05) and 2 (F(3, 77) = 2.22, p < .05). Results of the follow-up univariate analyses for Externalizing and Internalizing factors are presented in Table 3. Significant effects were found for Internalizing problems at both ages and Externalizing problems at age 2 only. Post hoc contrasts testing for differences among stressor groups indicated that for Internalizing problems, there were significant differences between Groups 1 and 4 at age 1, and at age 2, between Groups 1 and 4 and 2 and 4 (p < .05 for all contrasts). For age 3 Externalizing problems, significant differences were found at age 2 between Groups 1, 2, and 3, and Group 4 (p < .05).
To examine the linearity of the relationship between number of stressors and CBCL Internalizing and Externalizing problems more rigorously, Pearson correlation coefficients were computed between stressors at age 1 and 2 and CBCL behavior problems using the original stressor counts to increase the range of scores; thus, there were six stressor groups in this analysis with scores ranging from 0 to 5. Stressors at Year 1 and Year 2 were associated with increased risk of CBCL Externalizing and Internalizing behavior problems at age 3. For age 1 stressors, r = .27, p < .01 with Externalizing problems, and r = .42, p < .001 with Internalizing problems. For age 2 stressors, r = .25, p < .05 for Externalizing problems, and r = .32, p <.01 for Internalizing problems.
Finally, in order to examine whether stressors operate differently when children show high levels of problem behavior, CBCL Externalizing and Internalizing scores were dichotomized into two groups. Children with scores one standard deviation above the sample mean on the CBCL Externalizing and Internalizing factors were compared to those children with scores below this criteria. A series of chi square analyses was then computed between stressors at age 1 and age 2 and CBCL behavior problems at age 3. Though these relations were not significant for age 3 Externalizing problems, age 1 stressors were significantly associated with increased risk for high internalizing problems (X2(3, N = 81) = 11.47, p < .01), and there was nonsignificant trend in the expected direction for age 2 stressors and high internalizing problems at age 3 (X2(3, N = 81) = 6.77, p < .08).
Multivariate Relations Between Stressors and Age 3 Behavior Problems
In order to examine the effects of family stressors on child outcome from a multivariate perspective, a series of multiple regression procedures were computed. Stressors showing significant zero order correlations (p < .05) with age 3 Externalizing or Internalizing problems were entered into hierarchical multiple regression equations. Since sex differences were found in the correlational analysis, regression models were computed separately by gender and separately for Externalizing and Internalizing problem behavior. Predictors were entered in chronological order with earlier variables entered first (e.g., age 2 BDI and CRD scores). In cases where variables were assessed at the same time, stepwise ordering was used to determine which variable accounted for more variance. Variables that explained less variance were only retained in the equation if they added significant variance to the equation. Results are presented in Table 4.
For boys, different variables accounted for variance in explaining CBCL Externalizing and Internalizing factors. For Externalizing scores, maternal personality risk assessed at 18 months and age 2 childrearing practices conflict both accounted for an additional 12% of variance after maternal personality risk at recruitment and Year 1 maternal depressive symptoms had been entered in the regression equation. Unique variance in boys' Internalizing problems was accounted for by Year 1 maternal depressive symptoms, 24-month family income, and Year 2 perceived difficult infant temperament.
For girls, regression equations for Externalizing and Internalizing problems were similar. Marital satisfaction at 12 months and Year 1 perceived infant difficult temperament together accounted for 45% of the variance in explaining both CBCL Externalizing and Internalizing problems.
Discussion
At a broad level, the results of the present investigation are in accord with past findings on the impact of cumulative chronic family stress on children's adjustment. As with previous studies of school-age and preschool children, increase in family stress was associated with an increase in behavior problems (Blanz et al., 1991; Rutter et al., 1975a, 1975b; Sanson et al., 1991; Shaw & Emery, 1988). Relations between individual stressors and age 3 behavior problems varied by sex of child. The one exception to this was 24-month childrearing disagreements, which predicted age 3 externalizing and internalizing problems strongly for both boys and girls. For boys, externalizing problems also were predicted by earlier maternal depressive symptoms and maternal personality risk, while age 3 internalizing problems were related to earlier low income, difficult infant temperament, and maternal depressive symptoms. For girls, early difficult infant temperament and marital dissatisfaction were the strongest predictors of age 3 externalizing and internalizing problems, along with childrearing disagreements.
As mentioned above, the pattern of results regarding the relationship between family stress and child behavior problems broadly supports previous research, but the findings are more in accord with those of Shaw et al. (1988) than those of Rutter and colleagues' (1975a, 1975b). In the former, a linear pattern between family stress and children's adjustment emerged, while in the latter, a multiplicative or potentiating effect was found so that the interaction of two or more stressors was greater than their sum. In the present study, a linear model was consistent with the pattern of children's behavior problem scores, especially when stressor groups were kept in their unrecoded format.
The pattern of results also parallels those found by Richman, Stevenson, and Graham (1982) in their longitudinal study of the emergence of behavior problems from preschool to school-age. When these authors computed a multiple regression analysis between factors of adversity and age 3 behavior problems, including such stressors as poor marital functioning, maternal warmth and mental status, developmental delay of the child, and poor housing, two major findings emerged. First, factors reflecting relationships within the family appeared to exert the strongest influence on child behavior problems, the three strongest being quality of the marriage, maternal warmth, and maternal criticism. Second, the effects of these within-family factors were maximized in the presence of social disadvantage. That is, the effects of marital disharmony, low maternal warmth, and high maternal criticism on child behavior were greater in families from working class backgrounds. Why would the association between such within-family factors be more consistently associated with behavior problems among children from low SES backgrounds? As both the Richman et al. (1982) study and our own findings indicate, we believe that factors that more directly compromise the quality of parenting are more likely to affect child behavior (e.g., maternal depression, parental conflict), and that these factors are more likely to have a negative impact when the quality of parenting is further compromised by conditions of social adversity (Shaw & Bell, 1993). The buffers of the middle class child's ecosystem may soften the impact of individual within-family stressors on parental functioning, and thus because parenting abilities are also not being simultaneously affected by the hardships of poverty, child behavior is not as adversely influenced.
However, the magnitude of this relation might be slightly inflated for reasons discussed below.
In the present study, when relations between individual stressors and child behavior problems were examined longitudinally, only one variable was a consistent predictor of behavior problems across sex: childrearing disagreements. It is unfortunate that this variable was not assessed at age 1 to test its predictive power repeatedly; though a related factor, marital satisfaction, was strongly associated with girls' age 3 problem behavior. The consistency with which childrearing disagreements were associated with later behavior problems for both boys and girls confirms the findings of several investigators documenting the robustness of the relationship between parental conflict and children's adjustment (Emery, 1982, 1988; Hetherington, Cox, & Cox, 1985; O'Leary & Emery, 1984; Shaw & Emery, 1987; Shaw, Emery, & Tuer; 1993), including for preschool children (Block, Block, & Gjerde, 1986; Block, Block, & Morrison, 1981; Snyder, Klein, Godowski, Faulstich, & LaCombe, 1988). Moreover, as Jouriles et al. (1991) note, specific indices of childrearing disagreements appear to be more consistently related to behavior problems than global scales of marital adjustment. Though in the present study marital satisfaction at age 1 was strongly related to later behavior problems, this association was only true for girls. While age 2 childrearing disagreements predicted boy boys' and girls' age 3 behavior problems.
Alternatively, it is possible that higher rates of childrearing disagreements are a consequence rather than a cause of having a child who is more difficult behaviorally or temperamentally. Though behavior problems were not assessed at age 2, we were able to examine the relationship between perceived infant difficulty at age 1 and childrearing disagreements at age 2. For both boys and girls, there were nonsignificant trends in the expected direction, with age 1 Difficulty on the ICQ related to childrearing disagreements at age 2 (for boys, r = .24, p < .06; for girls, r = .27, p < .07). However, the magnitude of this rlation might be slightly inflated for reasons discussed below.
The study's findings are also in accord with the work of several investigators that have found early temperament to be related to later behavior problems alone (Cameron, 1978; Earls & Jung, 1987; Bates & Marvinney, 1993), and in combination with other risk factors (Sanson et al., 1991; Bates et al., 1985). However, we found early ratings of difficult temperament to be more consistently related to girls' rather than boys' age 3 behavior problems, and more consistently related to internalizing rather than externalizing problems.
The study is not without its limitations. First, there exists the potential that a response bias may partially explain the pattern of results given maternal report was the sole source of data. Though it is possible that mothers who report higher rates of marital dissatisfaction and/or conflict, depressive symptoms, personality risk factors, and/or perceive their infants as more difficult temperamentally would be more likely to report higher rates of behavior problems for children one-two years later, the pattern of results is not entirely consistent with such an explanation. Some of these associations were found, but few such instances were found across time and sex of child. Thus, the response bias appears to be, at the least, moderated by the child's sex, and at the most, of questionable significance given the differential association of specific variables with age 3 child problem behavior.
Second, the size of the sample is small by epidemiological standards. Though the results confirm several previous findings regarding predictors of child outcome, replication of the sex differences found between individual stressors and problem child behavior is in order, particularly among larger samples of low income, preschool-age children.
In sum, the results provide longitudinal evidence to support the notion that different components of family adversity exert cumulative adverse effects on young children's behavioral adjustment. The results also suggest that researchers should continue to examine the role of parental conflict over childrearing disuptes in the development of psychopathology, and that an expansion of the early assessment of this factor appears warranted. Finally, the results indicate that sex-specific models are needed to identify salient risk indicators of early psychopathology.
Acknowledgements
This study was supported by the following organizations within the University of Pittsburgh: the Mental Health Clinical Research Center for Affective Disorders, the Office of Child Development, the Central Research Development Fund, the School of Education in conjunction with the Buhl Foundation, and the Faculty of Arts and Sciences. Requests for reprints should be sent to the first author at the following address: Department of Psychology, Clinical Psychology Center, 604 Old Engineering Hall, 4015 O'Hara Street, University of Pittsburgh, Pittsburgh, PA, 15260. We wish to thank Anne Kolar, Barbara Radigan, Krista Wilhelm, and Mary Krieger for their assistance with data collection, and extend our appreciation to study participants for letting us watch and learn about family development.
References
Achenbach, T.M., Edelbrock, C., & Howell, C. (1987). Empirically-based assessment of the behavioral/emotional problems of 2-3 year old children. Journal of Abnormal Child Psychology, 15(4), 629-650.
Bates, J.E., & Marvinney, D. (1993). Temperament, mother-child relations and marital harmony as predictors of child adjustment at 3-5 years. Paper presented at the Society for Research in Child Development, New Orleans, LA.
Bates, J.E., Freeland, C.B., & Lounsbury, M.L. (1979). Measurement of infant difficultness. Child Development, 50, 794-803.
Bates, J.E., Maslin, C.A., & Frankel, K.A. (1985). Attachment security, mother-child interaction, and temperament as predictors of behavior-problem ratings at age three years. In I. Bretherton & E. Waters (Eds.), Growing points of attachment theory and research. Monographs of the Society for Research in Child Development, 50, Nos. 1-2, 167-193.
Beck, A.T., Ward, C.H., Mendelon, M., Mock, J.E., & Erbaugh, J.K. (1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 561-571.
Beck, A.T., & Beamesderfer, A. (1974). Assessment of depression: The Depression Inventory. In P. Pichot (Ed.), Psychological measurement in psychopharmacology: Modern problems in pharmacopsychiatry (Vol. 7). Basel, Switzerland: Kanger.
Blanz, B., Schmidt, M.H., & Esser, G. (1991). Familial adversities and child psychiatric disorders. Journal of Child Psychiatry and Psychology, 32, 939-950.
Cameron, J. (1977). Parental treatment, children's temperament, and the risk of childhood behavior problems. American Journal of Orthopsychiatry, 47, 568-576.
Christopoulos, C., Cohn, D.A., Shaw, D.S., Joyce, S. Sullivan-Hanson, J., Kraft, S.P., & Emery, R.E. (1987). Children of abused women: I. Adjustment at time of shelter residence. Journal of Marriage and the Family, 49, 611-619.
Egeland, B., & Farber, E.A. (1984). Infant-mother attachment: Factors related to its development and changes over time. Child Development, 55, 753-771.
Earls, F., & Jung, K.G. (1987). Temperament and home environment characteristics as causal factors in the early development of childhood psychopathology. Journal of the American Academy of Child Psychiatry, 26, 491-498.
Emery, R.E., & O'Leary, K.D. (1982). Children's perceptions of marital discord and behavior problems of boys and girls. Journal of Abnormal Child Psychology, 10, 11-24.
Hammen, C. (1991). Depression runs in families: The social context of risk and resilience in children of depressed mothers. New York: Springer-Verlag.
Hershorn, M. & Rosenbaum, A. (1983). A closer look at the unintended victims. Unpublished manuscript, Syracuse University.
Jackson, D.H. (1989). Personality research form manual (3rd Edition). New York: Research Psychologists Press.
Jackson, D.H., & Morf, M. (1973). An empirical evaluation of factor reliability. Multivariate Behavioral Research, 8, 439-459.
Jouriles, E.N., Murphy, C. M., Farris, A.M., Smith, D. A., Richters, J.E., & Waters, E. (1991). Marital adjustment, parental disagreements about child rearing, and behavior problems in boys: Increasing the specificity of the marital assessment. Child Development, 62, 1424-1433.
Locke, H.J., & Wallace, K.M. (1959). Short marital-adjustment and prediction tests: Their reliability and validity. Marriage and Family Living, 21, 251-255.
Lyons-Ruth, K., Connell, D.B., Zoll, D., & Stahl, J. (1987). Infants at social risk: Relations among infant maltreatment, maternal behavior, and infant attachment behavior. Developmental Psychology, 23, 223-232.
O'Leary, K.D., & Emery, R.E.. (1984). Marital discord and child behavior problems. In M.D. Levine & P. Satz (Eds.), Developmental variation and dysfunction (pp. 345-364). New York: Academic Press.
Prior, M., Smart, D., Sanson, A., & Oberklaid, F. (1993). Sex differences in psychological adjustment from infancy to 8 years. Journal of the American Academy of Child and Adolescent Psychiatry, 32, 291-304.
Radke-Yarrow, M., Cummings, E.M., Kuczynski, L, & Chapman, M. (1985). Patterns of attachment in two- and three-year-olds in normal families and families with parental depression. Child Development, 56, 884-893.
Reynolds, W.M. & Gould, J.W. (1981). Marital psychometric investigation of the standard and short form Beck Depression Inventory. Journal of Consulting and Clinical Psychology, 49, 63-71.
Richman, N., Stevenson, J., & Graham. P.J. (1982). Pre-school to school: A Behavioural Study. London: Academic Press.
Rosenbaum, A., & O'Leary, K.D. (1981). Marital violence: Characteristics of abusive couples. Journal of Consulting and Clinical Psychology, 12, 328-329.
Rutter, M. (1978). Family, area, and school influences in the genesis of conduct disorders. In L.A. Hersov & D. Schaffer (Eds.), Aggression and anti-social behavior in childhood and adolescence (pp. 95-114). Oxford: Pergamon Press.
Rutter, M. (1981). Stress, coping, and development: Some issues and some questions. Journal of Child Psychology and Psychiatry, 22, 323-356.
Rutter, M., Cox, A., Tupling, C., Berger, M., & Yule, W. (1975a). Attainment and adjustment in two geographical areas: 1. The prevalence of psychiatric disorder. British Journal of Psychiatry, 126, 493-509.
Rutter, M., Yule, B., Quinton, D., Rowlands, O., Yule, W., & Berger, W. (1975b). Attainment and adjustment in two geographical areas: 3. Some factors accounting for area differences. British Journal of Psychiatry, 126, 520-533.
Sanson, A., Oberklaid, F., Pedlow, R., & Prior, M. (1991). Risk indicators: Assessment of infancy predictors of pre-school behavioural maladjustment. Journal of Child Psychology and Psychiatry, 32, 609-626.
Shaw, D.S., & Bell, R.Q. (1993). Developmental theories of parental contributors to antisocial behavior. Journal of Abnormal Child Psychology, 21, 25-49
Shaw, D.S., & Emery, R.E. (1988). Chronic family adversity and school-age children's adjustment. Journal of the American Academy of Child and Adolescent Psychiatry, 27, 200-206.
Shaw, D.S., & Vondra, J.I. (1993). Chronic family adversity and infant attachment security. Journal of Child Psychology and Psychiatry and Allied Disciplines, 34, 1205-1215.
Stumpf, H. Wieck, T., & Jackson, D.H. (1976). Personality Research Form ( PRF). Revidierte Uebersetzung. Psychologisches Institut der Unversetaet Bonn.