Peer Relationships and ADHD 1 Anxiety in Children with Attention-Deficit/Hyperactivity Disorder: Influences on Peer Relationships and Social Skills

This investigation examined the associations of anxiety symptoms with peer relationships and social skills in a sample of children ages 6-12 with ADHD (n = 61; 42 boys, 19 girls), and age and sex-matched comparison children (n = 60; 41 boys, 19 girls). After controlling for ADHD and ODD diagnoses (present in 32% of children with ADHD and no comparison children), anxiety symptoms predicted poorer teacher reported social skills, increased parent and teacher reported social problems, lower teacher-reported peer acceptance and greater peer neglect, and more observed solitary play. Positive associations between anxiety and parent-rated social problems as well as peers’ negative sociometric nominations were stronger for children with ADHD relative to comparison children. However, associations between anxiety and observed social hesitance and decreased cooperation were stronger for comparison children than for children with ADHD. Findings suggest that anxiety symptoms may be linked to children’s social relationships, and that the effects of anxiety may differ depending on comorbid ADHD diagnosis. Peer Relationships and ADHD 3 Anxiety in Children with Attention-Deficit/Hyperactivity Disorder: Influences on Peer Relationships and Social Skills The core symptoms of Attention-Deficit /Hyperactivity Disorder (ADHD) center on inattentive and hyperactive/impulsive behaviors (American Psychiatric Association, 1994), but children with ADHD are also well-known to experience substantial peer difficulties (Hoza et al., 2005). Relative to typically-developing youth, children with ADHD are more peer-rejected, have fewer dyadic friendships (Blachman & Hinshaw, 2002; Stormont, 2000), and are rated by parents and teachers as having lower social skills (Hoza et al., 2005). Poor peer relationships warrant concern, as they predict serious adjustment problems in adolescence and adulthood, such as anxiety, depression (Bagwell, Molina, Kashdan, Pelham, & Hoza, 2006), low self-esteem, poor academic achievement, substance abuse (Mikami & Hinshaw, 2006), and school dropout (Hinshaw, 1992). Most children with ADHD have at least one comorbid disorder (Jensen, Martin, and Cantwell, 1997). Yet despite advances in knowledge about the ways in which children with comorbid conduct disorder (CD) and oppositional defiant disorder (ODD) may differ in their peer relationships, relative to those without these comorbidities, far less is known about the influence of anxiety in this regard (Pfiffner, Calzada, & McBurnett, 2000). Approximately 25% to 40% of children with ADHD have a comorbid anxiety disorder, and this prevalence persists after accounting for overlapping symptoms of restlessness and distractibility common to both ADHD and anxiety (Angold et al., 1999; Pliszka, 1999; Tannock). Herein, we review research on the contribution of anxiety symptoms to the peer relationships of children with ADHD, the theoretical rationale for potential interaction effects between anxiety and ADHD, and the hypotheses of the current study. Peer Relationships and ADHD 4 There is reason to believe that anxiety may add to the peer difficulties faced by youth with ADHD. Relative to children with ADHD alone, those who also have a co-occurring anxiety disorder may have more inattention and working memory problems, less impulsivity (Pliszka, 1989; Tannock, 2000), greater academic impairment, stressful life events, parental conflicts, and lower levels of self-esteem (Tannock, Ickowicz, & Schachar, 1995). Each of these domains has been linked to interpersonal problems, suggesting the potential for a differential pattern of peer relationships among children with both ADHD and anxiety. To the extent that ADHD and anxiety each share risk factors associated with peer problems, such as poor emotion regulation (Maegden & Carlson, 2000), self-control deficits (Capps, Sigman, Sena, Hencker, & Whalen, 1996; Hoza et al., 2000), and a hostile attribution bias (Barrett, Rappee, Dadds, & Ryan, 1996), these commonalities may contribute to poor peer status. Indeed, poor behavioral regulation is often negatively associated with prosocial behavior (Eisenberg et al., 1996), and this relationship appears especially strong in children prone to experiencing negative emotions (Diener & Kim, 2004). Thus, regulation problems and negativity common to both ADHD and anxiety may interfere with peer relations. On the other hand, some evidence suggests that comorbid anxiety may inhibit the core behavioral antecedents of peer rejection in children with ADHD – impulsivity and disruptive behavior (Carlson, Lahey, & Neeper, 1984; Pliszka, 1992; Newcorn, 2001). In this sense, comorbid anxiety in children with ADHD may lead to behaviors similar to those demonstrated in youth with anxiety alone, as these children are frequently socially withdrawn and neglected by peers (i.e., receive few liking or disliking peer nominations; Rubin et al., 2003; Strauss et al., 1988). Because neglected status may be favorable to peer rejection in terms of better long-term adjustment and a higher likelihood of changing to accepted status (Parker & Asher, 1987), comorbid anxiety might mitigate social impairment in youth with ADHD. Peer Relationships and ADHD 5 Previous Research on Comorbid ADHD and Anxiety Despite these theoretical rationales, very few studies have empirically assessed the contribution of comorbid anxiety to the peer relationships of children with ADHD. Nevertheless, research suggests that anxiety symptoms (either undiagnosed or diagnosed) in children with ADHD are similar to or more severe than those experienced by children with anxiety alone, and that such anxiety often centers on social concerns (Bowen et al., 2008; Perrin & Last, 1992). For example, in one study of boys with ADHD only, anxiety only, or no psychiatric disorder, childreported worry and oversensitivity did not differ between the ADHD group and the anxiety group (Perrin & Last, 1992). In fact, children with ADHD reported more worries about “having enough friends”, “meeting new kids”, and “meeting new adults” than did children with an actual diagnosis of anxiety, and reported significantly more school-based fears than youth without a psychiatric disorder. Other studies have reported similar levels of anxiety and fearfulness in children with ADHD and those with an anxiety disorder (Jensen, Shevrette, & Xenakis, & Richters, 1993). Such findings are concerning, as they suggest that children with ADHD may have clinically significant levels of worry, particularly surrounding social relationships and the school domain that may go undetected and perhaps exacerbate poor peer relationships. Similar concerns about social incompetence and peer problems have been reported by youth with ADHD and comorbid anxiety disorder, although the associated impairment appears more severe than it is in youth with ADHD or anxiety alone. One recent study of youth ages 817 found higher rates of social phobia and school phobia among youth with ADHD and parentreported anxiety than among those with either ADHD alone, anxiety alone, and nonclinical comparisons (Bowen et al., 2008). Notably, on a social competence measure about perceived popularity and rejection, the comorbid group had significantly lower (poorer) selfand parentPeer Relationships and ADHD 6 reported scores than any of the other groups. This suggests that ADHD and anxiety may have a significant additive impact on social impairment. Although it could be argued that greater peer adversity in youth with comorbid ADHD and anxiety could be attributed to greater severity of ADHD symptoms relative to that in youth with ADHD alone (as in the study above by Bowen and colleagues; 2008), recent evidence suggests the contrary. In fact, secondary analyses in the Multimodal Treatment Study of Children with ADHD (MTA; MTA Cooperative Group, 1999) found that heightened severity in ADHD symptoms in children with comorbid anxiety (parent-reported) disappeared altogether after accounting for dually comorbid CD (March et al., 2000) or ODD/CD (Jensen et al., 2001). Moreover, child self-reported anxiety (without CD) was negatively associated with total social skills (SSRS, parent-reported), but this association was not significant when comorbid CD was considered (March et al., 2000). Thus, isolating the independent effects of comorbid anxiety on the social functioning of children with ADHD may be essential. There is a lack of research on the ways in which peer behaviors typically associated with anxiety may persist in children with comorbid ADHD and anxiety. Children with anxiety are often reported to show withdrawal in new group situations (Rubin & Asendorpf, 1993), and importantly, during childhood such behavior is often associated with peer neglect (i.e., receiving few positive or negative nominations from peers). However, not all anxious children show withdrawal, and for those who do, the type expressed may be variable. For instance, whereas “solitary play” is a form of withdrawn behavior characterized by its goal-oriented nature, social hesitance (or reticence) can be conceptualized by the lack of goal-orientation or productive activity (i.e., simply looking onto the group; Rubin & Asendorpf, 1993). Few studies have examined solitary behavior in youth with ADHD, yet evidence from a sample of girls with and without ADHD suggested that it might increase maladjustment in adolescence Mikami & Peer Relationships and ADHD 7 Hinshaw, 2003). Studies including both sexes have predominantly focused on disruptive or externalizing behaviors (Newcorn et al., 2004; Pliszka, 1989; Abikoff et al., 2004). Results from a seven-year longitudinal investigation of youth with ADHD revealed that anxiety persisted and was associated with more self-reported social problems in adolescence, which increased in the presence of comorbid conduct disorder (Newcorn et al., 2004). In light of these findings, it may be important to examine different social behaviors that accompany social maladjustment in children with ADHD and anxiety. Observational measures have not yet been used to examine the potential contribution of anxiety to the peer interactions of youth with ADHD. Given speculation that anxiety may increase solitary behaviors, but decrease disruptive behaviors, it is important to assess whether this is in fact the case during peer interactions and whether such behaviors can account for differences in adult-rated social skills and peer sociometric status. Finally, the existing research has relied on assessments of children’s social functioning with classroom peers. This presents a limitation because children’s social standing in a familiar group of peers may be more influenced by their previously established reputation than by their actual competence while interacting with peers. As such, observing children’s interactions with novel peers may be a valuable method to capture behavioral variability and social preference. Interactions between Anxiety and ADHD. There is reason to believe that anxiety may have more detrimental effects on the social skills and peer status of youth with ADHD relative to typically developing peers. First, youth without ADHD may have more compensatory strategies to cope with peer rejection than do youth with the disorder. In addition to peer problems, children with ADHD often have conflictual relationships with parents (Johnston & Mash, 2001), siblings (Mikami & Pfiffner, 2008), and non-parental adults (Mikami, Chi, & Hinshaw, 2004). Non-peer relationships may Peer Relationships and ADHD 8 serve as “emotional buffers” (Furman, & Buhrmester, 1985) that can facilitate learning of social skills, and positive non-peer relationships can help children to overcome peer difficulties. Given these factors, children with ADHD may be less equipped to deal with the challenges associated with anxiety. Hypotheses We investigated the role of anxiety symptoms on the social functioning of children with ADHD and typically-developing children in three core domains: (1) parentand teacher-reported social skills and problems, (2) peer preference after a lab-based playgroup and reported by teachers, and (3) social behavior with peers in a lab-based playgroup. We hypothesized that, after statistical control of ADHD and ODD diagnoses, anxiety symptoms would predict poorer social skills and more social problems (parent and teacher report), less acceptance, more disliking, and more neglect reported by peers at playgroups and by teachers (about relations with classmates). Similarly, we predicted that anxiety symptoms would predict more solitary behaviors (solitary play, social hesitance), and fewer prosocial behaviors (cooperative play, prosocial play, peer seeking) at playgroups. Finally, we hypothesized interactions, such that the negative impact of anxiety in all domains would be greatest for children with ADHD. Method Participants One hundred and twenty-one children (84 boys, 37 girls) between 6-10 years of age (M = 8.2; SD = 1.2) participated. Sixty-one children were diagnosed with ADHD (42 boys, 19 girls). Of these, 45 had ADHD Combined type (ADHD-C) and 16 had ADHD Inattentive Type (ADHD-I). The remaining children (n = 60; 41 boys, 19 girls) were ageand gendermatched non-ADHD comparisons. Children with ADHD were recruited from schools, pediatricians, and local clinics. Comparison children were recruited from schools and a database of families who Peer Relationships and ADHD 9 participated in previous research studies at the same university. The ethnic distribution was as follows: 84% Caucasian, 7% Mixed, 5% African American, 3% Asian, and 1% Latino. All parents and children provided written informed consent and assent, and protocols were approved by the university’s Institutional Review Board. Children in the ADHD group surpassed clinical cutoffs for ADHD-C or ADHD-I, as reported by the parent and teacher on the Child Symptom Inventory (CSI; Gadow & Sprafkin). In accordance with DSM-IV field trials (Lahey et al., 1994), a symptom was counted when endorsed by either the parent or the teacher as “often” or “very often” on the CSI. A diagnosis of ADHD was then confirmed via the Schedule for Affective Disorders and Schizophrenia, schoolage version, a semi-structured DSM-IV-TR-based diagnostic parent interview(K-SADS; Kaufman et al., 1997). Children with ADHD-I displayed six or more symptoms of inattention and fewer than six symptoms of hyperactivity/impulsivity; children with ADHD-C had six or more symptoms of inattention and six or more symptoms of hyperactivity/impulsivity. Because of the lack of empirical support for the predominantly Hyperactive/Impulsive subtype beyond preschool age (e.g., Lahey et al., 1994), this subtype was not included. Children in the comparison group could not meet criteria for a diagnosis of ADHD (any subtype), as reported by parent or teacher on the CSI, and the parent confirmed that the child did not meet criteria for ADHD on the K-SADS. Exclusionary criteria were a diagnosis of Pervasive Developmental Disorder or Full Scale IQ below 70. Anxiety, depressive disorders, ODD and CD were not exclusion criteria for either the ADHD or the comparison group because of the high prevalence of these comorbidities in ADHD samples and the desire to have a “normal” but not “super-normal” comparison sample (see Hinshaw, 2002). Based on the K-SADS semi-structured interview given to the parent, 16% (n= 20) of children, all of whom had ADHD, were classified as ODD. No child met criteria for CD. Additionally, 9% (n=11) of children had at least one Peer Relationships and ADHD 10 anxiety disorder (Generalized Anxiety Disorder, Separation Anxiety Disorder, Social Phobia; and Specific Phobia); with the exception of one comparison child, all anxiety diagnoses were among children with ADHD. Due to evidence stating that many children with ADHD who take medications nonetheless remain impaired in their peer relationships (Hoza et al., 2005), the use of medication was not a criterion for exclusion. However, children were required to have been on the same dose of medication for at least three months before the start of the study to avoid the possibility of medication changes interfering with children’s typical behavior. Thirty-six children (54% of ADHD group), but no children in the comparison group, were taking medication to address ADHD symptoms or emotional/behavioral problems. The majority (n= 34; 93%) of medicationusers were taking a stimulant preparation. Procedure Interested parents were administered a ten-minute telephone screening interview during which they completed the CSI about their child’s ADHD symptoms and medication use. Teachers completed the CSI and returned this information by fax. If children passed initial symptom cutoffs (for the ADHD group) or were below clinical cutoffs (for the comparison group) on both parent and teacher rating scales, they then completed an intake session where parents were administered the K-SADS to confirm ADHD diagnosis and establish comorbid disorders, and children were administered the Verbal Comprehension and Perceptual Reasoning subtests from the Wechsler Intelligence Scale – Fourth Edition (WISC-IV; Wechsler, 2003) to provide an estimate of the child’s Full Scale IQ. Parents and children also completed questionnaires about the child’s behaviors and teachers completed this information by mail. Once children completed the intake and were deemed eligible for the study, they were assigned to playgroups with same-age, same-sex peers who were all previously unacquainted. Every Peer Relationships and ADHD 11 playgroup included two children with ADHD and two comparison children (four children in total). However, absenteeism in two playgroups (a child with ADHD on the first occasion, and a non-ADHD comparison child on the second) prevented the complete maintenance of this composition. Each child had one parent present with him/her. A structured activity (group completion of a maze) took place during the first half of the playgroup, but will not be described, as only the latter half of the playgroup (unstructured free play) was examined in this current study. During free play, children were presented with an assortment of board games and toys and were told they could choose any activity and play with whomever they wished. Parents were given the global instruction to help their child make friends. At the end of the playgroup, each child was individually interviewed using a standard sociometric procedure (Coie, Dodge, & Coppotelli, 1982). To aid recall, the interviewer presented the child with pictures of the other three children present. Children nominated the peers with whom they most liked to play (positive nominations) and those with whom they least liked to play (negative nominations). Unlimited nominations were accepted, such that children were allowed to nominate none, one, two, or all three of the other peers for each category. Children were asked not to disclose their nominations to the other children. We note that, after intake and playgroup procedures were completed, parents were randomly assigned to receive a novel intervention designed to provide them with strategies to improve their child’s peer relationships. Parents were present during all of the observations of children’s playgroup interactions because parental behaviors were viewed as central to the intervention. The results of this intervention are unpublished at this time, and the current manuscript is the first based on this sample. All measures used for the current manuscript focus on baseline assessments before parents were notified of their inclusion in the intervention. Peer Relationships and ADHD 12 Measures Child anxiety. Children’s anxiety symptoms were measured using the anxiety narrowband scale on the Teacher Report Form (TRF; Achenbach, 1991b). This 13-item measure assesses anxiety on a three-point scale (0= “not true”, 1= “somewhat/sometimes true”, and 2 = “very true/often true”). It includes questions such as “too fearful or anxious”, “worries”, and “fears going to school”. This scale has been found to discriminate between children with and without an anxiety disorder and is correlated with both childand parent-reported measures of anxiety (Kendall et al., 2007). In the current sample, the TRF was associated with the internalizing problems subscale on Social Skills Rating System (SSRS; Gresham & Elliot, 1990) reported by teachers (r = .62, p < .05) and parents (r= .46, p < .05). Although the SSRS internalizing scale contains items related to depression and anxiety, this high correlation was likely due to associations between anxiety items on the scales, as the TRF anxiety scale was not correlated with child-reported depression (r = .13, p > .10), supporting its discriminant validity in this sample. Each child’s raw score on the scale was converted into a t-score based on age and sex norms. Using parent-report on the K-SADS, children were also classified as having an anxiety disorder if they met diagnostic criteria for any of the following disorders: Separation Anxiety Disorder, Social Phobia, Agoraphobia, and Overanxious/Generalized Anxiety Disorder. The KSADS was administered and scored by students in clinical psychology who had received special training in this procedure. A licensed clinical psychologist oversaw all the interviews and agreement was 100% with the ascribed diagnoses. Adult-informant measures of social functioning. Parents completed the social problems subscale on the Child Behavioral Checklist (CBCL; Achenbach, 1991a) and teachers completed the parallel version (TRF). These scales include items measuring children’s emotional and Peer Relationships and ADHD 13 behavioral problems (rated from 0-2; 0 = not true, 2 = very true), have excellent test-retest reliability and validity, and are widely used (Vignoe, Berbue, & Achenbach, 1999). Each child’s raw score on the scale was converted into a t-score based on age and sex norms. Parents and teachers also reported children’s social skills on their respective versions of the Social Skills Rating System (SSRS; Gresham and Elliot, 1990). The SSRS assesses three domains (Cooperation, Assertion, and Self-control), and the parent version has a Responsibility subscale. Items are rated on a 3-point scale (1= “never”, 2 = “sometimes”, and 3 = “very often”). The SSRS has good criterion-related validity (Gresham & Elliot, 1990), and sufficient internal consistency (Pederson, Worrell, & French, 2001). Each child’s raw score on the total social skills scale was converted into a t-score based on age and sex norms. Peer sociometric status. Using the child sociometric interview procedure after the playgroup (as described above), we calculated each child’s proportion of positive nominations and negative nominations received by peers by dividing the number of nominations a child received by the number of playmates providing nominations. Teachers also reported on children’s sociometric status in their regular classrooms using the Dishion Social Acceptance Scale (Dishion, 1990). This measure has been used in the ADHD literature (Hinshaw, 2002), and Dishion (1990) reports moderate correlations with peer-reported sociometric measures. Teachers reported the percentage of peers who “like and accept”, “dislike and reject”, and “ignore or are neutral” towards the child being assessed. Teachers could choose one of five options: Less than 25% (“Almost none”), 25-50% (“A few”), 50% (“About half”), 50-70% (“Most”), to over 75% (“Nearly all”). Observed social behaviors in playgroups. All playgroup interactions were videotaped. Trained coders, blind to the children’s diagnostic status, used a standardized behavior coding system to assess children’s interactions. A random sample of 25% of videotapes were selected to Peer Relationships and ADHD 14 be double-coded, and inter-rater reliability estimates for the behavioral categories were calculated using the Intraclass Correlation Coefficient (ICC). Five operationally defined behaviors were coded during unstructured free play. Social hesitance, solitary play, and cooperative play were mutually exclusive and were coded on a 1-10 metric based on the coders’ estimate of the proportion of time each child spent doing each activity (1=less than 10% of the time; 10 = 91-100% of the time). Social hesitance was defined as being alone and not engaging in a constructive activity; children in this category could have been wandering aimlessly, looking at a child or group of children (without joining in) or “staring off into space” (ICC = .91). Solitary play was defined as being alone yet engaging in a constructive, goal-directed activity; children in this category could have been drawing alone, playing Legos, or playing another game alone (ICC = .73). Cooperative play referred to interacting with at least one other child; the quality of the interaction was not accounted for, just the presence of a peer interaction (ICC =.80). Prosocial behavior was defined as cooperative, praising, complementing, assisting, or other helpful behaviors, and was scored on a 0-3 metric (0 = no evidence of behavior; 1 = one minor incidence of behavior, but behavior is not a pervasive theme; 2 = more than one minor incident of behavior such that it was a pervasive theme, or one major incident; 3 = more than one severe incident or only one major incident but at least one minor incident. Coders’ reliability estimate for prosocial behavior was acceptable (ICC=.72). Lastly, peer seeking reflected the number of times a child was alone and then attempted to play with another child or enter another group (ICC=.61). Data Analytic Plan Hierarchical multiple regressions were performed to test our hypotheses that anxiety symptoms would be concurrently associated with poorer social functioning beyond the effects of ADHD and ODD. Measures of social skills, peer sociometric status, and observed behaviors Peer Relationships and ADHD 15 were entered as the outcome variables. To isolate the contribution of anxiety, we controlled for ADHD diagnosis (dichotomous, dummy coded) on Step 1. Because most children with ADHD also had ODD, which may affect social functioning, we controlled for ODD diagnosis (dichotomous, dummy coded) in Step 2. We introduced teacher-reported anxiety symptoms in step 3 and the interaction between ADHD and anxiety in Step 4. Significant interactions were probed in the manner recommended by Holmbeck (2002). Examination of the histogram for each dependent variable tested potential violations to assumptions of normality. The histograms for prosocial and cooperative play were skewed to the left, such that many participants often showed these behaviors. The histogram for social hesitance was skewed to the right such that few children displayed this behavior. To reduce the skews, square root transformations were performed. The results were significant before and after the transformations, yet our results represent analyses using the transformed values. If children’s anxiety symptoms were negatively associated with social skills (adultinformant rated), peer liking/acceptance, observed prosocial behavior, cooperative play, and peer seeking, then our first hypothesis would be confirmed. Similarly, our first hypothesis would also be confirmed if children’s anxiety symptoms were positively associated with social problems (adult-informant rated), and peer neglect by peers (peer and teacher report), as well as observed solitary play and social hesitance. If the interactions between ADHD and anxiety symptoms were significant, and post-hoc probing revealed that the correlations between anxiety and poor social functioning were stronger for the children with ADHD relative to the comparison children, then our second hypothesis would be confirmed. We repeated all analyses substituting anxiety disorder diagnosis (dichotomous, from the parent-reported K-SADS) in place of the continuous symptom measure at step 3. However, the dichotomous measure of anxiety diagnosis was not used as a primary measure of anxiety in this Peer Relationships and ADHD 16 study for several reasons. First, parent-reported anxiety explains a low proportion of variance in child-reported anxiety (1-2%; four times lower than that explained by teacher-report), and has failed to discriminate between children who report high versus low levels of anxiety (March et al., 2000; Mesman & Koot, 2000). It has also been suggested that parents may confound their report of externalizing symptoms with report of anxiety (March et al., 2000). Further, few participants met criteria for an anxiety disorder, particularly among the comparison sample. Using the continuous symptom measure of anxiety provided the most sensitive test of our central hypotheses involving potential interaction effects between anxiety and ADHD status. Results Participant demographics. As shown in Table 1, children with ADHD did not differ from children without ADHD with respect to age, sex, or ethnicity. However, comparison children had significantly higher Full Scale IQ scores than did children with ADHD. Although the overall sample mean of teacher-reported anxiety was low, anxiety scores were significantly higher in children with ADHD than they were in comparison children. These levels are similar to those reported for children with ADHD in previous investigations (March et al., 2000). No significant differences between the ADHD-I (n = 16) and ADHD-C (n = 45) subtypes emerged on the demographic variables or on levels of teacher-reported anxiety symptoms, which we note is consistent with the findings of Power et al. (2004). All children taking medication did not differ from the other non-medication users on any of the dependent measures. Adult-informant measures of social functioning. Results of analyses predicting adultreported social skills on the SSRS are displayed in Table 2. After statistical control of Adhd in Step 1 and ODD in Step 2, there was a significant negative association between teacher-reported continuous anxiety symptoms and teacher-reported social skills in Step 3. However, for parentreported social skills, the negative association between anxiety and social skills was only Peer Relationships and ADHD 17 marginally significant (p = .07). The interactions between ADHD and anxiety were not significant for social skills rated by parents or teachers. Analyses predicting children’s social problems reported by the parent on the CBCL and the teacher on the TRF are presented in Table 3. After statistical control of Adhd in Step 1 and ODD diagnostic status in Step 2, anxiety symptoms were significantly associated with more social problems on both parent and teacher report in Step 3. There was also a significant interaction between ADHD and anxiety in Step 4. Post-hoc probing in the manner suggested by Holmbeck (2002) revealed that the positive relationship between anxiety symptoms and parentreported social problems was stronger for children with ADHD (_ = .35; p < .001) than it was for comparison children (_ = .00; p > .10). The interaction between ADHD and anxiety was not significant in Step 4 for teacher-reported social problems. Peer Sociometric Status. Regarding peer status assessed sociometrically during playgroups, as shown in Table 4, ADHD was marginally associated with positive nominations with a negative beta weight in Step 1, but in Step 2 ODD was significantly associated with a higher number of positive nominations. Contrary to hypotheses, the contribution of anxiety in Step 3, and the interaction term in Step 4, were not significant. For negative nominations, ADHD was not significant in Step 1, but in Step 2 ODD was associated with fewer negative nominations, and anxiety symptoms were not significantly associated with negative nominations in Step 3. However, the interaction between ADHD and anxiety was significant. Probing revealed associations in the hypothesized directions, although they were marginal and not significant; elevated anxiety tended to be positively associated with negative nominations (_ = .19; p = .10) , whereas for the comparison group, anxiety tended to be negatively associated with negative nominations (_ = -.16; p > .10). Lastly, contrary to hypotheses, social impact was not associated with ADHD, anxiety, or the interaction. Peer Relationships and ADHD 18 Results for the criterion of classroom peer status (teacher-reported) are shown in Table 5. ADHD diagnosis was significantly associated with poorer peer status in all three domains (i.e., decreased peer acceptance, increased peer rejection, and increased peer neglect). Of particular importance, anxiety symptoms were significantly associated with lower peer acceptance and higher neglect, but marginally associated with more negative nominations. For all three variables of classroom peer status, the interaction between ADHD and anxiety was not significant. Observed Social Behaviors in Playgroups. Table 6 presents results for the observed social behaviors of solitary play and social hesitance. After statistical control of ADHD in Step 1 and ODD in Step 2, anxiety symptoms in Step 3 were associated with a higher frequency of solitary play. The interaction between anxiety and ADHD was no significant for solitary play. However, for social hesitance, the main effect of anxiety was not significant in Step 3 after statistical control of ADHD and ODD, yet the interaction in Step 4 was significant. Probing revealed non-significant associations for both groups, but anxiety to be positively associated with social hesitance among the comparison group (_ = .15; p >.10), but not among children with ADHD (_ = -.05 p >.10). In the final model, all main effects and the interaction term were significantly associated with social hesitance. Contrary to expectations, cooperative play was not associated with ADHD in Step 1, ODD in Step 2, or anxiety in Step 3. The hypothesized interaction was significant in Step 4; anxiety was associated with significantly lower levels of cooperative play in children without ADHD (_ = -.44; p < .05), but the association was not significant for children with ADHD (_ = .00; p > .10). For peer seeking, there was a significant positive relationship between ADHD and anxiety in Step 1, ODD was associated with lower levels of this behavior in Step 2, and anxiety in Step 3 was not significantly associated with total peer seeking. The interaction between Peer Relationships and ADHD 19 ADHD and anxiety was significant in Step 4. Among comparison children, elevated anxiety was associated with decreased peer seeking (_ = -.17; p >.10), but the association was weaker (and negative) for children with ADHD (_ = .07 p > .10). All analyses were repeated using children’s categorical anxiety diagnosis from the KSADS semi-structured parent interview. K-SADS anxiety was not correlated with teacherreported continuous anxiety symptoms on the TRF (r = .07, p > .10). Because only one of the 11 children with an anxiety disorder was a comparison child, it was not possible to test the interaction effects between K-SADS anxiety and ADHD status. For both parent and teacherreported social skills on the SSRS, after controlling for ADHD and ODD, K-SADS anxiety disorder was not significant. However, K-SADS anxiety was associated with more social problems reported by parents (_ = .24; p < .005) but not by teachers. Regarding playgroup sociometric nominations, K-SADS anxiety was positively associated with the number of positive nominations received (_ = .24; p < .05), and a higher social impact score (_ = .24; p < .05), but was not associated with negative nominations (_ = .01; p >.10). For teacher-reported classroom social status, anxiety was not associated with acceptance but was positively associated with neglect (_ = .28; p < .005) and rejection (_ = .32; p < .005). For observed behavior, after controlling for ADHD and ODD, K-SADS anxiety disorder was not found to be associated with solitary play, social hesitance, cooperative play, prosocial play, or peer seeking. In sum, results using parent-reported anxiety disorder on a diagnostic interview were largely consistent with those using teacher-reported continuous anxiety symptoms in suggesting that comorbid anxiety was associated with adult-informant reported social impairment and teacher-reported peer neglect. However, in contrast to results using teacher-reported anxiety, KSADS anxiety was positively associated with social impact but not with the observed behaviors in the playgroup. Peer Relationships and ADHD 20 Discussion We examined the associations between anxiety symptoms, social skills, peer status, and observed social behaviors among children with ADHD and comparison children. After statistical control of ADHD and ODD diagnoses, anxiety symptoms were significantly associated with more social problems as reported by parents and teachers, and poorer social skills according to teachers. Regarding main effects for peer status assessed via peer nominations in playgroups, anxiety was not significantly associated with peer liking, disliking, or neglect after accounting for ADHD and ODD diagnostic status. Anxiety symptoms were also negatively related to peer acceptance and positively related to neglect by classmates, based on teacher report. Solitary play was positively associated with anxiety symptoms for all children, but social hesitance, cooperative play, prosocial behavior, and peer seeking, were individually unrelated to anxiety. Importantly, several interactions emerged between anxiety symptoms and ADHD diagnostic status. Regarding parent-reported social problems, the positive association between anxiety and social problems was stronger for children with ADHD relative to comparison children. Nominations of whom children "least liked to play with" at playgroups were associated with higher levels of anxiety, but only among children with ADHD. The interaction between ADHD and anxiety for playgroup observations of prosocial behavior was not significant. Yet there was a significant interaction for social hesitance, and the direction of the beta weights (although both non-significant) suggested that anxiety tended to be positively associated with social hesitance in comparison children but not in children with ADHD. Higher cooperative play was significantly associated with increased anxiety in comparison children, but was not significant in children with ADHD. Finally, anxiety symptoms were differentially related to total peer seeking attempts; it appeared that the tendency for comparison to not seek peers was stronger than the tendency for children with ADHD to seek peers. Peer Relationships and ADHD 21 That teacher-reported anxiety symptoms were positively associated with poorer parent reported social skills and more social problems (reported by both parents and teachers) beyond the effects of ADHD and ODD is generally consistent with previous studies, but provides several important extensions. For instance, in keeping with the findings of this study, secondary MTA analyses revealed that children with ADHD and comorbid anxiety (based on parent interview) without ODD/CD had significantly poorer social skills than did those with ADHD only (Jensen et al., 2001). However, prior to this study, such findings had not been extended to include teacher reports of anxiety or social skills. Furthermore, our results suggest that anxiety may add to the social adversity of youth with ADHD and that this effect is not simply a product of parental attributions of negative affectivity when reporting their child’s anxiety (see March et al., 2000). Indeed, higher levels of anxiety among all children in this study were associated with more severe social problems on both parent (CBCL) and teacher (TRF) reports, which to our knowledge is an unprecedented finding. Anxiety may contribute to social problems through many routes, yet it is likely that its relationship with peer difficulties is bidirectional. Such a relationship may be particularly salient in anxious children with inappropriate knowledge about or performance of social skills, as social ineffectiveness may arouse expectations of future failures and negative peer evaluations (Muris, 2007). Such expectations may worsen anxiety and avoidance (Rubin et al., 2003), and further limit opportunities to learn social skills. It is also possible that anxiety symptoms may hinder poor social relations by interfering with adaptive coping strategies. Indeed, anxious children often have trouble managing worries, sadness, and anger; they experience emotions with “high intensity and have little confidence in their ability to regulate their arousal” (Suveg & Zeman, 2000). Such characteristics may limit the use of more effective alternative strategies (e.g., cooperation, self-control, and assertion) to manage conflicted peer interactions, should they arise. Peer Relationships and ADHD 22 Regarding the interaction between ADHD and anxiety for social problems reported by parents (on the CBCL), it appears that anxiety may exacerbate peer problems in children with ADHD to a greater degree than in comparison children. Interestingly, this runs counter to theories claiming that anxiety may inhibit disruptive/impulsive behavior in ADHD, possibly ameliorating peer problems (e.g., Pliszka, 1992). Our results could more likely suggest that children with ADHD possess fewer psychosocial buffers to handle anxiety-related impairment. Indeed, relative to typically developing children, these children may be less involved in extracurricular activities (e.g., less school and sport-related involvement; Lahey, 1994), have a weaker support system (e.g., poorer parent and non-peer relations), and higher levels of aggression may exacerbate the poor regulation that often comes with anxiety (Tannock, 2000). With the exception of the interaction between anxiety and ADHD for negative peer nominations at playgroups, all hypothesized effects for playgroup peer nominations were not significant. However, our prediction that anxiety symptoms would be associated with more peer neglect and less peer acceptance held more strongly for teacher reports of peer status on the Dishion Social Acceptance scale. That is, peers in children’s regular classroom were reported to more often ignore and less frequently accept children with anxiety. Why might there have been a discrepancy across informants of children's peer status? One possible explanation could concern teacher reporter bias on the Dishion Acceptance Scale. More specifically, considering that anxiety symptoms were based on teacher report, it is possible that the association between anxiety and classmate peer neglect and low acceptance, could have been an artifact of teacher bias. Teachers may have used their perceptions of children’s anxiety to guide their reports of peer status, or vice versa. Another possibility is that the results for peerand teacher-reported peer status may shed light on some the processes through which anxious youth elicit and respond to reactions from peers in familiar and unfamiliar contexts over time. Initially, peers may be less Peer Relationships and ADHD 23 responsive to anxious peers, but over time, these anxious children are recognized as being neglected and unpopular, which may lead to low acceptance. Yet this proposed trend could possibly depend on the presence or absence of ADHD. That is, children’s initial reactions to anxious peers with ADHD may be more negative instead of neutral, as anxiety in children with ADHD may be manifested in the form of more disruptive or aggressive behaviors associated with rejection (Coie & Kupersmidt, 1982; Tannock, 2000). While high levels of social hesitance and low levels of cooperative play, found among anxious comparison peers relative to those with ADHD in this study, may be considered less “social”, these behaviors may be less salient predictors of peer status in a new group of peers. Indeed, younger children pay more attention to peers’ aggression and negative behaviors than their withdrawal, and may not consider peers’ solitary behavior as merits for rejection (Ladd & Mars, 1986). Instead, reliable behaviors associated with rejection include disruptive attempts to enter a new group of peers, aggression, noisy and bothersome behaviors, and “quick-tempered” behavior, (Guevrement & Dumas, 1994; Coie & Kupersmidt, 1983). Perhaps such behaviors (unaccounted for in this study) contributed to the link between anxiety and negative nominations that were positively associated with anxiety in the ADHD group but not in the comparison group. Several explanations could be given as to why anxiety symptoms were associated with fewer observations of peer seeking in comparison children, yet more of this behavior in youth with ADHD, and may also explain why anxiety was associated with more social problems for children with ADHD. Specifically, evidence suggests that, relative to comparison peers, children with ADHD a) face greater peer rejection and exclusion (Hoza et al., 2005), and b) are more impulsive/hyperactive (Nigg, 2001). These characteristics may feed the impairment conferred by anxiety. Regarding peer seeking, it may be that anxious youth without ADHD may follow the more normative cycle of fearful behavior when faced with the task of approach a peer group – Peer Relationships and ADHD 24 they may become shy and retreat into solitude on the periphery of the group. Yet those with ADHD may deviate from this process, as they generally have a poorer ability to regulate their arousal and emotions, especially in the pursuit of a goal, than do children without ADHD (Barkley, 2006; Hinshaw & Melnick, 2000). Thus, the tendency for children with ADHD to “approach” impulsively might override the typical tendency to “avoid”, and as a result, they may act inappropriately initiate peer interactions with little forethought. Moreover, children with comorbid ADHD and anxiety may represent a subset of children with particularly poor social skills, and as such, their peer overtures may be largely unsuccessful. These encounters may exacerbate problems, as rejection or exclusion from peers may reinforce the anxiety. This study has several strengths. Our use of a variety of methods (rating scales, behavior observations) and informants (peers, parents, teachers) adds to knowledge about the social skills and peer group status among youth with and without ADHD and the ways in which they can be influenced by anxiety. Peer ratings may be especially important for identifying children at risk for later social maladjustment (Cowen et al., 1973; Parker & Asher, 1987). This study is also strengthened by our inclusion of girls and boys, children with ADHD-I and ADHD-C, and children with comorbid disorders. Participating youth were not only clinic-referred, but were also drawn from schools and were respondents to community advertisements, thus adding to the heterogeneity of the sample. Furthermore, because children with ADHD were matched with comparison children on sex, gender, and grade, we were able to more accurately compare behavioral and social characteristics that might otherwise have differed if the matched characteristics were free to vary. Limitations of this study include our primary measurement of anxiety, ascertained by teachers’ reports. Children may more accurately report their anxiety symptoms than parents (Bird et al., 1992), and teachers may over-report children’s anxiety (Furr, Tiwari, Suveg, & Peer Relationships and ADHD 25 Kendall, 2009). Furthermore, it is also important to note that peer status, as measured in this study, represents only one dimension of children’s peer relationships, and that dyadic friendships were not considered. Few studies have focused on the friendships of children with ADHD (for exceptions, see Blachman & Hinshaw, 2002; Hoza et al., 2003). However, our results, suggesting heightened social problems in the presence of elevated anxiety among children with and without ADHD, should provide a clear impetus for future research investigating these trends in the friendship domain. Playgroup results should also be interpreted in light of the group context in which they were measured. Half of the children nominating their playmates had ADHD, and this proportion far exceeds that in naturalistic contexts (i.e., 3% to 7% of school-age children have a diagnosis of ADHD; American Psychiatric Association, 2000). Because children with ADHD may be more tolerant of other peers’ behavior, our results may underestimate the peer adversity seen in this population (Hinshaw & Melnick, 1995). Furthermore, a more comprehensive analysis of observed behaviors would have been at the inclusion of negative behaviors (as opposed to prosocial and solitary behaviors only). Our results may thus limit the range of behaviors influenced by anxiety, especially given evidence that heightened aggression may accompany comorbid anxiety in children with ADHD (Tannock, 2000). Finally, it could be suggested that we used statistical over-control in our regression analyses by covarying ODD when testing the relation between anxiety and an outcome variable. Some evidence suggests that negative affectivity (NA) is tightly coupled with ODD, and that it may also be an important mode of linkage between ADHD and anxiety (Baldwin & Dadds, 2007). Thus, controlling for ODD could possibly have subtracted away from important covariance between ODD and anxiety. Evidence does however suggest that, regardless of shared negative affectivity between ODD and anxiety, comorbid subgroups defined by these disorders have distinct clinical profiles and treatment outcomes (Jensen et al., 2001). Peer Relationships and ADHD 26 The results of this investigation have important implications for future attempts to identify youth with poor peer status who are at risk for social maladjustment. Crucially, our results demonstrate the importance of incorporating the perspectives of multiple informants and utilizing a variety of methodologies when assessing children’s social skills and peer status, especially among youth with ADHD. Subsequent research will be needed to further identify the relative risks and potential benefits associated with anxiety symptoms among children with and without ADHD. Peer Relationships and ADHD27 ReferencesAchenbach, T.M. (1991a). Manual for the Child Behavior Checklist Form and Profile. Burlington, VT:University of Vermont, Department of Psychiatry.Achenbach, T. M. (1991b). Manual of the Teacher’s Report Form and Profile. Burlington, VT:University of Vermont, Department of Psychiatry.Asher, S.R. (1983). Social competence and peer status: Recent advances and future directions. ChildDevelopment, 54, 1427-1434.American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, FourthEdition, Text Revision. Washington, DC, American Psychiatric Association, 2000.Angold, A., Costello, E.J., & Erkanli, A. (1999). Comorbidity. Journal of Child Psychology andPsychiatry, 40, 57-87.Bagwell, C.L., Molina, B.S., Kashdan, T.B., Pelham, W.E., & Hoza, B. (2006). Anxiety and mooddisorders in adolescents with childhood attention deficit/hyperactivity disorder. Journal ofEmotional and Behavioral Disorders, 14, 178-187.Baldwin, J.S., & Dadds, M.R. (2008). Examining alternative explanations of ADHD and anxietysymptoms in children: A community study. Journal of Abnormal Child Psychology, 36, 67-79.Barett, P.M., Rappee, R.M., Dadds, M.M., & Ryan, S.M. (1996). Family enhancement of cognitivestyle in anxious and aggressive children. Journal of the American Academy of Child andAdolescent Psychiatry, 24, 187-203.Barkley, R.A. (2006). Attention deficit hyperactivity disorder: A handbook for diagnosis and treatment.Third edition. New York: Guilford.Berry, C.A. Shaywitz, S.E., Shaywitz, B.A. (1995). Girls with attention-deficit disorder: A silentmajority? A report on behavioral and cognitive characteristics. Pediatrics, 76, 801-809. Peer Relationships and ADHD28 Capps, L., Sigman, M,, Sena, R., Henker, B., & Whalen, C. (1996). Fear, anxiety, and perceivedcontrol in children of agoraphobic parents. Journal of Child Psychology, Psychiatry, and AlliedDisciplines, 37, 445-452.Coie, J.D., Dodge, K. A., & Coppotelli, H. (1982). Dimensions and types of social status: A cross-ageperspective. Developmental Psychology, 18, 557-570.Biederman, J., Newcorn, J., & Sprich, S. (1991). Comorbidity of attention deficit hyperactivitydisorder with conduct, depressive, anxiety, and other disorders. American Journal of Psychiatry,148, 564-577.Bishop, J.A., & Interbizen, H.M. (1995). Investigation of their relation to self-esteem. Journal ofEarly Adolescence, 4, 476-489.Bowen, R.C., Chavira, D.A., Bailey, K., Stein, M.T., Stein, M.B. (2008). Nature of anxiety comorbidwith attention deficit hyperactivity disorder in children from a pediatric primary care setting.Psychiatric Research, 157, 201-9.Boyle, M.H., Offord, D.R., Racine, Y., Szatmari, P., Fleming, J.E., & Sanford, M. (1996). Identifyingthresholds for classifying childhood psychiatric disorder: Issues and prospects. Journal ofAmerican Academy of Child and Adolescent Psychiatry, 35, 14401448.Carlson, C.L., Lahey, B.B., & Neeper, R. (1984). Peer assessment of the social behavior of accepted,rejected, and neglected children. Journal of Abnormal Psychology, 12, 189-198.Diener, M., & Kim, D.Y. (2004). Maternal and Child Predictors of Children’s Social Competence.Journal of Applied Developmental Psychology, 25, 3-24.Dishion, T.J. (1990). The peer context of troublesome child and adolescent behavior. In P.E. Leone(Ed.), Understanding troubled and troubling youth: Multiple perspectives. Thousand Oaks, CA:Sage. Peer Relationships and ADHD29 Dunn, J., & Bruner, J.S. (2004). Intrigues, Bullying, Rejection and Loneliness: Friendships,Depression, and Anxiety. In J. Dunn & J.S. Bruner (Eds.), Children's Friendships (pp. 96-97).Malden, MA: Blackwell.Dodge, K.A., Coie, J.D., & Brakke, N.P. (1982). Behavior patterns of socially rejected and neglectedpreadolescents: The roles of social approach and aggression. Journal of Abnormal ChildPsychology, 10, 389-410.Eisenberg, N., Fabes, R.A., Guthrie, I.K., Murphy, B.C., Maszk, P., Smith, M., et al. (1996). Therelations of regulation and emotionality to problem behavior in elementary school children.Development and Psychopathology, 8, 141-162.Eisenberg, N., Fabes, R., Nyman, M., Bernwieg, J., & Pinuelas, A. (1994). The relations ofemotionality and regulation to young children’s anger-related reactions. Child Development, 65,109-128.Erhardt, D., & Hinshaw, S.P. (2004). Initial sociometric impressions of attention-deficit hyperactivitydisorder and comparison boys: Predictors from social behaviors and from nonverbal variables.Journal of Consulting and Clinical Psychology, 62, 833-842.Furman, W., & Bierman, K.L. (1984). Children’s perceptions of the qualities of sibling relationships.Child Development, 56, 448-461.Furman, W., & Buhrmester, D. (1985). Children's perceptions of the personal relationships in theirsocial networks. Developmental Psychology, 21, 1016-1024.Furr, J.M., Tiwari, S., Suveg, C., & Kendall, P.C. (2009). Anxiety disorders in children andadolescents. In M.M Anthony & M.B. Stein (Eds.), Oxford Handbook of Anxiety and RelatedDisorders (pp.636-656). New York: Oxford.Gazelle, H., & Ladd, G.W. (2003). Anxious solitude and peer exclusion: A diathesis-stress model ofinternalizing trajectories in childhood. Child Development, 74, 257-258. Peer Relationships and ADHD30 Gazelle, H. (2008). Behavioral profiles of anxiety solitary children and heterogeneity in peerrelationships. Developmental Psychology, 44, 1604-1649.Gershon, J. (2002). A meta-analytic review of gender differences in ADHD. Journal of AttentionDisorders, 5, 143-154.Interbitzen-Nolan, & Walters, K.S. (2000). Social Anxiety Scale for adolescents: normative data andfurther evidence of construct validity, Journal of Clinical Child Psychology, 29, 360–371.Jensen, P.S., Martin, D., & Cantwell, D.P. (1997). Comorbidity in ADHD: Implications for Research,Practice, and DSM-V. Journal of the American Academy of Child and Adolescent Psychiatry, 36,1065-1078.Karustis, J.L., Power, T.J., Rescorla, L.A., Eiraldi, R.B., Gallagher, P.R. (2000). Anxiety anddepression in children with ADHD: Unique associations with academic and social functioning,Journal of Attention Disorders, 4, 133-149.Hartman, C.A., Rhee, S.H., Willicutt, W., Pennington, B.F. Modeling rater disagreement for ADHD:Are parents or teachers biased? Journal of Abnormal Child Psychology, 35, 536-542.Hodgens, J. B., Cole, J., & Boldizar, J. (2000). Peer-based differences among boys with ADHD.Journal of Child Clinical Psychology, 29, 443-452.Holmbeck, G.N. (2002). Post-hoc probing of significant moderational and meditational effects instudies of pediatric populations. Journal of Pediatric Psychology, 27, 87-96.Hinshaw, S.P. (1992). Externalizing behavior problems and academic underachievement in childhoodand adolescence: Causal relationships and underlying mechanisms. Psychological Bulletin, 111,127-155.Hoza, B., Mrug, S., Gerdes, A.C., Hinshaw, S.P., & Bukowski, W.M. (2005). What aspects of peerrelationships are impaired in children with attention-deficit/hyperactivity disorder? Journal ofConsulting and Clinical Psychology, 73, 411-423. Peer Relationships and ADHD31 Hoza, B., Mrug, S., Pelham, W.E., et al. (2003). A friendship intervention for children with attention-deficit/hyperactivity disorder: preliminary findings. Journal of Attention Disorders, 6, 87-98.Hoza, B., Waschbush, D.A., Pelham., W.E., Molina, B.S.G., & Milich, R. (2000). Attention-deficit/hyperactivity disordered and control boys’ responses to social success and failure. ChildDevelopment, 71, 432-446.Hymel, S., Wagner, E., & Butler, L. (1990). Reputation bias: View from the peer group. In S.R.Asher & J.D. Coie (Eds.), Peer rejection in childhood (pp. 156-186). Cambridge: CambridgeUniversity Press.Jensen, P.S., Shervette, R.E., Xenakis, S.N., Richter, J. (1993). Anxiety and depressive disorders inattention deficit disorder with hyperactivity: New findings. American Journal of Psychiatry,150, 1203-1209.Ladd, G. W. & Burgess, K. B. (1999). Charting the relationship trajectories of aggressive, withdrawn,and aggressive/withdrawn children during early grade school. Child Development, 70, 910-929.Lahey, B.B., Applegate, B., McBurnett, K., Biederman, J., Greenhill, L., Hynd, G.W., et al. (1994).DSM-IV field trials for attention deficit hyperactivity disorder in children and adolescents.American Journal of Psychiatry, 151, 1673-1685.Lahey, B.B., Pelham, W.E., Schaughency, E.A., Atkins, M.S., Murphy, A., Hynd, G., et al. (1988).Dimensions and types of attention deficit disorder. Journal of American Academy of Child andAdolescent Psychiatry, 27, 330-335.Landau, S., Milich, R., & Diener, M. B. (1998). Peer relations of children with attention-deficithyperactivity disorder. Reading and Writing Quarterly, 14, 83-105.March, J.S. (2002). Attention-deficit hyperactivity disorder. Program in Child and Adolescent AnxietyDisorders, 1-5. Peer Relationships and ADHD32 March, J. S., Swanson, J. M., Arnold, L. E., Hoza, B., & Conners, K.C., et al. (2000). Anxiety as apredictor and outcome variable in the multimodal treatment study of children with ADHD.Journal of Abnormal Child Psychology, 28, 527-541.Mikami, A.Y, & Hinshaw, S.P. (2003). Buffers of peer rejection among girls with and without ADHD:The role of popularity with adults and goal-directed solitary play. Journal of Abnormal ChildPsychology, 31, 381-397.Mikami, A.Y., Chi, T.C., & Hinshaw, S.P. (2004). Behavior ratings and observations of externalizingsymptoms in girls: The role of child popularity with adults. Journal of Attention Disorders, 11,1-11.Mikami, A.Y., & Hinshaw, S.P. (2006). Resilient adolescent adjustment among girls: Buffers of peerrejection and attention-deficit/hyperactivity disorder. Journal of Abnormal Child Psychology, 34,825-839.Mrug, S., Hoza, B., Pelham, W.E., Gnagy, E.M., & Greiner, A.R. (2007). Behavior and peer status inchildren with ADHD. Journal of Attention Disorders, 10, 359-371.Mrug, S.B., Hoza, B., Gerdes, A.C., Hinshaw, S.P., Arnold, E.A., et al. (2009). DiscriminatingBetween Children with ADHD and Classmates Using Peer Variables. Journal of AttentionDisorders, 12, 372-380.Muris, Peter. (2007). Normal and Abnormal Fear and Anxiety in Children and Adolescents:Environmental Influences. Elsevier Inc: Burlington, MA.Nangle, D. W., Erdley, C. A., & Gold, J. A. (1996). A reflection on the popularity construct: Theimportance of who likes or dislikes a child. Behavior Therapy, 27, 337-352.Newcorn, J.H., Halperin, J.M., Arnold, L.E., Abikoff, H.B., Cantwell, D.P., et al. (2001). Symptomprofiles in children with ADHD: Effects of comorbidity and gender. Journal of the AmericanAcademy of Child and Adolescent Psychiatry, 40, 137-146. Peer Relationships and ADHD33 Nigg, J.T. (2001). Is ADHD an inhibitory disorder? Psychological Bulletin, 125, 571-596.Perrin, S., & Last, C.G. (1997). Worrisome thoughts in children clinically referred for anxiety disorder.Journal of Clinical Child Psychology, 26, 181-189l.Parker, J.G., & Asher, S.R. (1993). Friendship and friendship quality in middle childhood: Links withpeer group acceptance and feelings of loneliness and social dissatisfaction. DevelopmentalPsychology, 29, 611-621.Pelham, W.E., & Bender, M.E. (1982). Peer relationships in hyperactive children: Description andTreatment. Advances in Learning and Behavioral Disabilities, 1, 365-436.Pliszka, S. R. (1989). Effect of anxiety on cognition, behavior, and stimulant response in ADHD.Journal of the American Academy of Child and Adolescent Psychiatry, 28, 882-887.Plizska, S.R. (1992). Comorbidity of attention deficit hyperactivity disorder and overanxious disorder.Journal of the American Academy of Child and Adolescent Psychiatry, 31, 197203.Puig-Antich J, Chambers W (1978), The Schedule for Affective Disorders and Schizophrenia forSchool-Age Children (Kiddie-SADS). New York: New York State Psychiatric Institute.Putallaz, M., & Gottman, J.M. (1981). An interactional model of children’s entry into peer groups.Child Development, 52, 986-994.Rubin, K.H., & Asendorpf, J. (1993). Social withdrawal inhibition, and shyness in childhood.(pp. 291-314). Hillsdale, NJ: Erlbaum.Rubin, K.H., Bukowski, W., & Parker, J.G. (1998). Peer interactions, relationships, and groups, In. W.Damon (Series Ed.) & N. Eisenberg (Vol. Ed.), Handbook of child psychology: Vol. 3. Social,emotional, and personality development (pp. 619-700). New York: Wiley.Rubin, K.H., Burgess, K.B., Kennedy, A.E., & Stewart, S.L. (2003). Social withdrawal in childhood. InE.J. Mash & R.A. Barkley (Eds.), Child Psychopathology (2nded., pp. 372-406). New York:Guilford. Peer Relationships and ADHD34 Rubin, K.H., Lemare, L.J., Lollis, S. (1990). Social withdrawal in childhood: Developmentalpathways to peer rejection. In S.R. Asher & J.D. Coie (Eds.), Peer rejection in childhood (pp.217-249). New York: Cambridge.Strauss, C.C., Lahey, B.B., Frick, P., Frame, C.L., & Hynd, G.W. (1988). Peer social status of childrenwith anxiety disorders. Journal of Consulting and Clinical Psychology, 56, 127141.Tannock, R. (2000). Attention-deficit/ hyperactivity disorder with anxiety disorders. In T.E Brown(Eds.), Attention-deficit disorders and comorbidities in children, adolescents, and adults(pp.125-170). Washington, DC: American Psychiatric Press.Stevenson-Hinde, J., & Shouldice, A. (1993). Wariness to strangers: a behavior systems perspectiverevisited. In K.H. Rubin & J.B. Asendorpf (Eds.), Social withdrawal, inhibition, and shyness inchildhood (pp. 101-116). Hillsdale, NJ: Lawrence Erlbaum.Suveg, C. & Zeman, J. (2004). Emotion regulation in children with anxiety disorders. Journal ofClinical Child and Adolescent Psychology, 33, 750-759.Tannock, R., & Schachar, R. (1992). Is ADHD with comorbid overanxious disorder different fromADHD? Presentation in Comorbidity of ADHD: Discriminating Features and MethodologicProblems. Symposium held at the 39thannual meeting of the American Academy of Child andAdolescent Psychiatry, Washington, DC, October, 1992.Vignoe, D., Berbue, R.L., & Achenback, T.M. (1999). Bibliography of published studies using thechild behavior checklist and related materials. Burlington: University of Vermont.Whalen, C.K., & Hencker, B. (1992). The social profile of attention-deficit hyperactivity disorder: Fivefundamental facets. Child and Adolescent Psychiatric Clinics of North America, 1, 395 – 410.Zongqui, Z., Dongmei, Z., Xiajun, S., & Xiafeng, D. (2006). Children’s experiences with peers andloneliness: A two-year longitudinal study. Acta Psychologica Sinica, 38, 743-750. Peer Relationships and ADHD35 Zarbatany, L., Van Brunschot, M., Meadows, K., & Penner, S. (1996). Effects of friendship and genderon peer group entry. Child Development, 67, 2287-2300. Peer Relationships and ADHD36 Table 1. Means and Standard Deviations for Demographic Variables and Child Diagnoses byGroup.ADHD (n =61)Comparison (n =60)tVariableADHD-C = 45; ADHD-I = 16Age8.23 (SD = 1.20)8.25 (SD = 1.19).08Gender !-.19 Female1818Male4342Anxiety Disorderb101-2.92**Oppositional Defiant Disorderb200-5.41**Full Scale IQc107.3 (SD = 14.5) 117.3 (SD = 11.5)4.17**TRF Anxietyd58.6 (SD = 7.9)52.6 (SD = 4.0)-5.10***p < .05. ** p < .01. ! t scores represent differences between groups using a dichotomously coded variable (1 = male,