|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 MODERATING FACTORS IN THE RELATIONSHIP BETWEEN SOCIAL STRESSORS AND SYMPTOMS OF PSYCHOPATHOLOGY By MARY E. SACZAWA A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2011
2 2011 Mary Saczawa
3 To my parents
4 ACKNOWLEDGMENTS I thank my chair and my committee for their guidance throughout this process. I also thank my family for their unwavering support
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURE S ................................ ................................ ................................ .......... 9 A BSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 Parenting and Family Context ................................ ................................ ................. 14 Individual Differences in Physiological Functioning ................................ ................ 16 The Present Study ................................ ................................ ................................ .. 21 2 METHOD ................................ ................................ ................................ ................ 23 Participants ................................ ................................ ................................ ............. 23 Sampling Procedures ................................ ................................ .............................. 25 Saliva S ampling ................................ ................................ ................................ 25 DHEAS Sampling ................................ ................................ ............................. 26 Measures ................................ ................................ ................................ ................ 27 Depressive Symptoms ................................ ................................ ...................... 27 Aggressive Behaviors ................................ ................................ ....................... 27 Peer Problems ................................ ................................ ................................ .. 28 Parenting Behaviors ................................ ................................ ......................... 28 Environmental Chaos ................................ ................................ ....................... 30 Hormone Levels ................................ ................................ ............................... 31 Covariates ................................ ................................ ................................ ........ 32 Analyses ................................ ................................ ................................ ................. 33 3 RESULTS ................................ ................................ ................................ ............... 35 Descriptives ................................ ................................ ................................ ............ 35 Regressions ................................ ................................ ................................ ............ 36 4 DISCUSSION ................................ ................................ ................................ ......... 51 Depressive S ymptoms ................................ ................................ ............................ 52 Aggression ................................ ................................ ................................ .............. 53 Peer P roblems ................................ ................................ ................................ ........ 54 Cortisol/DHEAS R atio ................................ ................................ ............................. 55 Parenting and Environmental Chaos ................................ ................................ ...... 56
6 Implications and Limitations ................................ ................................ .................... 57 APPENDIX A CBCL SOCIAL PROBLEMS SUBSCALE ................................ ............................... 61 B N INVENTORY ................................ ............................ 62 C CBCL AGGRESSIVE BEHA VIORS SUBSCALE ................................ .................... 64 D BREAKDOWN OF CHAOS V ARIABLES ................................ ................................ 65 E ADDITIONAL STATISTIC S ................................ ................................ ..................... 68 LI ST OF REFERENCES ................................ ................................ ............................... 72 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 84
7 LIST OF TABLES Table page 3 1 Means and Standard Deviations for Time 1 Primary Variables ......................... 41 3 2 Means and Standard Deviations for Time 2 Primary Variables .......................... 41 3 3 Cross Sectional Correlations of Time 1 Variables ................................ .............. 41 3 4 Cross Sectional Correlations of Time 2 Variables ................................ .............. 42 3 5 Cross sectional regression using Time 1 variables to predict Time 1 dep ressive symptoms, controlling for concurrent aggression ............................. 42 3 6 Cross sectional regression using Time 2 variables to predict Time 2 depressive symptoms, controlling for concurrent aggression ............................. 43 3 7 Longitudinal regression using Time 1 variables to predict c hange in depressive symptoms, controlling for concurrent aggression ................................ ............... 45 3 8 Cross sectional regression using Time 1 variables to predict Time 1 aggression, controlling for concurrent depressive symptoms ............................. 47 3 9 Cross sectional regression using Time 2 variables to predict Time 2 aggression, controlling for concurrent depressive symptoms ............................. 48 3 1 0 Longitudinal regression using Time 1 variables to predict change in aggression, controlling for concurrent depressive symptoms ............................. 49 D 1 Time 1 Chaos frequencies ................................ ................................ .................. 65 D 2 ................................ ................... 65 D 3 ................................ ................ 65 D 4 Time 1 number of places cared for by a non parent ................................ ........... 65 D 5 Time 1 number of non parent caregivers ................................ ............................ 66 D 6 Time 2 Chaos frequencies ................................ ................................ .................. 66 D 7 ................................ ................... 66 D 8 ................................ ................ 66 D 9 Time 2 number of places cared for by a non parent ................................ ........... 67 D 10 Time 2 number of non parent caregivers ................................ ............................ 67
8 E 1 Demographic frequencies ................................ ................................ ................... 68 E 2 SES, age, and puberty descriptives ................................ ................................ .... 68 E 3 Means and standard deviations for Time 1 primary variables ............................ 69 E 4 Means and standard deviations for Time 2 primary variables ............................ 69 E 5 Cross Sectional Corre lations of Time 1 Variables ................................ .............. 70 E 6 Cross Sectional Correlations of Time 2 Variables ................................ .............. 70 E 7 Longitudinal Correlations of Time 1 Predictors and Time 2 Outcome Variables 70 E 8 Frequency of ma ternal warmth at Time 1 ................................ ........................... 71 E 9 Frequency of maternal hostility at Time 1 ................................ ........................... 71 E 10 Frequency of maternal warmth at Time 2 ................................ ........................... 71 E 11 Frequency of maternal hostility at Time 2 ................................ ........................... 71
9 LIST OF FIGURES Figure page 1 1 Moderated regression model. ................................ ................................ ............ 22 3 1 The interaction between Parenting Characteristics and Rid icule in the prediction of depressive symptoms. ................................ ................................ ... 44 3 2 The interaction between Time 1 Cortisol/DHEAS Ratio and Time 1 Ridicule in the prediction of change in depressive symptoms ................................ .............. 46 3 3 The interaction between Time 1 Cortisol/DHEAS Ratio and Time 1 Ridicule in the prediction of the change in Aggression ................................ ......................... 50
10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science MODERATING FACTORS IN THE RELATIONSHIP BETWEEN SOCIAL STRESSORS AND SYMPTOMS OF PSYCHOPATHOLOGY By Mary E. Saczawa December 2011 Chair: Julia Graber Major: Psychology The experience of p sychological disorders during adolescence result s in impaired functioning concurrently and is a key predictor of problems later in life. At the same time, individuals may share similar experiences yet have very different outcomes, particularly in terms of psychopathology. While many young adolescents report some degree of distress associated with peer relationships, only a small portion develops lasting ps ychological problems, such as depression or aggression. This study assess ed the moderating effects of parenting, environmental chaos, and the cortisol/DHEAS ratio on the relationship between peer problems and symptoms of depression and aggression. Partic ipants were 194 young adolescent boys and girls (Time 1 Mage = 10.86) and their mothers, re assessed approximately one year later Parenting was coded from video taped interactions dail y routine was determined from child and maternal interviews The cortisol/DHEAS ratio was determined using saliva and urine samples, respectively, controlling for pubertal timing. Neither environmental chaos nor parenting were identified as significant pr edictors of symptoms of psychopathology; however, results indicated that the
11 cortisol/DHEAS ratio was a significant predictor of change in depressive symptoms over time, and the interaction between the cortisol/DHEAS ratio and child report of peer ridicule was a significant predictor of change in both aggression and depressive symptoms over time. Consistent with prior research on the ratio, high cortisol and low DHEAS was predictive of depressive symptoms while the opposite was found for aggression. This study is the first to examine the moderating effects of these three factors in the development of psychopathology in young adolescents.
12 CHAPTER 1 INTRODUCTION The experience of p sychological disorders during adolescence result s in impaired functioning concurrently and is a key predictor of problems later in life. Many studies have identified stressful experiences with peers as predictors of increases in both aggressive and depressive symptoms during the early adolescent years (Laird, Jordan, Dodge, Pet tit, & Bates, 2001; Panak & Garber, 1992 ; Sontag & Graber, 2010 ) While the relationship between social stressors during adolescence and abnormal psychological adjustment has been established in the li terature it is not yet understood why many adolescent s experience social stress with out lasting detrimental effects This study examine d environmental and biological factors that may influence the relationship between social stressors during adolescence and psychological well being Even without counting the time spent in school, adolescents spend more time socializing with similar aged peers than with any other group of people, including their families (Larson & Verma, 1999). When asked to list the most important people in their liv es, a majority of those listed by adolescents were their peers. This represents a shift from childhood, when more time is spent with family members (Larson & Richards, 1991) and the importance of family is greater than that of peers (Furman & Buhrmester, 1992). It is not until adolescence that friendships begin to display the emotional intimacy characteristic of adult relationships and become based on per sonal beliefs and mutual trust (Brown & Larson, 2009) While social interaction is pivotal for healthy emotional development, with increased tru st and personal investment come increased risk for emotional distress ( Aboud & Mendelson, 1998 ).
13 Peer relationships can have significant positive or negative influences on the young adolescent. Beneficial relationships are characterized by emoti onal support and intimacy; t hey help the individual develop a positive self image, thereby protecting against depression and the effects of future negative social influences ( Hartup, 1996; LaGreca & Harrison, 2005 ). However, many adolescents report that social interactions are a source of stress in their daily lives (Brooks Gunn, 1991). Negative peer relationships are unstable or hostile, causing the young adolescent to develop a poor or inconsistent self image ( Hart er, 2001 ). This puts the individual at a higher risk for depression ( Boivin, Hymel, & Bukowski, 1995; Panak & Garber, 1992 ) aggressive behavior disorders (Coie, Lochman, Terry, & Hyman, 1992; Laird et al. 2001; Werner & Crick, 2004) and substance abuse ( Skara et al., 2008; Sullivan, Farrell & Kliewer, 2006 ). A dolescents are typically identified as having negative peer relationships if they are often teased or bullied by other peers or have f ew or no same aged confidants (Achenbach & Edelbrock, 1981) The importance of friendships has been illustrated in many studies in which adolescents who reported having few or no close friendships were shown to fare worse than their peers in terms of psychological adjustment ( Nangle, Erdley, Newman, Mason, & Carp enter, 2003 ). From adolescence on, s trikingly more girls than boys display internalizing symptoms such as depression or anxiety ( Lewinsohn, Hops Roberts, Seeley, & Andrews, 1993 ). While some internalizing disorders, such as anxiety disorders, show a gender gap even in childhood, rates of depressive symptoms are not significantly different between boys and girls until adolescence, during which time the gender difference emerges a nd is maintained through adulthood (Cyranowski, Frank, Young, & Shear, 2000) A recent national study of adolescents found that over 25% of adolescents 12 17 years old
14 endorsed having had a period of significantly depressed mood in the last 12 months, a m ajority of whom were girls (62.6%; Centers for Disease Control and Prevention, 2010). Of adolescents who experience negative peer interactions, more girls than boys report feeling sad or lonely because of it ( LaGreca & Harrison, 2005; Rigby, 2002 ). In co ntrast, in at least one study, b o ys were found to display more delinquent behaviors when ridiculed by peers ( Sullivan, Farrell, & Kliwer, 2006 ) Overall, serious aggressive behaviors, such as physical fighting, are significantly more common in boys ( Lewin sohn et al., 1993 ). While it has come to be understood more recently that girls and boys engage in comparable rates of aggression (Card, Stucky, Sawalani, & Little, 2008), physical aggression, which is characteristic of problem behavior disorders, is more characteristic of the aggression displayed by boys. Parenting and Family Context Even though adolescents spend significantly more of their time socializing with peers than any other group, parents remain key factors in their social and emotional development. Much of the parenting literature has focused on responsiveness and demandingness as criti cal dimensions of parenting. These have been operationalize d via a wide range of parenting behaviors that may be exhibited during parent child interactions (e.g., Hetherin gton & Clingempeel, 1992 ). For example, y oung adolescents of parents who display wa rmth and support typically display lower rates of both internalizing and externalizing symptoms ( Bowes, Maughan, Caspi, Moffitt, & Arsenaut, 2010; DeVore, 2005 ). In addition children of hostile mothers display more aggression ( Ho, Bluestein, & Jenkins, 2008; Zadeh, Jenkins, & Pepler, 2010) and depressive symptoms (Sheeber, Davis, Leve, Hops, & Tildesley, 2007). A study by Ge and colleagues found that adolescents who had comorbid depressive symptoms and conduct problems had parents
15 who displaye d the highest levels of hostility and the lowest levels of parental warmth (Ge, Best, Conger, & Simons, 1996). Robert Weiss (1974 receive particular types of support (companionship, guidance, affec tion) from various sources in their lives (e.g. friend s, parents, siblings, teachers). Furman and Buhrmester (1985) found that there is a great deal of interplay among sources of support for most adolescents such that s ome sources (e.g. parents) may compe nsate for low support from others (e.g. peers). Additional research has further supported the findings of Furman and Buhrmester showing that warm and supportive parenting has a protective effect against bullying and rejection by peers ( Bowes et al., 2010; Patterson, Cohn, & Kao, 1989 ). A dolescents who lack support from two main sources, such as parents and peers, have been found to have the poorest adjustment (Gauze, Bukowski, Aquan Assee, & Sippola, 1996). In addition to the relations hip the adolescent has with his or her mother, other aspects of the home environment can have distinct effects on development. Recent studies have identified environmental chaos, or t he degr ee or lack of routine and organization that an adolescent experiences as salient to adolescent well being (Matheny, Wachs, Ludwig, & Phillips, 1995) A p ositive home environment, which includes low household chaos, is particularly important for adolescents who experience peer victimization (Bowes et al. 2010). Chaotic home environments that contain little or no consistency or few regular routine s have been shown to have negative effects on cognitive and psychosocial development (Deater Deckard, Mullineaux, Beekman, Petrill, Schatschneider, & Thomp son, 2009; Kliewer & Kluger, 1998) beyond
16 the effects of socioeconomic status (Evans, Gonnella, Marcynyszyn, Gentile, & Salpekar, 2005) Ratings of noise, foot traffic, and crowding are typically used as measures of environmental chaos ( Matheny et al., 1995 ), but only a small number of studies have examined additional information directly related to the individual such as bedtime consistency number of different childcare arrangements and related aspects of regular household routines These studies hav e found that such routines have a significant and cognitive development (Evans, 2006; Fiese et al., 2002 ; Johnson, Martin, Brooks Gunn, & Petrill, 2008; Wachs & Corapci, 2005). Hence, the present investigation considered e ffects of both parenting (as observed in a parent child interaction task) and household routine as moderators of the link between social problems or stress and symptoms of psychopathology. Individual D ifferences in P hysiological F unctioning In addition to the potential buffering effects of parenting on the relationship between social stress and psychopathology, individual differences in a ctivity in the hypothalamic pituitary adrenal (HPA) axis have been identified as potential sources of vulnerability to ps ychosocial stress In response to a stressor, the hypothalamus releases corticotropin releasing hormone (CRH). CRH acts on the pituitary gland, stimulating the release of adrenocorticotropic hormone (ACTH), which activates the adrenal glands. The adrena l glands then produce pregnenolone which is further metabolized into the adrenal hormones cortisol and dehydroepiandrosterone (DHEA; see Goodyer, Park, Netherton, & Herbert, 2001, and Wolf & Kirschbaum, 1 999 for review s ). Cortisol then acts in various par ts of the body to prepare the body to respond to the stressor DHEA and its sulfate, DHEAS, can act throughout the brain and body or can be further metabolized into 50 75%
17 19 97). A negative feedback mechanism regulate s activity of the HPA axis. Corticosteroids act on receptors in the hippocampus, which acts to decrease activity in the HPA axis following the termination of a stressful event. However, in the presence of chron ic stress, such as that experienced in continued social rejection, the hypothalamus also releases vasopressin (Ma & Lightman, 1998) ; v asopressin amplifies the effects of CRH in the pituitary gland, thereby increasing production of pregnenolone even with the effects of the negative feedback system (see Yu, Holsboer, & Almeida, 2008, for review). Adults Lavelle, & Scott, 2004), and blockage of that receptor has been sho wn to decrease depressive symptoms and aggressive behaviors (Heinrichs & Domes, 2008). A majority of the research on HPA axis dysregulation has focused on changes in cortisol. While cortisol typically displays diurnal fluctuations, with peak levels appear ing immediately after waking, abnormally high or low levels have been linked to internalizing and externalizing disorders, particularly depression and aggressive behavior disorders. In adults, h igh levels of cortisol after waking are associated with sever ity of depressive symptomatology as well as duration of depressive episode and likelihood of occurrence (Meador Woodruff et al., 1990) ; patterns of increased basal cortisol levels and increased HPA axis response to psychological stressors have also been se en in many studies in children and adolescents with depression (Lopez Duran, Kovacs, & George, 2009) In contrast, l ow cortisol levels after waking have been associated with increased aggression as well as accelerated appearance of adult like aggressive b ehaviors in human and
18 animal models (for review, see Soma, Scotti, Newman, Charlier, & Demas, 2008). A study of children and adolescents with conduct disorder found that low cortisol levels are more strongly associated with the aggressive behaviors relate d to conduct disorder than the non aggressive behaviors, such as delinquent behaviors (van Goozen, Matthys, Cohen Kettenis, Gispen de Wied, Weigant, & van Engeland, 1998). In contrast, low DHEA(S) levels have been associated with depressive symptoms an d high levels have been linked with aggression. Though the mechanism of action is less clear, studies in children and adolescents have found that low levels of DHEA(S) were associated with higher levels of depressive symptoms (see Angold, 2003, for review ). Other studies, however, have not shown this effect and in one study of girls, high levels of DHEA ( S ) interacted with other risk factors (i.e., early pubertal timing) to predict higher depressive symptoms (Graber, Brooks Gunn, & Warren, 2006). The role of high DHEA (S) in aggression is better understood (see Soma et al., 2008, for review) DHEA (S) acts in the brain to inhibit the activity of GABA, an inhibitory neurotransmitter. GABA has been identified as one of the neurotransmitters responsible for t he control of aggressive behaviors, such that high levels of GABA inhibition cause an increase in such behaviors (Majewska, 1992). While it can also be metabolized into androgens such as testosterone, which have also been associated with aggression (see S oma et al., 2008, for review), DHEA ( S ) has been linked to aggression independent of testosterone levels ( Pajer, Tabbah, Gardner, Rubin, Czambel, & Wang, 2006; van Goozen, Matthys, Cohen Kettenis, Thijssen, & van England, 1998 ; van Goozen et al., 2000 ). Be cause both cortisol and DHEA(S) are synthesized from pregnenolone, comparative ratios allow for the study of preferential production of one or the other
19 independent of overall HPA activity. It is typical and beneficial for h igh levels of cortisol to be secreted during transient stressful events ( Lephart, Baxter, & Parker, 1987 ; Luppa, Munker, Nagel, Weber, & Englehardt, 1991 ; Parker, Levin, & Lifrak, 1985 ; Wade et al., 1988 ), but long term exposure to high cortisol levels has been shown to cause neuronal death in the hippocampus and prefrontal cortex. While less is known about the physiological role of DHEA(S), it is believed to act as a protective agent, even promoting neurogenesis and thereby minimizing the negative effects of long term cortisol exposu re (see Maninger, Wolkowitz, Reus, Epel, & Mellon, 2009, for review). For example, high DHEA ( S ) levels have been shown to decrease the detrimental effects of high cortisol, suggesting an antidepressant property (Kaminska, Harris, Gijsbers, & Dubrovsky, 20 00). However, as previously mentioned, DHEA(S) can have a negative psychological effect through excessive inhibition of GABA neurotransmitters. A balance between the two hormones is preferred, and signific antly higher or lower cortisol/ DHEA(S) ratios have been associated with depression ( Goodyer et al., 1996; Goodyer, Herbert, & Tamplin, 2003) an d aggression ( Buydens Branchey & Branchey, 2004 ; Pajer et al., 2006 ), respectively. Many researchers have speculated as to why cortisol or DHEA(S) may be preferentially produced following HPA axis reactivity, but the exact mechanism is not yet known. Some have suggested that early experience of stress and/or long term exposure to stresso rs may cause a stress sensitization or inoculation; the former causing heightened cortisol production and the latter resulting in heightened DHEA(S) production (Ozbay, Fitterling, Charney, & Southwick, 2008). This lends explanation as to why
20 individuals r espond differently to similar social stressors, but it creates another question regarding what biological or psychosocial factors determine sensitization or inoculation. Little research has examined the association of peer stressors and cortisol and t he research conducted to date on this or related areas has been somewhat inconclusive. A meta analysis of 208 studies (Dickerson & Kemeny, 2004) revealed that laboratory stress evaluative thre associated with greater cortisol reactivity than were those tests limited to mental stress (i.e., cognitive challenge tasks) I n line with evolutionary theory and primate studies of cortisol response to social threats (Sapolsky, 1993), individuals high in defensiveness actually show lower cortisol levels following social rejection than do those low in defensiveness (Blackhart, Eckel, & Tice, 2007). In one of the first studies of peer stress, Gunnar and colleagues (2003) found that peer rejection in preschoolers was related to higher cortisol levels. To date, no studies (that the author is aware of) have examined whether social problems are associated with DHEA(S). In addition, w hile there has been extensive study of cortisol levels in c hild ren who have experienced maltreatment or abuse (MacMillan et al., 2009) few studies have looked at the effects of normative parenting styles on cortisol reactivity. A nimal studies have shown that the offspring of warm and attentive mothers (as indica ted by species specific behaviors) have lower cortisol response to stressors (Repetti, Taylor, & Seeman, 2002) and faster decline in cortisol levels following a stressor than offspring with low maternal warmth (Gunnar, Gonzalez, Goodlin, & Levine, 1981; Meaney, Aitken, van Berkel, Bhatnagar, & Sapolsky, 1988 ). Paralleling the animal studies, Chorpita and Barlow (1998) found that children from families low in warmth display abnormal cortisol
21 response. A long term study by Flinn and England (1995) showed that children in stable, affectionate families had moderate and more stable cortisol levels, while those households unstable in composition and parenting style had non normal (either high or low) and highly variable cortisol levels. The effects of stabili ty and routine have also shown effects separate from parenting behaviors, with lower levels of structure being associated with greater HPA axis activity (Ellenbogen & Hodgins, 2009). The Present Study The present study seeks to address gaps in the existing literature by examining the moderating effects of parenting characteristics, environmental chaos, and hormone levels on the relationship between peer problems and psychological adjustment both cross sectionally and long itudinally It was hypothesized that greater report of peer problems would be related to higher aggression and depressive symptoms and that this relationship would be moderated by biological and environmental factors (Figure 1) Based on the literature, it was also hypothesized that high or low cortisol/ DHEAS ratios, negative parenting and environmental chaos would be associated with higher reports of aggression or depressive symptoms. Specifically high cortisol/ DHEAS ratios were expected to predict higher depressive symptoms while lo w cortisol/ DHEAS ratios were expected to predict higher aggression when combined with the effects of negative parenting and /or environmental chaos
22 Figure 1 1. Moderated regression model.
23 CHAPTER 2 METHOD Participants This study utilizes data collected as part of the Girls and Boys Health and Development Project (GBHD) This larger study examine d the biological and psychosocial changes taking place during the transition from childhood to adolescence in a sample of children and their parents w ho were seen multiple times over the course of 3 4 years. Data from the girls were collected from 1996 2000, and data from the boys were collected from 2000 2005 Due to protocol changes over the course of the study, only the final two annual assessments were used for the girls, and the second and third assessments were used for the boys. These time points are most equally age matched and contain measures tapping the constructs of the present study For ease of assessment when girls and boys were 10.86 years of age on average (SD = 0.77) assessment when they were 12.04 years of age on average (SD = 0.77) interviews E nrollment data are used for demographic and retention information only All procedures in the GBHD were approved by the Institutional Review Board at Teachers College of Columbia University and a protocol review exemption was granted by the Institutional Review Board at the University of Florida for the present investigation To recruit for this project, flyers were distributed to students at school in ethnically diverse, working class area s in New York City. Students were instructed to gi ve the flyers to their mothers or primary caregivers who were to complete and return them if they were interested in learning more about the study. Mothers were then contacted, the study was
24 explained, and an in home visit was scheduled if they desired t o proceed. Due to the nature of the recruitment procedure, it is not known what percentage of students who received the flyers enrolled in the study, but approximately 85% of those mothers who returned cards enrolled. Two hundred sixty seven pre to ea rly adolescent boys and girls ( Ngirls = 138 N boys = 129 ) enrolled in the study. The racial /ethnic makeup of the participants at enrollment was consistent with that of the surrounding area, 39% Caucasian, 32% African American, 12% Hispanic, 2 % Asian or Pacific Islander, and 15 % multiracial or not specified. One hundred ninety four participants (Ngirls = 92 Nboys = 102 ) participated at Time 1 ( 27 % attrition) As all participants were contacted at each assessment 202 families (Ngirls = 111 N boys = 91 ) participated at Time 2 ( 24 % attrition from enrollment ) No significant differences were found on outcome variables between those who did and did not continue in the study, p > .05 Male participants were more likely than females to have discontinued from enrollment to time 2, but not from enrollment to time 1, likely because many females returned for the later time point who did not take part in the earlier one. Those who identifi ed as Asian or Pacific Islander, multiracial, or not specified were more likely to have discontinued by time 2, and participants with lower socioeconomic status were more likely to have discontinued at both time points from enrollment Each in home visit was approximately 1.5 2 hours long and the entire visit was videotaped for coding and reliability purposes After informed consent and assent were obtained, the mother and child completed self report surveys and a series of tasks. At the end of each visit, children received a gift such as a t shirt or tote bag
25 Additional questionnaires were completed by the mother and child after the in home visit and saliva and urine samples were collected from the child immediately u pon awakeni ng for two (boys) or three (girls) days following the in home visit A research assistant returne days after the home visit to collect the forms and biological samples At this time, mothers received a small payment Sampling Procedures Saliva Sampling Saliva samples were collected for the two (boys) or three (girls) day s following the in home visit. During the visit, research assistants explained the instr uctions for sample collection, and a detailed copy of the instructions was left with the family. Rather than setting a time, researchers asked that the samples be coll ected immediately upon the asked to assist the child in completing the procedure. Saliva samples were collected using Salivette kits (Sarstedt, Germany); t he instructio ns for the Salievette kits were altered slightly to assure sufficient amounts of saliva; the participant was instructed to remove the cotton swab from the tube and hold it in her/ his mouth for two minutes (instructions stated 40 seconds) without chewing on it. The child then returned the cotton swab to the tube and closed it securely. Participants were instructed to label the sample with the date and time of collection and store them in coolers provided for sample storage until the research assistant retu rned to pick them up. Once the saliva samples were retrieved by the research assistant, they were stored at 25 Celsius until they were analyzed at the Columbia Presbyterian Reproductive Endocrinology Department in New York City, NY. In order to stimulate saliva production, girls ingested a small amount of
26 sweetened Kool Aid At the time of these procedures, use of such substance s was commonplace (Granger, Weisz, & Kauneckis 1994; Gunnar Brodersen, Nachimas, Buss, & Rigatuso 1996; Hertsgaard, Gunnar, Larson, Brodersen, & Lehman 1992; Nachimas, Gunnar, Mangelsdorf, Parritz, & Buss 1996), but the amount used in this study was less than in other protocols ( Gunnar, Brodersen, Nachimas, Buss, & Gigatuso, 1996). I t was later found that this could alter the results of certain kits for salivary cortisol assays but t he kits used in this protocol were not affected by use of sweetened drink mix ( Gordon, Peloso, Auker, & Dozier, 2005; Schwartz, Granger, Susman, Gunnar, & Laird, 1998 ; Talge, Donzella, Kryzer, Gierens, & Gunnar, 2005 ) This practice was discontinued prior to collection of saliva samples for the boys. DHEAS S ampling Urine samples were also collected each morning immediately upon awakening, for the two (boys) or three (girls) days following the in home visit. Urine was collected from girls in a pediatric urine hat and from boys in a urine collection container a lid was secured, and the sample was placed into a supplied plastic bag. Mothers recorded the date and time of the sample on the containers and kept them in the cooler until a researcher returned to collect them. Participants were encouraged not to dr ink anything before bed in order to prevent the child from needing to urinate during the night. If the child did have to urinate during the night, both that and the morning sample were to be collected in the same container. The times for both samples wer e recorded on the lid. After the samples were picked up by a researcher, they were stored in a refrigerator until analysis at the Columbia Presbyterian Reproductive Endocrinology Department in New York City.
27 Measures Depressive S ymptoms Girls and boys Appendix C), a widely used measure of depressive symptoms (Nolen Hoeksema et al, 1992) adapted from the Beck Depression Inventory to assess depressive symptoms in children. As is common pra ctice in non clinical settings, the question regarding suicidality was removed for this study (Twenge & Nolen Hoeksema, 2002) resulting in a 26 item CDI. T he child selects which one of the three statements is most like how s/he has felt over the previous given scores ranging 0 2, and the score is summed with higher scores indicating higher lev els of depressive symptomatology. This measure has displayed validity and reliability in normative samples of this age group (Smucker, Craighe ad, Craighead, & Green, 1986). Aggressiv e B ehaviors Maternal report of aggression was assessed using the Aggressive Behavior subscale of the Child Behavior Checklist (CBCL; Achenbach, 1991 (Appendix D). The CBCL is a widely used measure of social and emotional development that has demonstrated reliability and validity in this age group. This subscale consists of 20 items Mothers rate statements from 0 ( not at all ) to 2 ( very much ) according to how true they are of their child. Items are summed such that higher scores indicate higher rates of aggressive behaviors. cale in this study at Time 1 is .87 and at Time 2 is .86.
28 Peer Problems Reports of problematic peer interactions were measured using the Social Problems subscale of the CBCL; ( Achenbach, 1991; Appendix B) completed by the mother. The an ms subscale in this study at Time 1 is .73 and at Time 2 is .58. Examination of individual questions did not indicate any items responsible for the decrease in alpha value; Time 1 and Time 2 So cial Problem subscale scores exhibited a strong correlation, r = .61, p < .001, and performance in the final analyses was comparable. Peer problems were also assessed from participant reports of ridicule. During an interview in the home visit, b oys and girls reported if they had ever been insulted, made fun of, or avoided for any of the following 12 reasons: size, skin color, clothes, hair, language, religion, age, weight, eye color, where they live and the food they eat responses for each item were summed to create a total score reflecting ridicule experienced This measure was created for the larger study to assess typical experiences of being picked on or teased and has not been validated. A subset of participants completed the Youth Self Re port in one of the years of the larger project; correlation of self reported ridicule and the Social Problems scale of the YSR was r = .408, p < .001 ; N = 107 Parenting B ehaviors Parenting characteristics were coded from video taped mother child interactions completed during the home visits. In separate rooms, mother s completed a self report questionnaire and participants answered interview questions on issues about which parents and children typically disagree, such as household chores, interactions with
29 From that list, the research assistant selected two items that were endorsed as being the topics of frequent or heated disagreements. The research assistant presented these topics to the mother and child and asked them to discuss ways to resolve each problem. They were given a list of done to avoid t the research assistant left them for five to seven minutes to discuss the topics. Interactions were coded on scales developed by Melby et al. (1998) and Graber et al. (1999) and adapted for this study. T he mother and child were rated separate ly by coders using a five point Likert scale for characteristics including warmth and hostility. an interest in the other person, and enjo such as touching and making eye contact, positive verbal communication, and emotional expressions such as smiling, laughing, and showing interest through asking questions are examples of warmth behaviors. H [and] has the potential for harming or hostility include grimace s or smirks, yelling or speaking in a sarcastic tone, assuming a defensive posture, or pushing. Due to the non normal distribution of ratings, kappa statistics show only moderate inter warmth = .63 and hostility = .33 In order to establish reliability unaffected by the truncated ratings, percentages within one point of the gold standard coder were calculated resulting in warmth = 95% and hostility = 97.25% Because hostility ratings were highly positively skewed, and in order to reduce
30 the number of variables entered into the models due to power considerations, both parenting characteristics were collapsed into one. Preliminary analyses indicated that the two variables were moderately correlated (Time 1 r = .36, p <.001; Time 2 r = .45, p < .001), so m aternal warmth was reverse coded and summed with hostility to form an overall measure of negative parenting. Environmental C haos To determine the level of environme ntal chaos the child experienced in the home, a measure was developed using questions from the in home interview with the mother survey completed by the mother, and the interview with the child Th ese measure s include questions on regularity caregiving. For child caregiving, m others were asked to indicate who cared for the target child an d where this child was cared for when the parent ( s ) were not home Frequency analyses identified roughly one quarter of the sample with more variable childcare arrangements; specifically, 28% of the sample report ed child self care or two or more caregivers other than the parents and 23% report ed two or more locations (outside of being cared for by the parents at home) in a typical week ( Appendix D ) These items were scored as 1 = self care or two or more caregivers per week and 0 = all others and 1 = two or more locations and 0 = less than two locations for out of home care. Although some of the non parental caregivers listed may currently reside in the home wi th the removed from the analysis. The purpose of this measure is to assess the degree of egardless of their familiarity, and only 22 participants (approximately 11%) resided with adult extended family members. The child was also asked with whom he or she usually eats dinner. Fourteen percent of participants reported
31 that they usually did not eat dinner with either parent ; this was coded as 1, and all other responses were coded as 0. For the items regarding regular bedtimes, 24% of the mothers reported that their child did not have a regular bedtime on weeknights, and 66% reported a lack of regular bedtime on weekends. Because of th is, only the weeknight item was used in the analysis (1 = no regular bedtime and 0 = regular bedtime ) These four items (multiple caregivers, multiple locations, dinner without parent and no regular bedtime) were summed to create an overall score for envi ronmental chaos ranging from 0 4 Hormone L evels Cortisol was assayed from saliva samples and DHEAS was assayed from urine samples at the Columbia Presbyterian Reproductive Endocrinology Department in New York City, NY. Cortisol s amples were centrifuged at 3000 rpm for 10 minutes. Cortisol level was determined using a radioimmunoassay adapted for use with saliva (Diagnostic Products Company) with a lower detection limit of 0.02 l/dl per 200 l of saliva. Every sample provided at least 400 l of saliva; a 200 l sample was used for duplicate analysis. All samples from each participant were analyzed in one assay run. The inter and intra assay variation coefficients were less than 3% and 5%, respectively. DHEAS level was determined using a commercial s olid phase, competitive chemiluminescent immunoassay (Immulite, Siemens, Los Angeles, CA) with a sensitivity of 3 g/dl. The inter and intra assay variation coefficients were less than 8.2% and 12.0%, respectively. Due to the importance of time of day eff ects on cortisol values, the data were checked for significantly different sampling times prior to calculating the mean cortisol value When outlying sampling times occurred with participants who had provided three samples, the two most similar times were retained. When only two samples were
32 provided, the sample with the highest cortisol value which was typically the earliest sample, was retained. Individual samples from only 19 participants in Time 1 and 31 participants in Time 2 were removed After t his process was complete, the remaining scores were averaged to give a mean cortisol score for each time point. While cortisol has a daily rhythm, DHEAS fluctuates throughout the lifetime, and very drastically during puberty. As the individual progresses through puberty, DHEAS levels increase. To control for this effect, pubertal status was residualized out of the mean DHEAS values. Pubertal status was measured using Tanner scoring (Marshall & Tanner, 1969, 1970), a five point Likert scale based on drawings of pubic hair growth and gonadal development for boys and pubic hair growth and breast development for girls (Morris & Udry, 19 80 ) Ratings for the two indicators (for each gender) were averaged to create a mean Tanner score Mother reports were used; if mother reports were not available child ratings of Tanner scores were used. Mean Tanner scores were entered as the sole predictor of DHEAS score in a regression, and unstandardized residualized were saved. This procedure was repeated for both ti me points. Mean cortisol values were then divided by residualized mean DHEAS values to create a ratio, and an inverse transformation was performed to normalize the distribution for analyses Higher scores indicate high cortisol and low DHEAS values, and l ower scores indicate low cortisol and high DHEAS values. Covariates /ethnicity gender, and socioeconomic status were entered as ethnicity were reported by the mother during the in home interview. Age was entered as a continuous variable, and race /ethnicity was dummy coded into three racial groups: African American/ Black, Hispanic, and other with
33 Caucasian as the omitted group al, and other or not specified. S ocioeconomic status (SES) was determined using based on answers given during the in home interview Hollingshead scores for this sample ranged from 9 to 66, with higher scores indicating a higher socioeconomic status. Analyses Hierarchical regression analyses were used to test the hypothes e s that parenting characteristics, chaos, and cortisol/DHEAS ratio moderated the relationship between peer problems and psychological adjustm ent (Baron & Kenny, 1986) Prior to conducting these analyses, descriptive statistics including means, skewness, and kurtosis were run to ensure that all variables were normally distributed. The only variable requiring transformation was the cortisol/DHE AS ratio, as noted above. Correlations were also examined to assess the associations between the variables as well as to test for multicollinearity between predictors. Hierarchical regressions were performed using parenting, environmental chaos, cortisol/DHEAS ratio, and peer measures predicting symptoms both cross sectionally and longitudinally. The first block included covariates consisting of socioeconomic status, gend er, race /ethnicity and age at the outcome time point. For the Time 2 cross sectional and the longitudinal analyses, Time 1 symptoms were entered in Block 2. Because of high rates of comorbidity between internalizing and externalizing disorders and symptoms particularly between aggression and depression ( R owe, Maughan, & Eley, 2006 ), comorbid symptoms were entered into the analysis in the block pri or to the parenting, chaos and the hormone ratio The se predictors were entered into the next
34 block, followed by centered interaction terms in the final block a s per the specifications of Baron and Kenny (1986) for testing moderation All analyses were run for each dependent variable separately, resulting in six final analyses. Due to power concerns over entering all of the interaction terms into the same model simultaneously, separate analyses were run for mother report of social problems and child report of ridicule and their respective interaction terms in order to determine which interaction terms should be entered into the final analyses
35 CHAPTER 3 RESULTS Descriptives The means and standard deviations for the predictor and outcome variables are reported in Table 3 1. The mean CDI total scores for Time 1 and Time 2 were 7.02 and 6.27, respectively, which are lower than expected for this age group. In particular, prior studies have reported CDI scores ranging from 8.36 to 9.08 for nine year olds and 9.30 to 9.91, for ten year olds (Smucker, Craighead, Craighead, & Green, 1986; Twenge & Nolen Hoeksema, 2002). Serious levels of depressive symptoms are indicated by scores over 19 (Kovacs, 1980; Smucker et al., 1986). For this sample, high levels of depressive symptoms were reported by 4.9% of the Time 1 and 3.8% of the Time 2 participants Again, this is fewer than expected The cutoff of 19 was calcu lated from the top 10% of a normative sample, though those samples included older adolescents. Studies on younger samples have suggested cutoffs of 13 (Larson & Melin, 1992) and 15 (Almqvist et al., 1999). This would increase the percentage of the current sample endorsing high levels of depressive symptoms to 12.6% (Time 1) and 11.4% (Time 2) or 9.3% (Time 1) and 7.1% (Time 2), respectively. CDI scores did not change significantly from Time 1 to Time 2 for the total sample, t (151) = .97 NS or by gender. Cross sectional analyses also indicated no significant gender differences at either time point. The change from Time 1 to Time 2 CDI score ranged from 29 to 15 with a mean change statistic of 0.46 SD = 5.80 Change in CDI score was sli ghtly negatively skewed ( 0.97) and moderately kurtotic (4.85). Aggressive Behavior Subscale scores from the CBCL were somewhat higher than scores previously reported for children/young adolescents of this age (Bongers, Koot, van
36 der Ende, & Verhulst, 2 003). Longitudinally, Time 1 aggression was significantly higher than Time 2 aggression, t (168) = 2.11, p < .05. No significant gender difference was found in aggression cross sectionally and neither males, t (85) = 1.90, NS nor females, t (82) = 1.08, NS showed significant differences in aggression across time points (Table 3 2). The change from Time 1 to Time 2 aggression score ranged from 11 to 17 with a mean change statistic of 0.63 SD = 3.87 Change in aggression score was slightly positively sk ewed (0.24) and moderately kurtotic (2.44). Cross sectional correlations among core variables in this study are shown in Tables 3.3 3.6. Child report of ridicule and parent report of social problems were not significantly correlated at either time poin t ( r = .11, at both times); the size of this correlation indicat es that there is no concern of multicollinearity when entering both into the same model. Parent report of social problems was moderately correlated with aggression ( Time 1 r = .51, p < .001; Time 2 r = .52, p < .001) and weakly with depressive symptoms ( r s = .24, p < .01) at both times. Child report of ridicule was weakly but significantly correlated with depressive symptoms ( Time 1 r = .18, p < .05; Time 2 r = .27 p < .0 5 ) but not with aggression at both times. Chaos at Time 2 was weakly but significantly correlated with both aggression and depressive symptoms at Time 2, r = .28, p < .001 and r = 16, p < .05, respectively, indicating that higher levels of chaos were corr elated with higher levels of psychological symptoms. However, these associations were not seen at Time 1. Few other significant associations among variables were found (see Tables 3 3 and 3 4 ). Regressions A hierarchical regression analyzing the cross sectional relationship between the Time 1 predictors and Time 1 depressive symptoms (Table 3 5) indicated that concurrent
37 aggression and problematic peer relationships are both significant predictors of depressive symptoms. C orrelation analyses had indicated that aggression and depressive symptoms were weakly though significantly positively correlated so it is not unexpected that aggression is a significant predictor of depressive symptoms in the model, explaining 4.1% of the variance ( = .215, p < .05). Also as expected, both child and mother report of peer problems were predictive of depressive symptoms with children who experienced more problematic peer relationships reporting significantly higher rates of depressive symptoms ( = .278, p < .01; = .252, p < .01, respectively); these reports together explain 13.1% of the variance in depressive symptoms. The step including the R 2 = 34, NS ). Results from cross sectional analysis of the Time 2 data using hierarchical regress ion is presented in T able 3 6. After entering the control variables, a ggression showed only a trend R 2 = .022 p = .08 ). The step including the reports of peer R 2 = .058 p < .05) ; mother report of social problems had a significant effect ( = .220 p < .05) ; however child report of ridicule demonstrated a trend for an association with depressive symptoms ( = .141 p < .09) T he block containing the biological and envir onmental main effects was not significant R 2 = .00 1 NS ). However, the interaction between child report of ridicule and parenting characteristics also demonstrated a trend for an association with depressive symptoms R 2 = .022 p = .08), suggesting tha t individuals who experience less warmth and more hostility from their mothers were more affected by peer ridicule (Figure 3 1 ). To create the graph, parenting
38 below % or 46 of 182 participants identified as having experienced nega tive parenting characteristics. Participants in the sponse to Results from the longitudinal analysis of depressive symptoms are presented in Table 3 7 Both the main effects and interactions steps were significant predictors of change in depre R 2 = .071, p R 2 = .041, p < .01, respectively). The cortisol/DHEAS ratio had a significant effect ( = .255, p < .01), with higher cortisol and lower DHEAS levels being associated with higher levels of depress ive symptoms over time There was a trend for an effect of the parenting variable ( = .144, p = .065), with less warmth and more hostility being associated with higher levels of depressive symptoms over time The interaction term between child report of ridicule and the cortisol/DHEAS had a significant effect ( = .220, p < .01), indicating that in adolescents whose cortisol/DHEAS ratios are higher, peer ridicule is associated with higher rates of depressive symptoms, while in those whose ratios are lowe r, ridicule does not have as much of an impact on e motional development (Figure 3 2 ). Participants were divided into quartiles based on the cortisol/DHEAS ratio, and the highest and lowest quartiles were graphed in Figure 3 2 contains the participants were in the lowest 25% of ratio values. The high ratio group reported greater depressive symptoms at higher levels of ridicule than did the low ratio group. The cross sectional regression models of Time 1 variables predicting Time 1 aggression are presented in Table 3 8 The control block predicted a significant amount
39 of the variance in aggressive symptoms ( R 2 = .113, p < .05). Socioeconom ic status and race/ethnicity were both significant predictors with lower socioeconomic status being associated with higher rates of aggression ( = .261, p < .01) and Hispanic children being significantly more likely than white children to exhibit aggress ive symptoms ( = .200, p < .05). As would be expected from prior analyses showing a concurrent association between depressive and aggressive symptoms, depressive symptoms were a significant predictor of aggressive symptoms ( = .203, p < .05). The block including child R 2 = .114, p < .001), but only mother report of social problems was a significant predictor ( = .424, p < .001). Neither R 2 = .005, NS R 2 = .003, NS respectively). Results from the cross sectional regression predicting Time 2 aggre ssion are presented in Table 3 9 R 2 = .131 p < .05 ); low socioeconomic status was a significant predictor of higher aggression ( = .307 p < .0 1). Time 2 depressive symptoms demonstrated a trend for an association with aggressive symptoms when added to the model R 2 = 020 p = .0 8 ). As in the Time 1 analysis, the peer problems block was signif R 2 = .232 p social problems was a significant predictor of aggression ( = .521 p < .001). Neither the main effects of nor interactions with biological and environmental predictors were R 2 = .027 NS R 2 = .001 NS respectively ). Results from the longitudinal regression predicting change in aggression are presented in T able 3 10 As in the previous two analyses on aggression, the control block was R 2 = .102, p < .05), and low economic status was a significant
40 predictor of high aggression ( = .297, p < .001). Concurrent depressive symptoms, peer problems, and the main effects of the moderator variables were not significant predictors of change in aggression in this analysis. However, the block for the interaction R 2 = .021, p < .05); the interaction at Time 1 between child report of ridicule and the cortisol/DHEAS ratio was a significant predictor of the change in aggression from Time 1 to Time 2 ( = .117, p < .05). According to Baron and Kenny (1986), the interaction term may be interpreted regardless of whether or not the individual main effects are significant. This indicates that in participants who had low cortisol/DHEAS ratios, or low cortisol and high DHEAS, the effects of ridicule were associated with a greater increase in aggression than in those whose ratios were not as low (Figure 3 3 ) Participants were divided into quartiles based on the cortisol/DHEAS ratio, and the high est and lowest quartiles were graphed in Figure 3 3 group contains participants who were in the lowest 25% of ratio values. The low ratio group reported higher levels of aggression at higher levels of ridicule than did the high ratio group. However, this difference was only notable at the highest levels of ridicule; group differences in aggression were not indicated at lower levels of ridicule, a ccording to the standard errors for each group.
41 Table 3 1. Means and Standar d Deviations for Time 1 Primary V ariables Variable n Min Max M SD T1 Aggression 190 .00 26.00 6.65 5.70 T1 CDI (Adjusted) 182 .00 39.00 7.02 6.33 T1 Ridicule 190 .00 7.00 1.16 1.36 T1 CBCL Social Problems 190 .00 15.00 1.83 2.29 T1 Cortisol/DHEAS a 163 .32 .53 .50 .019 T1 Parenting b 188 2.00 6.00 4.06 1.06 T1 Chaos c 187 .00 4.00 .88 .94 Note: a lower values indicate lower cortisol and higher DHEAS values. b higher values indicate more negative parenting characteristics. c higher values indicate higher levels of chaos Table 3 2. Means and Standard Deviations for Time 2 Primary Variables Variable n Min Max M SD T2 Aggression 195 .00 29.00 5.96 5.32 T2 CDI 184 .00 26.00 6.04 5.71 T2 Ridicule 191 .00 6.00 1.16 1.30 T2 CBCL Social Problems 195 .00 9.00 1.61 1.83 T2 Cortisol/DHEAS a 173 .36 .55 .50 .017 T2 Parenting b 182 2.00 6.00 3.88 .97 T2 Chaos c 196 .00 4.00 .91 .89 Note: a lower values indicate lower cortisol and higher DHEAS values. b higher values indicate more negative parenting characteristics. c higher values indicate higher levels of chaos Table 3 3 Cross Sectional Correlations of Time 1 Variables Variable 1 2 3 4 5 6 7 T1 Aggression T1 CDI .227** T1 Ridicule .073 .266*** T1 CBCL Social Problems .512*** .239** .110 T1 Cortisol/DHEAS .102 .088 .017 .116 T1 Parenting .027 .094 .062 .012 .094 T1 Chaos .076 .008 .019 .069 .039 .167* Note: p < .05. ** p < .01. *** p < .001
42 Table 3 4 Cross Sectional Correlations of Time 2 Variables Variable 1 2 3 4 5 6 7 T2 Aggression T2 CDI .211** T2 Ridicule .033 .182* T2 CBCL Social Problems .523*** .241** .108 T2 Cortisol/DHEAS .003 .035 .127 .016 T2 Parenting .094 .025 .148 .008 .038 T2 Chaos .277*** .161* .127 .093 .100 .146 Note: p < .05. ** p < .01. *** p < .001 Table 3 5. Cross sectional regression using Time 1 variables to predict Time 1 depressive symptoms, controlling for concurrent aggression Variable Model 1 Model 2 Model 3 Model 4 Model 5 Step 1 SES .087 .031 .017 .014 .029 Black .078 .074 .088 .057 .033 Hispanic .158 .115 .075 .077 .088 Other .109 .093 .055 .051 .041 Sex .013 .010 .087 .119 .109 T1 Age .066 .088 .057 .055 .072 Step 2 T1 Aggression .215* .064 .066 .068 Step 3 T1 Ridicule (R) .278** .260** .250** T1 Social Problems (SP) .252** .280** .294** Step 4 T1 Cortisol/DHEAS .158* .165* T1 Parenting .120 .156 T1 Chaos .015 .001 Step 5 SP*Parenting .103 SP*Chaos .122 R 2 .041* .131*** .034 .018 Final R 2 .060 .101 .231 .265 .283 Final Model F 1.384 2.077 4.275*** 3.761*** 3.467*** Note: Standardized beta weights are shown p < .10 p < .05. ** p < .01. *** p < .001
43 Table 3 6 Cross s ectional regression using Time 2 variables to predict Time 2 depressive symptoms, controlling for concurrent aggression Variable Model 1 Model 2 Model 3 Model 4 Model 5 Step 1 SES .014 .035 .018 .020 .040 Black .085 .081 .071 .066 .085 Hispanic .104 .082 .086 .084 .087 Other .034 .040 .011 .009 .012 Sex .028 .032 .016 .011 .016 T2 Age .193 .180 .223 .225 .219 Step 2 T2 Aggression .159 .027 .023 .038 Step 3 T2 Ridicule (R) .141 .146 .113 T2 Social Problems (SP) .220 .221 .187 Step 4 T2 Cortisol/DHEAS .005 .002 T2 Parenting .032 .030 T2 Chaos .002 .000 Step 5 R*Parenting .156 R 2 .022 .058* .001 .022 Final R 2 .06 5 .087 .145 .146 .167 Final Model F 1.497 1.748 2.389* 1.762 1.902* Note: Standardized beta weights are shown p < .10 p < .05.
44 Figur e 3 1. The interaction between p arenting c haracteristics and r idicule in the prediction of depressive symptoms When ridicule was high, participants with more negative parenting exhibited higher levels of depressive symptoms than did those with more positive parenting, p < .10. Error bars indicate the standard error. 0 2 4 6 8 10 12 14 16 18 20 0 1 2 3 4 5 6 7 CDI Score Ridicule negative parenting positive parenting
45 Table 3 7 Longitudinal regression using Tim e 1 variables to predict change in depressive symptoms, controlling for concurrent aggression Variable Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Step 1 SES .042 .038 .009 .004 .006 .008 Black .101 .101 .102 .105 .146 .143 Hispanic .134 .030 .023 .023 .028 .030 Other .112 .015 .018 .022 .030 .021 Sex .072 .097 .092 .080 .015 .039 T2 Age .143 .134 .129 .119 .103 .097 Step 2 T1 CDI .598*** .580*** .556*** .494*** .510*** Step 3 T2 Aggression .097 .096 .115 .125 Step 4 T1 Ridicule (R) .0 70 .059 .000 T1 Social Problems (SP) .002 .060 .078 Step 5 T1 Cortisol/DHEAS .255** .244** T1 Parenting .144 .165* T1 Chaos .009 .009 Step 6 R*Cortisol/DHEAS .220** R 2 .338*** .008 .004 .071** .041** Final R 2 .054 .393 .401 .405 .476 .517 Final Model F 1.070 10.246 9.190*** 7.345*** 7.324*** 7.938*** Note : Standardized beta weights are shown p < .10 p < .05. ** p < .01. *** p < .001
46 Figure 3 2. The interaction between Time 1 c ortisol/DHEAS r atio and Time 1 r idicule in the prediction of change in depressive symptoms Values represent the predicted CDI score when controlling for other variables, including Time 1 CDI score. When ridicule was high, participants with a higher cortisol/DHEAS ratio exhibited higher levels of depressive symptoms than did those with a lower cortisol/DHEAS ratio, p < .01. Error bars indicate the standar d error. 3 3.5 4 4.5 5 5.5 6 6.5 0 1 2 3 4 5 6 7 Change in CDI Score Ridicule High Ratio Low Ratio
47 Table 3 8. Cross sectional regression using Time 1 variables to predict Time 1 aggression, controlling for concurrent depressive symptoms Variable Model 1 Model 2 Model 3 Model 4 Model 5 Step 1 SES .261** .243** .163* .152 .163 Black .017 .002 .044 .045 .054 Hispanic .200* .168 .098 .100 .100 Other .074 .052 .019 .017 .018 Sex .017 .014 .062 .054 .053 T1 Age .103 .116 .062 .061 .051 Step 2 T1 CDI .203* .059 .063 .069 Step 3 T1 Ridicule (R) .040 .040 .047 T1 Social Problems (SP) .424*** .415*** .400*** Step 4 T1 Cortisol/DHEAS .048 .052 T1 Parenting .026 .016 T1 Chaos .052 .046 Step 5 SP*Parenting .063 R 2 .039* .114*** .005 .003 Final R 2 .113* .151 .295 .300 .304 Final Model F 2.772* 3.312** 5.961*** 4.475*** 4.166*** Note: Standardized beta weights are shown p < .10 p < .05. ** p < .01. *** p < .001
48 Table 3 9 Cross s ectional regression using Time 2 variables to predict Time 2 aggression, controlling for concurrent depressive symptoms Variable Model 1 Model 2 Model 3 Model 4 Model 5 Step 1 SES .307 ** .304 ** .235 ** .200 ** .197 Black .023 .011 .036 .080 .076 Hispanic .136 .120 .065 .050 .052 Other .040 .045 .134 .144 .139 Sex .025 .030 .072 .095 .093 T2 Age .080 .052 .168 .140 .139 Step 2 T1 CDI .148 .020 .016 .011 Step 3 T2 Ridicule (R) .021 .036 .034 T2 Social Problems (SP) .521 *** .499 *** .489 *** Step 4 T2 Cortisol/DHEAS .005 .007 T2 Parenting .076 .075 T2 Chaos .157 .161 Step 5 SP* Chaos .036 R 2 .020 .232*** .027 .001 Final R 2 .131 .152 .384 .411 .412 Final Model F 3.271** 3.293** 8.791*** 7.215*** 6.637*** Note: Standardized beta weights are shown p < .10 p < .05. ** p < .01. *** p < .001
49 Table 3 10 Longitudinal regression using Time 1 variables to predict change in aggression, controlling for concurrent depressive symptoms Variable Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Step 1 SES 0.297*** .058 .058 .066 .063 .062 Black 0.030 .018 .009 .001 .014 .013 Hispanic 0.121 .022 .032 .040 .041 .012 Other 0.003 .042 .051 .066 .066 .061 Sex 0.054 .065 .057 .063 .071 .041 T2 Age 0.051 .119 .103 .123 .119 .096 Step 2 T1 Aggression .785** .770*** .751*** .742*** .756*** Step 3 T2 CDI .094 .109 .130* .149* Step 4 T1 Ridicule (R) .105 .108 .086 T1 Social Problems (SP) .058 .043 .040 Step 5 T1 Cortisol/DHEAS .058 .001 T1 Parenting .009 .000 T1 Chaos .048 .048 Step 6 R*Cortisol/DHEAS .117* SP*Cortisol/DHEAS .107 R 2 .543*** .008 .012 .005 .021* Final R 2 .102* .645 .653 .665 .669 .690 Final Model F 2.214* 30.094*** 27.058*** 22.401*** 17.135*** 16.025*** Note : Standardized beta weights are shown p < 1 0 p < .05. ** p < .01. *** p < .001
50 Figure 3 3 The interaction between Time 1 c ortisol/DHEAS Ratio and Time 1 r idicule in the prediction of the change in a ggression Values represent the predicted aggression score when controlling for other variables, including Time 1 aggression score. When ridicule was high, participants with a higher cortisol/DHEAS ratio exhibited higher levels of aggression than did those with a lower cortisol/DHEAS ratio, p < .05. Error bars indicate the standard error. 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 0 1 2 3 4 5 6 7 Change in Aggression Ridicule Low Ratio High Ratio
51 CHAPTER 4 DISCUSSION This study examined parenting characteristics, environmental chaos, and the cortisol/DHEAS ratio as possible moderators of the relationship between social stress and symptoms of psychopathology Peer problems predicted depressive symptoms and aggression in all of the cross sectional analyses, though it did not predict change in symptoms from Time 1 to Time 2. The first hypothesis for this study was that the environmental and biological factors wo uld moderate the effects of peer problems on depressive symptoms or aggression; this hypo thesis was partially supported. Specifically, the interaction between the cortisol/DHEAS ratio and child report of ridicule exhibited a significant moderating effect on peer problems for both depressive and aggressive symptoms in the longitudinal analyses. The second hypothesis in this study was that high cortisol/DHEAS ratios would be predictive of depressive symptoms while low cortisol/DHEAS would be predictive of a ggression. This hypothesis was supported in some analyses. The main effect of the cortisol/DHEAS ratio was only significant in predicting longitudinal change in depressive symptoms, but the interaction of the cortisol/DHEAS ratio with child report of ridi cule was significant in the prediction of change in both aggression and depressive symptoms. P rior research on the cortisol/DHEAS ratio in psychopathology has been limited primarily to high risk or clinical populations (Gallagher & Ritsner, 2009) ; notably the present finding in a community sample suggests that this ratio may be used more broadly as an indicator of physiological vulnerability during the transition into adolescence.
52 Depressive symptoms Many previous studies in young adolescents have found significant associations between problematic peer relationships and depressive symptoms. In the Time 1 and Time 2 cross sectional analyses, maternal report of social problems was a significant predi cto r of depressive symptoms supporting prior research on peer relations and depression ( Boivin, Hymel, & Bukowski, 1995; Panak & Garber, 1992 ) at the younger age (time 1) child report of ridicule was also a significant predictor of concurrent symptoms of de pression. Because of the potential comorbidity of depressive and aggressive symptoms often reported in studies of adolescents, effects of peer problems were tested after controlling for concurrent aggression. Note that in the present study correlations o f aggressive behaviors and depressive symptoms were similar at both times but demonstrated a weak association (Time 1 r = .23, p < .01; Time 2 r = .21, p < .01). Concurrent aggression was a significant predictor of depressive symptoms in the Time 1 cross sectional analysis ; at Time 2, this association was not significant at the .05 level In the longitudinal model, after controlling for previous depressive symptoms, none of the main effects for prior aggression or social problems were present. This indicates that peer problems and concurrent aggression are predictive of depressive symptoms in this sample but not change in symptoms over time. Contrary to previous research, none of the biological or environmental variables had a significant mai n effect on depressive symptoms in the cross sectional analyses, and only the cortisol/DHEAS ratio was significant in the longitudinal analysis. A significant moderating effect was present in the interaction between the cortisol/DHEAS ratio and child repo rt of ridicule in the longitudinal analysis. However, a trend for the interaction between parenting and child report of ridicule was found in the cross sectional
53 Time 2 analysis, suggesting that, as ridicule is more prevalent and parenting is more negativ e, the adolescent experiences higher levels of depressive symptoms. This finding their peers, they would compensate by seeking parental support. If the parent child relationship is also harsh or cold, the adolescent may experience a negative emotional reaction, such as the development of depressive symptoms Aggression As has been reported in prior studies, socioeconomic status was a significant predictor of current aggression as well as change in aggression over time, with lower SES predicting higher aggression. In the Time 1 cross sectional analysis, comorbid depressi ve symptoms predicted higher rates of aggression, as was discussed above, but this relationship was not present in either the Time 2 cross sectional or the longitudinal analyses. This inconsistency may be due to the difference in symptoms; while aggressio n, specifically irritability, is a symptom of depression comorbid depressive symptoms may not be present in adolescents with high levels of aggression As was reported above for depressive symptoms maternal report of social problems was a significant pr edictor of aggressive behaviors in both cross sectional analyses. Interestingly, child report of ridicule was not pre dictive of aggressive behaviors as a main effect. This may be due to the salience of behaviors problems ( Sourander, Helst & Helenius, 1999 ) or to the differences in the types of rejection being assessed in each measure as discussed in the following section N o main effects of the biological or environmental factors were found in the cross sectional or lon gitudinal anal ysis However, the interaction between child report of ridicule and the
54 cortisol/DHEAS ratio was significant in the longitudinal analysis such that, as ridicule increased and the cortisol/DHEAS ratio decreased, adolescents experienced an inc rease in aggressive behavior. The difference between the groups, while significant, is small and is only apparent at higher levels of ridicule. However interaction has not been previously studied in children or adolescent s, the direction of these effects is consistent with prior studies of the cortisol/DHEAS and its role as a predictor of aggression in adolescents ( Buydens Branchey & Branchey, 2004 ; Pajer et al., 2006 ) The direction of the cortisol/DHEAS ratio interaction for aggression is opposite of that for depression, Peer problems Given the absence of correlation between child and mother reports of peer problems, it was not unexpected that the predictive ability of the two would differ While mother report of social problems was a significant predictor in all of the cross sectional analyses for both aggression and depressive symptoms, child report of ridicule was only a significant predictor in the Time 1 cross sectional analysis for depressive symptoms. Interestingly, all of the significant interaction terms included child rather than mother report of peer problems This may be due to the difference in the constructs assessed in each of these measures. Child report of ridicule measures strictly the amount of rejection the child has experienced, while mother report of social problems using the CBCL characteristics. The results fr om this study suggest that the interactions with parenting (CDI Time 2) and the cortisol/DHE AS ratio (longitudinal CDI and A ggression) were dependent on the amount of rejection the child experienced rejection rather than the ot her aspects measured by the CBCL. This may be due to the
55 child experiencing more distress from rejection than from the other aspects of social problems that the CBCL measures. Another possible explanation for these differences is that mothers were not aw are of the extent to which the child experienced r idicule from his or her peers. The social problems subscale scores rated by parents and youth have been shown in many studies to have a low correlation ( Achenbach, McConaughy, & Howell, 1987; Ferdinand, van der Ende, & Verhulst, 2006) suggesting that the parent may be experience of social problems or focusing on social behaviors in different contexts than those reported on by youth Studies have indicated that parent ch ild agreement of social abilities and social problems is greater during middle childhood than during early and middle adolescence (Achenbach, McConaughy, & Howell, 1987; Renk & Phares, 2004) uch problems. Cortisol/DHEAS ratio While Goodyer and his colleagues have previously published studies on the cortisol/DHEAS in adolescents ( see Angold, 2003, for review ), this is the first study to examine this relationship while controlling for puberty. Controlling for puberty in adolescent studies of the cortisol/DHEAS ratio is crucial due to the effects of pubertal development on DHEAS (Matchock, Dorn, & Susman, 2007) As the individual progresses through puberty, DHEAS levels increase significantly; if pubertal stage was not accounted for in the calculation of the ratio, more developed adolescents would have an artificially lower cortisol/DHEAS ratio value In the present study, when DHEAS levels were not adjusted for pubertal stage, effects were no longer significant. As has been found in previous studies in adults ( Michael, Jenaway, Paykel, & Herbert, 2000 ) and adolescents (Goodyer, Herbert, & Tamplin, 2003) the present study found that high
56 cortisol and low DHEAS values at Time 1 were significant predictors of change in depressive symptoms. The interaction between ridicule and the cortisol/DHEAS ratio was also significant in the longitudinal analysis ; adolescents who experienced more ridicule and had high cortisol/DHEAS ratios were more likely to have higher depressive symptoms at Time 2, which was not true of adolescents without high cortisol/DHEAS ratios. While many studies have been published on the effects of the cortisol/DHEAS ratio on depression, very few have examined these effects on aggr essive behaviors. While the main effect of the cortisol/DHEAS ratio in the current study was not significant, contrary to previous research ( Buydens Branchey & Branchey, 2004 ; Pajer et al., 2006), adolescents who had experienced ridicule and had a low cortisol/DHEAS ratio were more likely to display more aggression at Time 2 compared to those without a low cortisol/DHEAS ratio. While the interaction is significant, it is small, and difference between the groups is only apparent at higher levels of ridi cule. examined the effect of the interaction between peer ridicule and the cortisol/DHEAS ratio on aggression. Parenting and Environmental Chaos Though v arious types of chaos including lack of daily rou tine, have been extensively linked to adolescent aggression ( Deater Deckard, Mullineaux, Beekman, Petrill, Schatschneider, & Thompson, 2009 ), this effect was not found in the current study. While t he main effect of environmental chaos in the Time 2 cross sectional analysis was significant this finding cannot be reliably interpreted because the step was not significant None of the interaction terms including chaos were significant, contrary to the hypothesis that chaos would act as a moderator of the effect of peer problems on symptoms of psychopathology One explanation for this lack of significance is that many of t he
57 previous studies were conducted with at risk adolescents while t his study utilized a normative community sample recruited from a working class neighborhood Additionally, because the chaos variable was calculated by summing dichotomous values for four items, it produced a truncated range of scores, possibly affecting the accuracy of the chaos measure. This may be particularly true in adolescents who experienced the highest levels of chaos, who were also expected to be the most affected by chaos. As par t of a large study of early adolescent boys and girls, Sentse and colleagues (2010) found a significant interaction between the quality of parental and peer relationships in the prediction of psychopathology. For both internalizing and externalizing sympt oms, the authors found that adolescents who reported less warmth and more hostility from their parents and who were also rated as the least accepted by their classmates had the highest symptom levels. In the present study, t he interaction between parentin g and child report of ridicule demonstrated a trend for an association in the Time 2 cross sectional prediction of depressive symptoms, but no main effects or interactions were significant for depressive symptoms or aggres sion. Implications and Limitatio ns Despite the strengths of this study, limitations to the study merit discussion. First, a lthough the experimenter left the room during the conflict discussion tasks, the parent presence in the home or the presence of video camera. This could have resulted in the coded interaction styles being atypical for the mother child dyad. Ad ditionally, the full range of parenting behaviors were not captured by the task used, and other parent behaviors may have been more salient to adjustment. Though mothers are not always the primary caregivers, they were used in the parent child interaction task, which could
58 present a different parenting experience than what is typical for the child However, while the mother is not always the primary caregiver, Baumrind (1991) found that mothers and fathers have similar parenting styles in rou ghly 75% of ho useholds studied. As indicated, this study was conducted in a home rather than laboratory settings, and as such there are unique challenges to collecting hormone data. I n an effort to gather accurate awakening cortisol data, the researchers did not speci fy a time for saliva collection Though the collection time was consistent for most of the samples, some samples were collected on weekends, when children may be permitted to sleep later. Samples with drastically different collection times were removed in order to obtain the most accurate values, though this process may not have been sufficient Additionally, because of the typical pattern of cortisol change during sleep, participants who have little consistency in bedtime may have shown greater cortisol variability, even if samples were collected at the same time each morning. Due to the nature of collection, some of the samples may not have been collected or stored according to the Although s amples were eliminated from analysis if problems were identified unidentified problems likely contribute to error variance. Also, p revious studies of the cortisol/DHEAS level have obtained each hormone value from the same source, typically either saliva or blood. While this study u sed different samples for cortisol and DHEAS levels, saliva and urine, respectively, previous studies have indicated that DHEAS levels in overnight urine samples are comparable to those in saliva samples. Although most previous studies on the cortisol/DHE AS ratio have focused on high risk or clinical samples the current study using a community sample, found effects consistent with prior studies
59 Another concern is that the alpha level for CBCL social problems at Time 2 was much lower than at Time 1; th e reason for this difference is unknown, but the findings at both time points were similar. Additionally, both social problems and aggression were measured via maternal report on subscales of the CBCL; this may have caused the relationship to be inflated Inclusion of the child report of rejection offsets concerns about use of the same reporter for predictor and outcome to some extent. B ecause these data are from a community sample in an urban area, findings may not be applicable to a general young adole scent population. T here is also the possibility initiated four years apart, so any substantial events that occurred between the collection Although many studies have demonstrated gender differences during adolescence in key variables used in this study ( see Zahn Waxler, Shirtcliff, & Marceau, 2008, for review ), no gender diff erences were found in the current sample. This may be due to the early stage of adolescence in which this study took place for most participants. G ender differences in aggression and depressive symptoms are typically not as apparent unti l middle or later adolescence and the current sample included individuals in late childhood and early adolescence. Despite these limitations, studying the effects of environmental and hormonal factors on the interaction between social stressors and aggres sion and depressive symptoms may help us to better unders tand why most adolescents experience similar difficulties in social interactions while only a few develop lasting psychosocial difficultie s The results from this study suggest that intervention programs in home or school
60 environments designed to help children and their parents communicate effectively or to teach children the lasting effects of teasing and bullying may help to reduce risk for psychopathology Moderators such as the cortisol/DHEAS ratio may provide a means by which to predict whether an adolescent, upon experiencing significant peer rejection, develops an internalizing or externalizing disorder. The cortisol/DHEAS ratio may be an indicator of a biological vulnerability which emerges during e arly adolescence due to the increasing complexity and significance of peer relationships or due to additional biological or psychosocial changes taking place during this time. Additional research is needed to determine the developmental significance of th e cortisol/DHEAS ratio and the strength of the interaction effect with peer problems across different ages.
61 APPENDIX A CBCL SOCIAL PROBLEMS SUBSCALE 1. Acts too young for his/her age 11. Clings to adults or too dependent 25. other kids 38. Gets teased a lot 48. Not liked by other kids 55. Overweight 62. Poorly coordinated or clumsy 64. Prefers being with younger kids
62 APPENDIX B N INVENTORY 1. I am sad once in a while. I am sad many times. I am sad all the time. 2. Nothing will ever work out for me. I am not sure if things will work out for me. Things will work out for me O.K. 3. I do most things O.K. I do many things wrong. I do everything wrong. 4. I have fun in many things. I have fun in som e things. Nothing is fun at all. 5. I am bad all the time. I am bad many times. I am bad once in a while. 6. I think about bad things happening to me once in a while. I worry that bad things will happen to me. I am sure that terrible things will happen to me. 7. I hate myself. I do not like myself. I like myself. 8. All bad things are my fault. Many bad things are my fault. Bad things are not usually my fault. 9. I feel like crying everyday. I feel like crying many days. I feel like crying once in a while. 10. Things bother me all the time. Things bother me many times. Things bother me once in a while. 11. I like being with people. I do not like being with people many times. I do not like being with people at all. 12. I cannot make up my mind about things. It is hard to make up my mind about things. I make up my mind about things easily. 13. I look O.K. There are some bad things about my looks. I look ugly. 14. I have to push myself all the time to do my schoolwork. I have to push my self many times to do my schoolwork. Doing schoolwork is not a big problem. 15. I have trouble sleeping every night. I have trouble sleeping many nights.
63 I sleep pretty well. 16. I am tired once in a while. I am tired many days. I am tired all the time. 17. Most days I do not feel like eating Many days I do not feel like eating. I eat pretty well. 18. I do not worry about aches and pains. I worry about aches and pains many times. I worry about aches and pains all the time. 19. I do not feel alo ne. I feel alone many times. I feel alone all the time. 20. I never have fun at school. I have fun at school only once in a while. I have fun in school many times. 21. I have plenty of friends. I have some friends, but I wish I had more. I do not ha ve any friends. 22. My school work is alright. My school work is not as good as before. I do very badly in subjects I used to be good in. 23. I can never be as good as other kids. I can be as good as other kids if I want to. I am just as good as other kids. 24. Nobody loves me. I am not sure if anybody loves me. I am sure that somebody loves me. 25. I usually do what I am told. I do not do what I am told most times. I never do what I am told. 26. I get along with people. I get into fights many times. I get into fights all the time.
64 APPENDIX C CBCL AGGRESSIVE BEHA VIORS SUBSCALE 3. Argues a lot 7. Bragging, boasting 16. Cruelty, bullying, or meanness to others 19. Demands a lot of attention 20. Destroys his/her own things 21. Destroys thi ngs belonging to his/her family or others 22. Disobedient at home 23. Disobedient at school 27. Easily jealous 37. Gets in many fights 57. Physically attacks people 68. Screams a lot 74. Showing off or clowning 86. Stubborn, sullen, or irritable 87 Sudden changes in mood or feelings 93. Talks too much 94. Teases a lot 95. Temper tantrums or hot temper 97. Threatens people 104. Unusually loud
65 APPENDIX D BREAKDOWN OF CHAOS V ARIABLES Table D 1. Time 1 Chaos frequencies Frequency Percent Valid Percent Valid .00 82 37.4 43.9 1.00 57 26.0 30.5 2.00 39 17.8 20.9 3.00 7 3.2 3.7 4.00 2 .9 1.1 Total 187 85.4 100.0 Missing System 32 14.6 Total 219 100.0 Table D Frequency Percent Valid Percent Valid Yes 164 74.9 84.5 No 30 13.7 15.5 Total 194 88.6 100.0 Missing System 25 11.4 Total 219 100.0 Table D Frequency Percent Valid Percent Valid Yes 143 65.3 76.1 No 45 20.5 23.9 Total 188 85.8 100.0 Missing System 31 14.2 Total 219 100.0 Table D 4. Time 1 number of places cared for by a non parent Frequency Percent Valid Percent Valid .00 49 22.4 26.5 1.00 94 42.9 50.8 2.00 35 16.0 18.9 3.00 6 2.7 3.2 4.00 1 .5 .5 Missing 34 15.5 Total 219 100.0 100.0
66 Table D 5. Time 1 number of non parent caregivers Frequency Percent Valid Percent Valid .00 62 28.3 33.5 1.00 73 33.3 39.5 2.00 36 16.4 19.5 3.00 9 4.1 4.9 4.00 3 1.4 1.6 5.00 2 .9 1.1 Missing 34 15.5 Total 219 100.0 100.0 Table D 6. Time 2 Chaos frequencies Frequency Percent Valid Percent Valid .00 77 35.2 39.3 1.00 69 31.5 35.2 2.00 42 19.2 21.4 3.00 7 3.2 3.6 4.00 1 .5 .5 Total 196 89.5 100.0 Missing System 23 10.5 Total 219 100.0 Table D Frequency Percent Valid Percent Valid Yes 161 73.5 81.7 No 36 16.4 18.3 Total 197 90.0 100.0 Missing System 22 10.0 Total 219 100.0 Table D Frequency Percent Valid Percent Valid Yes 150 68.5 74.6 No 51 23.3 25.4 Total 201 91.8 100.0 Missing System 18 8.2 Total 219 100.0
67 Table D 9. Time 2 number of places cared for by a non parent Frequency Percent Valid Percent Valid .00 36 16.4 19.1 1.00 113 51.6 60.1 2.00 36 16.4 19.1 3.00 2 .9 1.1 4.00 1 .5 .5 Missing 31 14.2 Total 219 100.0 100.0 Table D 10. Time 2 number of non parent caregivers Frequency Percent Valid Percent Valid .00 58 26.5 30.9 1.00 76 34.7 40.4 2.00 37 16.9 19.7 3.00 13 5.9 6.9 4.00 3 1.4 1.6 5.00 1 .5 .5 Missing 31 14.2 Total 219 100.0 100.0
68 APPENDIX E ADDITIONAL STATISTICS Table E 1. Demographic frequencies Frequency Percent Sex Male 103 47.0 Female 116 53.0 Total 219 100.0 Ethnicity White 74 33.8 Black: African American 49 22.4 Black: Caribbean/Island origin 7 3.2 Black: non specified/other 22 10.0 Asian or Pacific Islander 6 2. 7 Hispanic: Puerto Rican 20 9.1 Hispanic: Dominican 2 .9 Hispanic: non specified/other 9 4.1 other/non specified 7 3.2 Multiracial 23 10.5 Total 219 Table E 2. SES, age, and puberty descriptives N Minimum Maximum Mean Std. Deviation SES 217 9.00 66.00 37.58 13.22 T1 Age 195 8.81 13.86 10.86 .77 Boys 103 9.17 13,86 11.13 .77 Girls 92 8.81 12.16 10.56 .64 T2 Age 202 10.16 14.09 12.04 .77 Boys 91 10.98 14.09 12.28 .69 Girls 111 10.16 13.74 11.84 .78 T1 Tanner Score 190 1.00 5.00 2.12 .95 Boys 100 1.00 5.00 1.93 .86 Girls 90 1.00 4.50 2.33 1.00 T2 Tanner Score 198 1.00 5.00 2.82 1.00 Boys 87 1.00 5.00 2,72 .91 Girls 111 1.00 5.00 2.91 1.07
69 Table E 3. Means and s tandard d eviations for Time 1 p rimary v ariables Variable n Min Max M SD T1 Aggression 190 .00 26.00 6.65 5.70 T1 Aggression Girls 89 .00 26.00 6.58 5.68 T1 Aggression Boys 101 .00 26.32 6.09 5.43 T1 CDI (Adjusted) a 182 .00 39.00 7.02 6.33 T1 CDI Girls (Adj) 88 .00 32.00 6.98 6.68 T1 CDI Boys (Adj) 94 .00 39.00 7.06 6.02 T1 Ridicule 190 .00 7.00 1.16 1.36 T1 CBCL Social Problems 190 .00 15.00 1.83 2.29 T1 Cortisol/DHEAS b 163 .47 .61 .50 .02 T1 Parenting c 188 2.00 6.00 4.06 1.06 T1 Chaos d 187 .00 4.00 .88 .94 a CDI scores adjusted to reflect removal of suicidality item b lower values indicate lower cortisol and higher DHEAS values c higher values indicate more negative parenting characteristics d higher values indicate higher levels of chaos Table E 4. Means and s tandard de viations for Time 2 p rimary v ariables Variable n Min Max M SD T2 Aggression 195 .00 29.00 5.96 5.32 T2 Aggression Girls 108 .00 29.00 5.85 5.25 T2 Aggression Boys 87 .00 26.32 6.09 5.43 T2 CDI (Adjusted) a 184 .00 26.00 6.04 5.71 T2 CDI Girls (Adj) 103 .00 25.00 6.08 5.90 T2 CDI Boys (Adj) 81 .00 26.00 6.52 5.48 T2 Ridicule 191 .00 6.00 1.16 1.30 T2 CBCL Social Problems 195 .00 9.00 1.61 1.83 T2 Cortisol/DHEAS b 173 .45 .80 .50 .03 T2 Parenting c 182 2.00 6.00 3.88 .97 T2 Chaos d 196 .00 4.00 .91 .89 a CDI scores adjusted to reflect removal of suicidality item b lower values indicate lower cortisol and higher DHEAS values c higher values indicate more negative parenting characteristics d higher values indicate higher levels of chaos
70 Table E 5. Cross Sectional Correlations of Time 1 Variables Variable 1 2 3 4 5 6 7 T1 Aggression T1 CDI .227** T1 Ridicule .073 .266*** T1 CBCL Social Problems .512*** .239** .110 T1 Cortisol/DHEAS .102 .088 .017 .116 T1 Parenting .027 .094 .062 .012 .094 T1 Chaos .076 .008 .019 .069 .039 .167* p < .05. ** p < .01. *** p < .001 Table E 6. Cross Sectional Correlations of Time 2 Variables Variable 1 2 3 4 5 6 7 T2 Aggression T2 CDI .211** T2 Ridicule .033 .182* T2 CBCL Social Problems .523*** .241** .108 T2 Cortisol/DHEAS .003 .035 .127 .016 T2 Parenting .094 .025 .148 .008 .038 T2 Chaos .277*** .161* .127 .093 .100 .146 p < .01. p < .05. ** p < .01. *** p < .001 Table E 7. Longitudinal Correlations of Time 1 Predictors and Time 2 Outcome Variables Variable 1 2 3 4 5 6 7 T2 Aggression T2 CDI .211** T1 Ridicule .065 .264** T1 CBCL Social Problems .480*** .260** .110 T1 Cortisol/DHEAS .079 .236** .017 .116 T1 Parenting .018 .177* .062 .012 .094 T1 Chaos .129 .075 .019 .069 .039 .167* p < .01. p < .05. ** p < .01. *** p < .001
71 Table E 8. Frequency of maternal warmth at Time 1 Frequency Percent Valid Percent Cumulative Percent Valid 1.00 32 14.6 17.0 17.0 2.00 43 19.6 22.9 39.9 3.00 65 29.7 34.6 74.5 4.00 42 19.2 22.3 96.8 5.00 6 2.7 3.2 100.0 Total 188 85.8 100.0 Missing System 31 14.2 Total 219 100.0 Table E 9 Frequency of maternal hostility at Time 1 Frequency Percent Valid Percent Cumulative Percent Valid 1.00 150 68.5 79.8 79.8 2.00 20 9.1 10.6 90.4 3.00 13 5.9 6.9 97.3 4.00 2 .9 1.1 98.4 5.00 3 1.4 1.6 100.0 Total 188 85.8 100.0 Missing System 31 14.2 Total 219 100.0 Table E 10 Frequency of maternal warmth at Time 2 Frequency Percent Valid Percent Cumulative Percent Valid 1.00 35 16.0 19.2 19.2 2.00 58 26.5 31.9 51.1 3.00 59 26.9 32.4 83.5 4.00 23 10.5 12.6 96.2 5.00 7 3.2 3.8 100.0 Total 182 83.1 100.0 Missing System 37 16.9 Total 219 100.0 Table E 11 Frequency of maternal hostility at Time 2 Frequency Percent Valid Percent Cumulative Percent Valid 1.00 136 62.1 74.7 74.7 2.00 29 13.2 15.9 90.7 3.00 11 5.0 6.0 96.7 4.00 6 2.7 3.3 100.0 Total 182 83.1 100.0 Missing System 37 16.9 Total 219 100.0
72 LIST OF REFERENCES Aboud, F.E. & Mendelson, M.J. (1998). Determinants of friendship selection and quality: Developmental perspectives. In W.M. Bukowski, A.F. Newcomb, & W.W. Hartup (Eds.), The Company They Keep: Friendship in Childhood and Adolescence (87 111). New York: Cambridge University Press. Achenbach, T.M. (1991). Manual for the Child Behavior Checklist 4 18 Burlington, VT: University of Vermont, Department of Psychiatry Achenbac h, T.M. (1991). Manual for the Youth Self Report for Ages 11 18 and 1991 Profile. Burlington, VT: University of Vermont, Department of Psychiatry. Achenbach, T.M. & Edelbock C.S. (1981). Behavioral problems and competencies reported by parents of normal and disturbed children aged four through sixteen. Monographs of the Society for Research in Child Development, 46(1), 1 82. Achenbach, T. M., McConaughy, S. H., & Howell, C. T. (1987). Child/adolescent behavioral and emotional problems: Implications of cross informant correlations for situational specificity. Psychological Bulletin, 101 (2), 213 232. doi:10.1037/0033 2909.101.2.213 Almqvist, F., Kumpulainen, K., Ik heimo Moilanen, I. (1999). Behavioural and emotional symptoms in 8 9 year old children. European Child and Adolescent Psychiatry, 8, IV8 IV16. Angold, A. (2003). Adolescent depression, cortisol, and DHEA. Psychol ogical Medicine, 33, 573 581. doi: 10.1017/S003329170300775X Baron, R.M. & Kenny, D.A. (1986). The moderator mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173 1182. doi: 10.1037/0022 3518.104.22.1683 Baumrind, D. (1991). Parenting styles and adolescent development. In J. Brooks Gunn, R. Lerner, & A.C. Peterson (Eds.), The encyclopedia of adolescence. New York: Garland. Bjrkqvist, K., Lagerspetz, K.M.J., & Kaukiainen, A. (1992). Do girls manipulate and boys fight? Developmental trends in regard to direct and indirect aggression. Aggressive Behavior, 18, 117 127. doi: 10.1002/1098 2337(1992)18:2<117::AID AB2480180205>3.0.CO;2 3 Blackha rt, G.C., Eckel, L.A., & Tice, D.M. (2007). Salivary cortisol in response to acute social rejection and acceptance by peers. Biological Psychology, 75, 267 276. doi: 10.1016/j.biopsycho.2007.03.005
73 Boivin, M., Hymel, S., & Bukowski, W.M. (1995). The roles of social withdrawal, peer rejection, and victimization by peers in predicting loneliness and depressed mood in childhood. Development & Psychopathology, 7, 765 785. doi: 10.1017/S0954579400006830 Bongers, I. L., Koot, H. M., van der Ende, J., & Verhulst F. C. (2003). The normative development of child and adolescent problem behavior. Journal of Abnormal Psychology, 112 179 192. doi:10.1037/0021 843X.112.2.179 Bowes, L., Maughan, B., Caspi, A., Moffitt, T.E., & Arseneault, L. (2010). Families promote e motional and behavioral resilience to bullying: Evidence of an environmental effect. Journal of Child Psychology & Psychiatry, 51, 809 817 doi: 10.1111/j.1469 7610.2010.02216.x Brooks Gunn, J. (1991). How stressful is the transition to adolescence in gir ls? In M.E. Colten & S. Gore (Eds.), Adolescent Stress: Causes and Consequences (pp. 131 149). Hawthorne, NY: Aldine de Gruyter. Brown, B.B. & Larson, J. (2009). Peer relationships in adolescence. In R.M. Lerner & L. Steinberg (Eds.), Handbook of Adolescent Psychology, Third Edition (74 103). Hoboken, New Jersey: John Wiley & Sons, Inc. Buydens Branchey, L. & Branchey, M. (2004). Cocaine addicts with conduct disorder are typified by decreased cortisol responsivity and high plasma levels of DHEA S. Neuropsychobiology, 50, 161 166. doi: 10.1159/000079109 Card, N. A., Stucky, B. D., Sawalani, G. M., & Little, T. D. (2008). Direct and indirect aggression during childhood and adolescence: A meta analytic review of gender differences, intercorrelations, and relations to maladjustment. Child Development, 79, 1185 1229. Centers for Disease Control and Prevention. (2010). 2009 National Youth Risk Behavior Survey [Data file]. Retrieved from http://www.cdc.gov/HealthyYouth/yrbs/data/index.htm Chorpita, B. F., & Barlow, D. H. (1998). The development of anxiety: The role of control in the early environment. Psychological Bulletin, 124, 3 21. Coie, J.D., Lochman, J.E., Terry, R., & Hyman, C. (1992). Predicting early adolescent disorder from childhood aggression and peer rejection. Journal of Consulting & Clinical Psychology, 60, 783 792. doi: 10.1037/0022 006X.60.5.783 Cole, D.A. & Rehm, L.P. (1986). Family interaction patterns and childhood depression. Journal of Abnormal Child Psychology, 14, 297 314. doi: 10.1007/BF00915448
74 Cyranowski, J.M., Frank, E., Young, E., & Shear, K. (2000). Adolescent onset of the gender difference in lifetime rates of major depression: A theoretical model. Archiv es of General Psychiatry, 57, 21 27. Deater Deckard, K. Millineaux, P.Y., Beekman, C., Petrill, S.A., Schatschneider, C., & Thompson, L.A. (2009). Conduct problems, IQ, and household chaos: a longitudinal multi informant study. Journal of Child Psycholog y and Psychiatry, 50, 1301 1308. doi: 10.1111/j.1469 7610.2009.02108.x DeVore, E.R. (2005). The protective effects of good parenting on adolescents. Current Opinions in Pediatrics, 17, 460 464. doi: 10.1097/01.mop.0000170514.27649.c9 Dickerson, S.S. & K emeny, M.E. (2004). Acute stressors and cortisol responses: A theoretical integration and synthesis of laboratory research. Psychological Bulletin, 130, 355 391. doi: 10.1037/0033 2909.130.3.355 Further neuroendocrine evidence of enhanced vasopressin V3 receptor responses in melancholic depression. Psychological Medicine, 34, 169 172. doi: 10.1017/S0033291703001004 Ellenbogen, M.A. & Hodgins, S. (2009). Structure provided by parents in middle ch ildhood predicts cortisol reactivity in adolescence among the offspring of parents with bipolar disorder and controls. Psychoneuroendocrinology, 34, 773 785. doi: 10.1016/j.psyneuen.2008.12.011 Evans, G.W., Gonnella, C., Marcynyszyn, L.A., Gentile, L., & Salpekar, N. (2005). The Psychological Science, 16, 560 565. doi: 10.1111/j.0956 7976.2005.01575.x Evans, G.W. (2006). Child development and the physical environment. Annual Review of Psyc hology, 57, 423 451. doi: : 10.1146/annurev.psych.57.102904.190057 Ferdinand, R.F., van der Ende, J., & Verhulst, F.C. (2006). Prognostic value of parent adolescent disagreement in a referred sample. European Child & Adolescent Psychiatry, 113, 198 206. d oi: 10.1037/0021 843X.113.2.198 Fiese, B.H., Tomocho, T.J., Douglas, M., Josephs, K., Poltrock, S., & Baker, T. (2002). A review of 50 years of research on naturally occurring family routines and rituals: Cause for celebration? Journal of Family Psycholog y, 16, 381 390. doi: 10.1037//0893 322.214.171.124 Flinn, M.V. & England, B.G. (1995). Childhood stress and family environment. Current Anthropology, 36, 854 866.
75 Ford, M.B. & Collins, N.L. (2010). Self esteem moderates neuroendocrine and psychological resp onses to interpersonal rejection. Journal of Personality & Social Psychology, 98, 405 419. doi: 10.1037/a0017345 in their social networks. Developmental Psychology, 2 1, 1016 1024. doi: 10.1037/0012 16126.96.36.1996 Furman, W. & Buhrmester, D. (1992). Age and sex differences in perceptions of networks of personal relationships. Child Development, 63, 103 115. doi: 10.1111/j.1467 8624.1992.tb03599.x Gallagher, P. & Ritsn er, M.S. (2009). Can the cortisol to DHEA molar ratio be used as a peripheral biomarker for schizophrenia and mood disorders? In M.S. Ritsner (Ed.) The Handbook of Neuropsychiatric Biomarkers, Endophenotypes and Genes (pp.27 45). DOI : 10.1007/978 1 4020 9838 3_3 Gauze, C., Bukowski, W.M., Aquan Assee, J., & Sippola, L.K. (1996). Interactions between family environment and friendship and associations with self perceived well being during early adolescence. Child Development, 67, 2201 22 16. doi: 10.1111/j.1467 8624.1996.tb01852.x Ge, X., Best, K.M., Conger, R.D., & Simons, R.L. (1996). Parenting behaviors and the occurrence and co occurrence of adolescent depressive symptoms and conduct problems. Developmental Psychology, 32, 717 731. do i: 10.1037/0012 16188.8.131.527 Goodyer, I.M., Herbert, J., Altham, P.M.E., Pearson, J., Secher, S.M., & Shiers, H.M. (1996). Adrenal secretions during major depression in 8 to 16 year olds, I. Altered diurnal rhythms in salivary cortisol and dehydroepiand rosterone (DHEA) at presentation. Psychological Medicine, 26, 245 256. doi: 10.1017/S0033291700034644 Goodyer, I.M., Herbert, J., & Tamplin, A. (2003). Psychoendocrine antecedents of persistent first episode major depression in adolescents: A community b ased longitudinal study. Psychological Medicine, 33, 601 610. doi: 10.1017/S0033291702007286 Goodyer, I.M., Park, R.J., Netherton, C.M., & Herbert, J. (2001). Possible role of cortisol and dehydroepiandrosterone in human development and psychopathology. British Journal of Psychiatry, 179, 243 249. Gordon, K.M., Peloso, E., Auker, A., & Dozier, M. (2005). Effect of flavored beverage crystals on salivary cortisol enzyme immunoreactive assay mea surements. Developmental Psychobiology, 47, 189 195. doi: 10.1 002/dev.20081
76 Graber, J.A., Brooks Gunn, J., & Warren, M.P. (2006). Pubertal effects on adjustment in girls: Moving from demonstrating effects to identifying pathways. Journal of Youth and Adolescents, 35, 391 401. doi: 10.1007/s10964 006 9049 2 Graber, J.A., Nichols, T., Luppino, C., Galen, B., Brooks Gunn, J., Schinke, S., & Botvin, G.J. (1999). Life skill training: Videotape role play codebook. Unpublished manuscript, Center for Children & Families, Teachers College, Columbia University. Grang er, D.A., Weisz, J.R., & Kauneckis, D. (1994). Neuroendocrine reactivity, internalizing behavior problems, and control related cognitions in clinic referred children and adolescents. Journal of Abnormal Psychology, 103, 267 276. Gunnar, M.R., Brodersen, L ., Nachimas, M., Buss, K., & Rigatuso, J. (1996). Stress reactivity and attachment security. Developmental Psychobiology, 29, 191 204. doi: 10.1002/(SICI)1098 2302(199604)29:3<191::AID DEV1>3.0.CO;2 M Gunnar, M. R., Gonzalez, C. A., Goodlin, B. L., & Levi ne, S. (1981). Behavioral and pituitary adrenal responses during a prolonged separation period in rhesus monkeys. Psychoneuroendocrinology, 6, 65 75. Gunnar, M.R., Sebanc, A.M., Tout, K., Donzella, B., & van Dulmen, M.M.H. (2003). Peer rejection, temperam ent, and cortisol reactivity in preschoolers. Developmental Psychobiology, 43, 346 368. doi: 10.1002/dev.10144. Harter, S. (2001). The Construction of the Self: A Developmental Perspective. New York, NY: The Guilford Press. Hartup, W.W. (1996). The compa ny they keep: Friendships and their developmental significance. Child Development, 67, 1 13. doi: 10.1111/j.1467 8624.1996.tb01714.x Heinrichs, M. & Domes, G. (2008). Neuropeptides and social behavior: Effects of oxytocin and vasopressin in humans. Progre ss in Brain Research, 170, 337 350. doi: 10.1016/S0079 6123(08)00428 7 Hertsgaard, L., Gunnar, M., Larson, M., Brodersen, L., & Lehman, H. (1992). First time experiences in infancy: When they appear to be pleasant, do they activate the adrenocortical stre ss response? Developmental Psychobiology, 25, 319 333. doi: 10.1002/dev.420250503 Hetherington, E.M. & Clingempeel, W.G. (1992). Coping with martial transitions: A family systems perspective. Monographs of the Society for Research on Child Development, 57 1 238. Ho, C., Bluestein, D.N., & Jenkins, J.M. (2008). Cultural differences in the relationship 522. doi: 10.1037/0012 16184.108.40.2067
77 Hollingshead, A.B. (1975). Four Factor Index of Social Status. Unpublished Manuscript, Department of Sociology, Yale University, New Haven, Connecticut, U.S.A. Johnson, A. D., Martin, A., Brooks Gunn, J., & Petrill, S. A. (2008). Order in the house! Associations among household cha os, the home literacy environment, maternal reading ability, and children's early reading. Merrill Palmer Quarterly: Journal of Developmental Psychology 54 445 472. doi:10.1353/mpq.0.0009 Kaminska, M., Harris, J.H., Gijsbers, K., & Dubrovsky, B. (2000). Dehydroepiandrosterone sulfate (DHEAS) counteracts detrimental effects of corticosterone on dentate gyrus LTP: Implications for depression. Brain Research Bulletin, 52, 229 234. doi: 10.1016/S0361 9230(00)00251 3 Kliewer, W. & Kluger, E. (1998). Family m oderators of the relation between hassles and behavior problems in inner city youth. Journal of Clinical Child Psychology, 27, 278 292. doi: 10.1207/s15374424jccp2703_5 Kovacs, M. (1980). Rating scales to assess depression in school aged children. Acta P aedopsychiatrica: International Journal of Child & Adolescent Psychiatry, 46, 305 315. Kovacs, M. (1982). New York: Multi Health Systems, Inc. Labrie, F., Blanger, A., Cusan, L., Gomez, J.L., & Candas B. (1997). Marked decline in serum concentrations of adrenal C19 sex steroid precursors and conjugated androgen metabolites during aging. The Journal of Clinical Endocrinology and Metabolism, 82, 2396 2402. LaGreca, A.M. & Harrison, H.M. (2005). Adoles cent peer relationships, friendships, and romantic relationships: Do they predict social anxiety and depression? Journal of Clinical Child & Adolescent Psychology, 34, 49 61. doi: 10.1207/s15374424jccp3401_5 Laird, R.D., Jordan, K.Y., Dodge, K.A., Pettit, G.S., & Bates, J.E. (2001). Peer rejection in childhood, involvement with antisocial peers in early adolescence, and the development of externalizing behavior problems. Development and Psychopathology, 13, 337 354. doi: 10.1017/S0954579401002085 Larson, R.W. & Richards, M.H. (1991). Daily companionship in late childhood and early adolescence: Changing developmental contexts. Child Development, 62, 284 300. doi: 10.1111/j.1467 8624.1991.tb01531.x
78 Larson, R.W. & Verma, S. (1999). How children and adolescen ts spend time across the world: Work, play, and developmental opportunities. Psychological Bulletin, 125, 701 736. doi: 10.1037/0033 2909.125.6.701 Larsson, B. & Melin, L. (1992). Prevalence and short term stability of depressive symptoms in schoolchildre n. Acta Psychiatrica Scandinavica, 85, 17 22. doi: 10.1111/j.1600 0447.1992.tb01436.x Lephart, E.D., Baxter, C.R., & Parker, C.R. Jr. (1987). Effect of burn trauma on adrenal and testicular steroid hormone production. Journal of Clinical Endocrinology and Metabolism, 64, 842 848. Lewinsohn, P.M., Hops, H., Roberts, R.E., Seeley, J.R., & Andrews, J.A. (1993). Adolescent psychopathology: I. Prevalence and incidence of depression and other DSM III R disorders in high school students. Journal of Abnormal Psychology, 102, 133 144. doi: 10.1037/0021 843X.102.1.133 Lopez Duran, N.L., Kovacs, M., & George, C.J. (2009). Hypothalamic pituitary adrenal axis dysregulation in depressed children and adolescen ts: A meta analysis. Psychoneuroendocrinology, 34, 1272 1283. doi: 10.1016/j.psyneuen.2009.03.016 Luppa, P., Munker, R., Nagel, D., Weber, M., & Englehardt, D. (1991). Serum androgens in intensive care patients: Correlations with clinical findings. Clinic al Endocrinology, 34, 305 310. Ma, X.M. & Lightman, S.L. (1998). The arginine vasopressin and corticotropin releasing hormone gene transcription responses to varied frequencies of repeated stress in rats. The Journal of Physiology, 510, 605 614. doi: 10.1 111/j.1469 7793.1998.605bk.x MacMillan, H.L., Georgiades, K., Duku, E.K., Shea, A., Steiner, M., Niec, A., Tanaka, M., et al. (2009). Cortisol response to stress in female youths exposed to childhood maltreatment: results of the youth mood project. Biolog ical Psychiatry, 66, 62 68. doi: 10.1016/j.biopsych.2008.12.014 Majewska, M.D. (1992). Neurosteroids: Endogenous bimodal modulators of the GABAA receptor. Mechanisms of action and physiological significance. Progress in Neurobiology, 38, 379 395. Maninger, N., Wolkowitz, O.M., Reus, V.I., Epel, E.S., & Mellon, S.H. (2009). Neurobiological and neuropsychiatric effects of dehydroepiandrosterone (DHEA) and DHEA sulfate (DHEAS). Frontiers in Neuroscience, 30, 65 91. doi: 10.1016/j.yfrne.2008.11.002 Ma rshall, W.A. & Tanner, J.M. (1969). Variations in the pattern of pubertal changes in girls. Archives of Disease in Childhood, 44, 291 303.
79 Marshall, W.A. & Tanner, J.M. (1970). Variations in the pattern of pubertal changes in boys. Archives of Disease in Childhood, 45, 13 23. Matchock, R.L., Dorn, L.D., & Susman, E.J. (2007). Diurnal and seasonal cortisol, testosterone, and DHEA rhythms in boys and girls duri ng puberty. Chronobiology International, 24, 969 990. doi: 10.1080/07420520701649471 Matheny, A.P ., Wachs, T.D., Ludwig, J.L., & Phillips, K. (1995). Bringing order out of chaos: Psychometric characteristics of the confusion, hubbub, and order scale. Journal of Applied Developmental Psychology, 16, 429 444. doi: 10.1016/0193 3973(95)90028 4 Meador Wo odruff, J.H., Greden, J.F., Grunhaus, L., & Haskett, R.F. (1990). Severity of depression and hypothalamic pituitary adrenal axis dysregulation: Identification of contributing factors. Acta Psychiatrica Scandinavia, 81, 364 371. Meaney, M.J., Aitken, D.H. van Berkel, C., Bhatnagar, S., & Sapolsky, R.M. (1988). Effect of neonatal handling on age related impairments associated with the hippocampus. Science, 239, 766 768. doi: 10.1126/science.3340858 Melby, J.N., Conger, R.D., Book, R., Reuter, M., Lucy, L. Reinski, D. et al. (1998). The Iowa Family Interaction Rating Scales (5 th ed.). Unpublished manuscript, Institute for Social and Behavioral Research, Iowa State University. Michael, A., Jenaway, A., Paykel, E.S., & Herbert, J. (2000). Altered salivary dehydroepiandrosterone levels in major depression in adults. Biological Psychiatry, 48, 989 995. doi: 10.1016/S0006 3223(00)00955 0 Morris, N.M. & Udry, J.R. (1980). Validation of a self administered instrument to assess stage of adolescent development. Journal of Youth and Adolescence, 9, 271 280. doi : 10.1007/BF02088471 Nachmias, M., Gunnar, M.R., Mangelsdorf, S., Parritz, R.H., & Buss, K. (1996). Behavioral inhibition and stress reactivity: The moderating role of attachment security. Child Development 67, 508 522. doi: 10.1111/j.1467 8624.1996.tb01748.x Nangle, D.W., Erdley, C.A., Newman, J.E., Mason, C.A., & Carpenter, E.M. (2003). Popularity, friendship quantity, and friendship quality: Interactive influences on Journal of Clinical Child & Adolescent Psychology, 32, 546 555. doi: 10.1207/S15374424JCCP3204_7 Nolen Hoeksema, S., Girgus, J.S., Seligman, M.E. (1992). Predictors and consequences of childhood depressive symptoms: A 5 year longitudinal study. Journal of Abnormal Psychology, 101, 405 422.
80 Ozbay, F., Fitterling, H., Charney, D., & Southwick, S. (2008). Social support and resilience to stress across the life span: A neurobiologic framework. Current Psychiatry Reports, 10, 304 310. doi: 10.1007/s11920 00 8 0049 7 Pajer, K., Tabbah, R., Gardner, W., Rubin, R.T., Czambel, R.K., & Wang, Y. (2006). Adrenal androgen and gonadal hormone levels in adolescent girls with conduct disorder. Psychoneuroendocrinology, 31, 1245 1256. doi: 10.1016/j.psyneuen.2006.09.005 Panak, W.F. & Garber, J. (1992). Role of aggression, rejection and attributions in the prediction of depression in children. Development & Psychopathology, 4, 145 165. doi: 10.1017/S095457940000561 Parker, L.N., Levin, E.R., & Lifrak E.T. (1985). Evidence for adrenocortical adaptation to severe illness. Journal of Clinical Endocrinology and Metabolism, 60, 947 952. Patterson, C.J., Cohn, D.A., & Kao, B.T. (1989). Maternal warmth as a protective factor against risk associated with pe er rejection among children. Development & Psychopathology, 1, 21 38. doi: 10.1017/S0954579400000225 Plunkett, S.W., Henry, C.S., Robinson, L.C., Behnke, A., & Falcon, P.C. (2007). Adolescent perceptions of parental behaviors, adolescent self esteem, and adolescent depressed mood. Journal of Child & Family Studies, 16, 760 772. doi: 10.1007/s10826 006 9123 0 Renk, K. & Phares, V. (2004). Cross informant ratings of social competence in children and adolescents. Clinical Psychology Review, 24, 239 254. doi: 10.1016/j.cpr.2004.01.004 Repetti, R. L., Taylor, S. E., & Seeman, T. E. (2002). Risky families: Family social environments and the mental and physical health of offspring. Psychological Bulletin, 128 330 366. doi:10.1037/0033 2909.128.2.330 Rigby, K. (2000). Effects of peer victimization in schools and perceived social support on adolescent well being. Journal of Adolescence, 23, 57 68. doi: 10.1006/jado.1999.0289 Rowe, R., Maughan, B., & Eley, T.C., (2006). Links between antisocial behavior and depressed mood: The role of life events and attributional style. Journal of Abnormal Child Psychology, 34, 283 292. doi : 10.1007/s10802 006 9032 0 Sapolsky, R.M. (1993). Endocrinology alfresco: Psychoendocrine studies of wild baboons. Recent Progress in Hormone Research, 48, 437 468.
81 Schwartz, E.B., Granger, D.A., Susman, E.J., Gunnar, M.R., & Laird, B. (1998). Assessing salivary cortisol in studies of child development. Child Development, 69, 1503 1513. doi: 10.1111/j.1467 8624.1998.t b06173.x Sentse, M., Lindenberg, S., Omvlee, A., Ormel, J., & Veenstra, R. (2010). Rejection and acceptance across contexts: Parents and peers as risks and buffers for early adolescent psychopathology. The TRAILS study. Journal of Abnormal Child Psycholog y, 38, 119 130. doi: 10.1007/s10802 009 9351 z relationships with their mothers and fathers: Associations with depressive disorder and subdiagnostic symptomatology. Journal of Abnormal Psychology, 116, 144 154. doi: 10.1037/0021 843X.116.1.144 Skara, S., Pokhrel, P., Weiner, M.D., Sun, P., Dent, C.W., & Sussman, S. (2008). Physical and relational aggression as predictors of drug use: Gender differences among high school stu dents. Addictive Behaviors, 33, 1507 1515. doi: 10.1016/j.addbeh.2008.05.014 Smucker, M.R., Craighead, W.E., Craighead, L.W., & Green, B.J. (1986). Normative and Journal of Abnormal Child Psycholog y, 14, 25 3 9 doi: 10.1007/BF00917219 Soma, K.K., Scotti, M.A.L., Newman, A.E.M., Charlier, T.D., & Demas, G.E. (2008). Novel mechanisms for neuroendocrine regulation of aggression. Frontiers in Neuroendocrinology, 29, 476 489. doi: 10.1016/j.yfrne.2007.1 2.003 Sontag, L.M. & Graber, J.A. (2010). Coping with perceived peer stress: Gender specific and common pathways to symptoms of psychopatholo gy. Developmental Psychopathology, 46, 1605 1620. doi: 10.1037/a0020617 Sourander, A., Helstel L., & Helenius H. (1999). Parent adolescent agreement on emotional and behavioral problems. Social Psychiatry and Psychiatric Epidemiology, 34, 657 663 doi: 10.1007/s001270050189 Steinberg, L., Mounts, N., Lamborn, S., & Dornbusch, S. (1991). Authoritative parenting and adolescent adjustment across various ecological niches. Journal of Research on Adolescents, 1, 19 36. doi: 10.1111/1532 7795.ep11522650 Sullivan, T.N., Farrell, A.D., & Kliewer, W. (2006). Peer victimization in early adolescence: Association between p hysical and relational victimization and drug use, aggression, and delinquent behaviors among urban high school students. Development & Psychopathology, 18, 119 137. doi: 10.1017/S095457940606007X
82 Talge, N.M., Donzella, B., Kryzer, E.M., Gierens, A., & bad: Error introduced by oral stimulants in salivary cortisol research. Developmental Psychobiology, 47, 369 376. doi : 10.1002/dev.20097 Twenge, J.M. & Nolen Hoeksema S. (2002). Age, gender, race, socioeconomic status, analysis. Journal of Abnormal Psychology, 111, 578 588. doi: 10.1037/0021 843X.111.4.578 van Goozen, S.H.M., Matthys, W., Cohe n Kettenis, P.T., Gispen de Weid, C., Weigant, V.M., & van Engeland, H. (1998). Salivary cortisol and cardiovascular activity during stress in oppositional defiant disorder boys and normal controls. Biological Psychiatry, 43, 531 539. Van Goozen, S.H.M., Matthys, W., Cohen Kettenis, P.T., Thijssen, J.H.H., & van England, H. (1998). Adrenal androgens and aggression in conduct disorder: Prepubertal boys and normal controls. Biological Psychiatry, 43, 156 158. doi: 10.1016/S0006 3223(98)00360 6 Van Goozen S.H.M., Van Den Ban, E.L.S., Matthys, W., Cohen Kettenis, P.T., Thijssen, J.H.H., & van England, H. (2000). Increased adrenal androgen functioning in children with oppositional defiant disorder: A comparison with psychiatric and normal controls. Journal of the American Academy of Child & Adolescent Psychiatry, 39, 1446 1451. doi: 10.1097/00004583 200011000 00020 Wade, C.E., Lindberg, J.S., Cockrell, J.L., Lamiell, J.M., Hunt, M.M., Ducey, J., et al. (1988). Upon admission adrenal steroidogenesis is adapt ed to the degree of illness in intensive care unit patients. Journal of Clinical Endocrinology and Metabolism, 67, 223 227. Wachs, T.D. & Corapci, F. (2005). Environmental chaos, development and parenting across cultures. In C. Raeff & J.B. Benson (Eds.) Social and cognitive development in the context of individual, social, and cultural processes (pp.54 83). New York, NY: Routledge. Weiss, R.S. (1974). The provisions of social relationships. In Z. Rubin (Ed.) Doing Unto Others Englewood Cliffs, NJ: Prent ice Hall. Werner, N.E. & Crick, N.R. (2004). Maladaptive peer relationships and the development of relational and physical aggression during middle childhood. Social Development, 13, 495 514. doi: 1 0.1111/j.1467 9507.2004.00280.x Wolf, O. T., & Kirschbaum, C. (1999). Actions of dehydroepiandrosterone and its sulfate in the central nervous system: Effects on cognition and emotion in animals and humans. Brain Research Reviews, 30 264 288. doi: 10.1016/S0165 0173(99)00021 1
83 Yu, S., Holsboer, F., & Almedia, O.F.X. (2008). Neuronal actions of glucocorticoids: Focus on depression. The Journal of Steroid Biochemistry & Molecular Biology, 108, 300 309. doi: 10.1016/j.jsbmb.2007.09.014 Zahn Waxler, C., Shirtcliff, E.A., & Marceau, K. (2008). Disorders of childhood and adolescence: Gender and psychopathology. Annual Review of Clinical Psychology, 4, 275 303 doi: 10.1146/annurev.clinpsy.3.022806.091358 Zadeh, Z.Y., Jenkins, J., & Pepler D. (2010). A transactional analysis of maternal negativity and child externalizing behavior. International Journal of Behavioral Development, 34, 218 228. doi: 10.1177/0165025409350953
84 BIOGRAPHICAL SKETCH Mary Eileen Saczawa was born in Huntsville, Alabama as the youngest of two children. She graduated from East Limestone High School in Athens, Alabama in 2004. She attended Oxford College of Emory University, where she received her A.A in 2006 before con tinuing on to Emory College to receive her B.S. in n euroscience and b ehavioral b iology in 2007. While completing her undergraduate education and for a year and a half after graduating, she worked as a clinical research coordinator at the Institute for Beh avioral Medicine in Smyrna, Georgia. She moved to Florida in 2009 to pursue a doctorate in developmental psychology at the University of Florida.