INTERNALIZING SYMPTOMS AND PERSONALITY  COLOR PARENTAL REPORTS OF CHILD TEMPERAMENT  By  David Angus Clark                  A THESIS  Submitted to  Michigan State University  in partial fulfillment of the requirements for the degree of  Psychology - Master of Arts  2016         ABSTRACT INTERNALIZING SYMPTOMS AND PERSONALITY  COLOR PARENTAL REPORTS OF CHILD TEMPERAMENT  By  David Angus Clark  Depressed mothers have negatively distorted views of the personalities and behaviors of their children. This is potentially problematic as parental reports of child behavior are a popular method of collecting data on children. The current study sought to extend the existing literature by investigating the degree to which internalizing symptomology and personality traits influence both and . This was done by applying the tri-factor model (Bauer et al., 2013) to data collected from the parents of 222 children (ages 3-5 years). Personality and psychopathological symptoms were related to perspectives of their children. However, the effects varied somewhat across dimensions of child temperament, and mothers and fathers. These results support concerns that psychological characteristics influence parental perceptions of their children. This work provides guidance may be biased informants of child temperament.       TABLE OF CONTENTS  LIST OF TABLES            iv LIST OF FIGURES            v INTRODUCTION            1  Temperament and its Assessment         2  Informant Disagreement          3           5  The Current Study             6  METHOD             7  Participants             7  Measures            8   Child Temperament          8   Parent Internalizing Symptomatology       9   Parent Personality        10  Data Analytic Strategy        10    RESULTS           12  Preliminary Analyses         12  Effortful Control         13  Negative Affectivity          15  Surgency          17  DISCUSSION           19  Summary          19  Implications          22  Limitations and Future Directions        23  REFERENCES          25  APPENDICES          32                      iv LIST OF TABLES    Table 1: CBQ Means, Standard Deviations, and Mother-Father Scale Correlations    34  Table 2: Parent Internalizing Symptomatology and Personality Means, Standard    Deviations, and Mother-Father Correlations       35  Table 3: Unconditional Trifactor Model for Effortful Control       36  Table 4: Unconditional Trifactor Model for Negative Affectivity      37  Table 5: Unconditional Trifactor Model for Surgency        38  Table 6: Paths From Parent Characteristics to Consensus and Unique Perspective Factors   39       v LIST OF FIGURES   Figure 1: The Trifactor Model          33    1    Introduction Childhood temperament represents the foundation of adult personality (Shiner & DeYoung, 2013). Research on temperament therefore has the potential to both contribute to basic knowledge about human development, and inform programs and policies designed to promote positive outcomes across the lifespan (Moffitt et al., 2011). However, for this potential to be realized, researchers must be confident that individual differences in temperament are adequately assessed. This has proven to be a challenge, and debates over the assessment of temperament have generated decades of research and commentary.  One of the more contentious issues is the usefulness of parent reports of child temperamental attributes. A frequently levied criticism is that parents are biased informants, prone to over- or under-report certain behaviors. These criticisms have been fueled, in part, by (e.g., Duhig et al., 2000). Moreover, certain parental characteristics might accentuate general reporting biases, and thus contribute to greater disagreement between parents regarding the attributes of their children. A number of investigations have evaluated whether depressive symptomatology represents one source of parental bias (Richters, 1992). This work indicates that depressed mothersmay provide overly negative reports of their children (e.g., Muller, Achtergarde, & Furniss, 2011). The goal of the current study was to extend existing work on depression and the parent report method by examining the extent to which different kinds of internalizing symptomatology and personality traits relate to disagreement between maternal and paternal informants of temperament. This was done by applying the trifactor model  a latent    2  variable model that can efficiently isolate rater specific biases (Bauer et al., 2013) - to the parental ratings of 222 young (3-5 years) children.  Temperament and Its Assessment  emperament refers to the constitutionally based individual differences in reactivity and self-regulation that emerge early in life (Rothbart, 2011). There are a number of approaches for classifying the diverse array of individual differences evident in children. One prominent model posits that temperament during early childhood (age 3 to 7 years) can be characterized at the highest level by three broad dimensions: Effortful Control, Negative Affectivity, and Surgency (Rothbart, 2011). The dimension of Effortful Control captures an experience fear, anger, and other types of psychological distress (Rothbart, 2011). Last, the domain of Surgency captures how likely individuals are to experience positive emotions and rapidly approach potential rewards (Rothbart, 2011).These three dimensions of temperament are  Emotionality described in the adult personality literature (Tellegen & Waller, 2008).  Approaches to assessing temperament tend to fall into one of three broad categories: informant report, home observation, and laboratory assessment (Rothbart & Goldsmith, 1985). Informant report refers to when researchers ask individuals familiar with the target child (e.g., parents, teachers) to answer questions . In home observation, a trained researcher/research assistant spends time in th their behavior in a variety of situations, and codes that behavior. Finally, for laboratory assessment, children are    3  brought into a lab and their behaviors are observed and coded across controlled situations designed to elicit certain trait-relevant behaviors and emotions.  Of the various approaches to assessing child temperament/behavior, using parents as informants is the most popular (Goldsmith & Gagne, 2012). This is understandable considering that parent reports are quick, inexpensive, and relatively easy to obtain. Furthermore, parental reports may be more ecologically valid than the other methods of assessment (e.g., Rothbart & Goldsmith, 1985; but see Lo, Vroman, & Durbin, 2015). Despite these advantages, the parent report method has often been criticized on the grounds that parents might be biased informants of child behavior (Seifer, 2002).  Informant Disagreement Criticisms of the parent report method are often motivated by the finding that parent reports of temperament are only modestly (at best) correlated with alternative methods for assessing temperament. For example, parental reports converge only modestly (rs around .01 -.30) with other approaches means of assessing child behavior, such as home observation and laboratory assessment (e.g., Bates & Bales, 1982; Durbin & Wilson, 2012; Gartstein & Marmion, 2008; Sameroff, Seifer, & Elias, 1982; Seifer, Sameroff, Dickstein, Schiller, & Hayden, 2004; Vaughn, Bradley, Joffe, Seifer, & Barglow, 1987; Vaughn, Taraldson, Crichton, Egeland, 1981).Achenbach and colleagues (1987) meta-analyzed over 119 studies that included multiple informants of childhood behavior problems and found varying, yet generally moderate, levels of agreement between various informants (e.g., r = .60 between parents, r = .27 between parents and teachers, r = .24 between parents and mental health workers). Agreement tended to be higher for more overt behaviors such as hyperactivity and    4  aggression (e.g., r = .32 between parents and teachers), than convert behaviors such as anxiety and depression (e.g., r = .21 between parents and teachers). Another meta-analysis - exclusively looking at agreement between mothers and fathers - found moderate correlations between  of problem behavior (r = .46-.66; Duhig, Renk, Epstein, & Phares, 2000). When investigating parental agreement on fifteen scales designed to measure child temperament, Rothbart and colleagues (2001) reported only moderate correspondence (most rs between .3 and .4).  Although these studies reveal that mother-father agreement is generally higher than agreement between a parent and a non-parent informant (Achenbach, McConaughy, & Howell, 1987; Duhig et al., 2000), the degree of correlation between parents still usually only hovers in the .3 to .6 range. This range is comparable to results from a meta-analysis of inter-rater agreement on the Big Five personality traits (Extraversion, Agreeableness, Conscientiousness, Neuroticism, and Openness) in adulthood (Connelly & Ones, 2010), and raises questions about moderators of agreement (see Funder, 2012). If systematic causes for disagreement between informants can be identified, this information could be leveraged to enhance the quality of parental reports of temperament. Several models of cross informant agreement have been proposed in the literature (e.g., De Los Reyes & Kazdin, 2005; Funder, 2012; Kenny, 1994). One insight common across these models is that certain informant characteristics are likely to increase inter-informant disagreement. Thus, one promising avenue for understanding disagreement is to investigate the characteristics of raters that may lead to differences in their ratings. Parental psychopathology, particularly depressive symptomatology, has often been proposed as an informant characteristic that might serve to bias reports, and thus increase disagreement between informants.     5    Several concerns have been raised about the validity of the reports of child behavior obtained from depressed parents. The two traditionally competing models in this area are known as the depression distortion model, and the accuracy model (Fergusson, Lynskey, & Horwood, 1993; Richters, 1992). The distortion model proposes that depressive symptomatology warps parental views such that maladaptive child behaviors are over-reported. Conversely, the accuracy model posits that depressed parents are actually more accurate informants than their non-distressed peers (Fergusson et al., 1993; Richters, 1992). The accuracy model is in part based on the ideas that the typical way of viewing offspring is with a positive bias (e.g., Seifer et al., 2004), and that the children of depressed parents really do have more behavior problems. Generally speaking, empirical evidence has tended to converge on the depression distortion model, suggesting that depressive symptomology does indeed bias maternal reports to some degree (e.g., Boyle & Pickels, 1997; Durbin & Wilson, 2012; Fergusson et al., 1993; Leerkes & Crockenberg, 2003; Muller, Achtergarde, & Furniss, 2011 Najman et al., 2000; Youngstrom, Loeber, & Stouthamer-Loeber, 2000). Although the preponderance of evidence supports the depression distortion model, several details remain unclear. For example, how large is the depression-distortion effect? Some studies indicate that depression considerably distorts maternal reports (e.g., Muller, Achtergarde, & Furniss, 2011), yet others have suggested that the effect may be only modest in size (e.g., Briggs-Gowan, Carter, & Schwab-Stone, 1996). Furthermore, there is the issue of what kinds of child behaviors are most likely to be distorted by depressive symptoms - all behaviors, or only particular ones?  Gartstein and colleagues (2009) found support for the depression distortion model exclusively in gender stereotyped problem behaviors; depressed mothers over-   6  behaviors, but not vice-versa. Additionally, the existing work leaves open the issues of whether also affected by depressive symptomatology, and if other parental characteristics besides depression can distort parental views. Work on the depression distortion model is limited by a somewhat narrow focus on maternal informants and depressive symptomatology. Extending the boundaries of this model is important for increasing the quality of temperament assessment. The underrepresentation of fathers is not unique to the depression distortion literature (Phares, 1992; Phares, Fields, Kamboukos, & Lopez, 2005), but it is problematic because fathers are frequently used as informants of child behavior alongside mothers (Duhig et al., 2000). Additionally, many of the potential mechanisms underlying the depression distortion effect (e.g., preferential attention to negative stimuli, greater recall of negative events; Gelfand & Teti, 1990; Mathews & MacLeod, 2005; White & Barrowclough, 1998) may not be unique to depression per se (Chan, Goodwin, & Harmer, 2007; Isbell, 2013). For example, Youngstrom, Izard, and Ackerman (1999) found that general maternal dysphoria (both self-reported and other-observed) was associated with biases in rating child. Findings such as these raise the possibility that the depression distortion model might also extend to characteristics similar to depression such as anxiety, and general personality attributes like Negative Emotionality.  The Current Study   There were two primary aims of the current study. The first was to evaluate the extent to which depressive symptomatology is related to  and s of child temperament, and any disagreement between parents. The second was to test whether related constructs such as anxiety, and the personality trait of Negative Emotionality, also relate to parental reports. Finally, in an exploratory effort to understand whether there are parental    7  attributes not characterized by negative affect that are related to reporting and disagreement, the effects of other parent personality traits (Positive Emotionality and Constraint) were also examined. Associations between parental attributes and parental reports were examined for all three of the broad dimensions of temperament described above (Effortful Control, Negative Affectivity, and Surgency). This allowed for an investigation of whether distortion effects are specific to certain child traits/behaviors.     Method Participants The sample comes from the second and third generations of the Family Transitions Project (FTP). The FTP is an ongoing longitudinal study of 559 target individuals (the second generation) and their families of destination in adulthood. The FTP represents a merging of two previously distinct longitudinal studies, the Iowa Youth and Families Project (IYFP), and the Iowa Single Parents Project (ISPP). Participants for the IYFP were initially recruited when they were in seventh grade; to be eligible for participation individuals had to come from a two parent family, and have at least one sibling within four years of age (Neppl et al., 2010). Participants for the ISPP were initially recruited when they were in ninth grade; to be eligible for participation they needed to have come from a family headed by a mother whom had experienced divorce within the past two years, and have a sibling within four years of age (Neppl et al., 2010). The IYFP and ISPP were combined into the FTP in 1994 when the target participants were in 12th grade.  Data for the FTP is collected regularly by trained interviewers who visit the target participants in their home. During these visits targets provide responses to a number of    8  questionnaires that span a range of topics. Similar information is also collected from individuals close to the target participant, individuals such as parents, siblings, romantic partners, and when are older than 18 months).  The sample used in this study consists of the families of 222 children, aged 3-5 years (102 female, 120 male). These children are the offspring of the original FTP targets, and thus the families include one original FTP target, her/his child, and the other parent of the child. In almost all cases, both parental informants are the biological parents of the child. The parent-reported temperament data are taken from when the Child Behavior Questionnaire (CBQ; Rothbart, Ahadi, Hershey, & Fisher, 2001) was first administered to each child. This occurred at age 3 for 190 children, age 4 for 24 children, and age 5 for 8 children. In total there were 211 maternal reports of child temperament, and 186 paternal reports of child temperament. Given the geographic origins of the FTP (rural Iowa), the sample is almost exclusively European American. Furthermore, participants are primarily of middle and lower-middle class background. For a more thorough description of the FTP, IYFP, and ISPP, see Elder and Conger (2000), and Neppl et al. (2010).      Measures  Child Temperament. The Child Behavior Questionnaire (CBQ) is a caregiver report instrument that contains 195 items which tap one of 16 facets of temperament: Activity Level, Anger/Frustration, Attentional Focusing, Attentional Shifting, Discomfort, Falling Reactivity/Soothability, Fear, High Intensity Pleasure, Impulsivity, Inhibitory Control, Low Intensity Pleasure, Perceptual Sensitivity, Positive Anticipation/Approach, Sadness, Shyness, and Smiling/Laughter (see Rothbart et al., 2001 for conceptual definitions). These 16 facets of    9  temperament are subsumed by the aforementioned superordinate dimensions of Effortful Control, Negative Affectivity, and Surgency (Rothbart et al., 2001).  The CBQ presents respondents with particular behaviors and asks them to rate the degree to which those behaviors were characteristic of the target child over the past six months. Responses are given on a 7-point Likert-type scale that ranges from 1: extremely untrue of your 7: . has not recently been in the item-relevant situation. Item content was derived both rationally from the conceptual definitions of the 16 facets, and from extensive parental interviews (Rothbart et al., 2001). Items were selected to be developmentally appropriate for children aged 3-8 years.  ice when asked . .  Parent Internalizing Symptomatology. The Mini-Mood and Anxiety Symptom Questionnaire (Mini-MASQ) is a 26 item self-report inventory of anxious and depressive symptomatology (Casillas & Clark, 2000). Respondents rate the extent to which they have experienced various symptomatic feelings over the past week on a 5-point Likert-type scale 1: 5: . Items fall into one of four categories: Generalized Distress-Depression, Anhedonic Depression, Generalized Distress-Anxiety, and Anxious Arousal. .   Mini-MASQ data was taken from the same wave of data collection as their child temperament data. Anhedonic depression and anxious arousal scores were computed by averaging together the items of each respective scale. The items from the generalized distress-   10  depression and generalized distress-anxiety scales were averaged together to create one A also computed by averaging together all items of the questionnaire.  Parent Personality. The Multidimensional Personality Questionnaire  Brief Form (MPQ-BF) is a 156 item self-report inventory of personality (Patrick, Curtin, & Tellegen, 2002). The MPQ-BF contains several scales for capturing various facets of personality (e.g. stress reactivity, harm avoidance). Like the CBQ, these scales are subsumed by three higher order dimensions: Negative ENegative Affectivity), Positive ESurgency), and CEffortful Control). The items of the MPQ-BF come from the full, 300 item Multidimensional Personality Questionnaire (Tellegen & Waller, 2008). Scale scores derived from the full and brief versions of the MPQ are highly correlated (e.g. .85-.95; Donnellan, Conger, & Burzette, 2007).   The MPQ-BF presents respondents with a trait and asks them to indicate whether they think that that trait is true of them (1), or not true of them (0). enthusiasticremain calm, even in difficult situations. All MPQ-BF data was taken from the tenth wave of data collection (when FTP targets were roughly 27 years old), regardless of the age of the focal child. Personality dimension scores were calculated by averaging together the items of the scales that make up a particular dimension.  Data Analytic Strategy   The primary analyses of this study were based on rifactor model (see Figure 1). The trifactor model is an extension of the bifactor model (e.g., Chen, Hayes, Carver, Laurenceau, & Zhang, 2012) that allows for the integration of ratings across multiple informants (Bauer et al., 2013). In this model, the ratings from multiple informants    11  (e.g., mothers and fathers) are used as indicators of three major types of latent factors. The first is a general consensus factor that represents the overlap between informantss of the target on the construct of interest (i.e., agreement); this factor captures covariation that is common across ngs on all indicators (CON in figure 1). The second type of factor is the unique perspective factor. There is one unique perspective factor for each informant (MP and FP in Figure 1). These factors capture covariance that is common across a single informanindicators. The unique perspective factors represent a specific informantthe target on the construct of interest, a view that is not shared by the other informant (i.e., disagreement). Finally, there are rating scale (i.e., indicator) specific factors (S1 through S4 in Figure 1). There is one of these factors for each scale. These factors capture the covariance in indicators that is unique to the specific indicator, but shared across informants. All of the latent factors described here are conceptualized as being orthogonal (i.e., uncorrelated). Importantly, the factors in this model may be regressed onto external variables (Bauer et al., 2013; see Figure 1) to examine the predictors of informantof the target (i.e., agreement and disagreement). When the unique perspective factors are regressed onto external variables (e.g., depressive symptomatology), results can be interpreted as the extent to which that particular external variable predicts disagreement via its relation to the idiosyncratic impression of the target. When the consensus factor is regressed onto external variables, results can be interpreted as the extent to which that particular variable predicts  of the target, or their agreement.  The goals of this study were accomplished by first estimating the unconditional trifactor model (i.e., the trifactor model without external predictor variables) for a major dimension of temperament, and then subsequently regressing the consensus and unique perspective factors    12  onto the parental characteristics of interest. Each higher-order dimension of temperament was examined in turn. Specific details regarding model specification are provided below. All models were estimated in Mplus version 7.4 (Muthen & Muthen, 1998-2012) using full information maximum likelihood techniques (FIML). This estimation procedure is recommended in the face of missing data (Allison, 2009; Graham & Coffman, 2012).  Results  Preliminary Analyses  Means and standard deviations for mother and father reported child temperament can be found in Table 1. Generally, children were rated higher on scales assessing positive affect (e.g., smiling and laughter, low intensity pleasure) than they were on scales measuring negative affect (e.g., sadness, fear) in terms of the average scores. ds between .1 and .2; see Table 1). The largest discrepancies were observed for the smiling and laughter (d = .50) and low intensity pleasure (d = .42) scales. Correlations between the mother and father reported temperament scales (see Table 1) ranged from .20 (sadness) to .62 (shyness). Most correlations were around .3 to .4 in size, consistent with prior work (e.g., Rothbart et al., 2001). As expected, correlations were often higher for scales focused on more overt behaviors (e.g., high intensity pleasure, shyness).   Means and standard deviations for  internalizing symptomatology and personality traits can be found in Table 2. The most highly endorsed type of symptomatology was anhedonic depression whereas anxious arousal was the least. Mothers and fathers scored similarly on the measures of internalizing symptomatology and personality (most ds between .01 and .10; see Table 4). The largest observed difference was that mothers    13  reported more Constraint than fathers (d = .61). attributes ranged from .09 (Positive emotionality) to .39 (anxious arousal).  Effortful Control The Effortful Control dimension comprises 4 CBQ scales: attentional focusing, inhibitory control, low intensity pleasure, and perceptual sensitivity (Clark et al., 2016; Rothbart et al., 2001)1. To estimate the unconditional trifactor model for Effortful Control, both mothers 4 Effortful Control scales were specified to load on the latent consensus factor.  on the mother perspective factorto the 4 scales were specified to load on the father perspective factor. Finally, scale factors were specified for each individual scale aded on a single scale factor. At this stage, the variance of all latent factors was set to 1, and mean of all latent factors was set to 0. All covariances for the latent factors were fixed to 0.  Initially, all factor loadings and indicator intercepts in the unconditional model were allowed to freely estimate. A series of constraints on the factor loadings and indicator intercepts were then tested (via the significance of the change in 2) with the goal of obtaining the most parsimonious model possible2consensus (e.g., maternal and paternal reported attentional focusing were set to equality) and scale factors, as well as the indicator intercepts, to equality without observing a significant drop in fit. All unique perspective factor loadings were able to be constrained across mothers and fathers except for inhibitory control. In this, the most justifiably constrained model, the father unique perspective factor variance and mean were allowed to freely vary. The final unconditional                                                  1 Preliminary exploratory factor analyses of the CBQ generally supported the approach to constructing dimensions used in the literature. Full results available upon request.  2 Full results from this model fitting process (as well as for the other dimensions) are available upon request    14  trifactor model for Effortful Control fit the data at: 2 = 16.90, df = 17 p = .46; RMSEA = .00; SRMR = .07; CFI = 1.00; TLI = 1.00.  The unstandardized and standardized parameter estimates for this model can be found in table 3.  The attentional focusing and inhibitory control scales loaded higher on the consensus factor than the unique perspective factors, whereas the low intensity pleasure and perceptual sensitivity scales loaded more strongly on the unique perspective factors. This suggests that lected more strongly in their ratings of attentional focusing and inhibitory control, whereas ratings of low intensity pleasure and Effortful Control. It is also worth noting that loadings on the scale factors were non-trivial.  Parental characteristics were introduced as predictors of the consensus and unique perspective factors one at a time. In these models the mother unique perspective factor was regreswas regressed on the . The consensus factor was regressed on both the attributes. This procedure is illustrated in Figure 1. In all, 7 conditional trifactor models were estimated, one for each parental characteristic. Standardized regression coefficients and 95% confidence intervals can be found in table 6. ith -.21 to -.45). On the other hand, maternal Constraint and Positively Emotionality were positively associated with the consensus factor to a modest nality was also significantly    15  - Constraint was positively associated with the unique  The father unique perspective factor was significantly, negatively -.25), and significantly, positively predicted by .25).  Importantly, despite differences across parents in the pattern of overall significance, the associations between significantly across parents (see Table 6). That is, it was possible to constrain most corresponding maternal and paternal paths (i.e., the paths from parent characteristics to unique perspective factors, and the paths from parent characteristics to the consensus factor) to the same values without observing a significant drop in model fit (full results available upon request). Only the paths from general distress and Constraint to the consensus factor significantly differed across mothers and fathers. In both instances, the maternal characteristic was more strongly associated with the consensus factor than was the corresponding paternal characteristic.  Gender differences across boys and girls were also examined by estimating the conditional trifactor model separately for boys and girls and imposing equality constraints on the paths from parental characteristics to the factors. There were few significant differences across girls and boys in how parent symptomatology and personality related to shared and unique parental views of Effortful Control. Positive Emotionality wperspectives of boys (full results available upon request).           Negative Affectivity  The Negative Affectivity dimension includes 5 CBQ scales: sadness, anger/frustration, discomfort, fear, and soothability (Clark et al., 2016; Rothbart et al., 2001). The unconditional    16  trifactor model for Negative Affectivity was specified in the same general manner as described above for Effortful Control. The only major difference was that the loadings for one of the scale factors - anger/frustration - were set to 0 (effectively eliminating it)loaded only trivially on this factor, and allowing these loadings to freely estimate led to problems in estimation (specifically, issues of linear dependency between other scale factors). Factor loadings and intercepts were initially allowed to freely estimate, and constraints were systematically applied and tested as above to achieve the most parsimonious model. In the final model, factor loadings on the consensus factor, unique perspective factors, and scale factors were constrained to equality across mothers and fathers. Furthermore, all intercepts except for the soothability intercept were constrained to equality across parents. Every factor except for the father unique perspective factor was fixed to have a variance of 1 and mean of 0; the father unique perspective factor had a freely estimating mean and variance.   The unconditional trifactor model for Negative Affectivity fit the data at: 2 = 56.47, df = 33,  p < .01, df = 28; RMSEA = .06; SRMR = .06; CFI = .95; TLI = .93. Unstandardized and standardized parameter estimates for this model can be found in Table 4. The sadness, anger/frustration, and soothability scales loaded modestly to strongly on the consensus factor, their responses on these three scales. Conversely, the discomfort and fear scales effectively did Affectivity is unrelated to responses on these scales. With the exception of the anger/frustration scale, all scales loaded more strongly on the unique perspective factors than the consensus factor (anger/frustration still loaded moderately on these factors). Thus, on these five scales are more a reflection of parent specific impressions of Negative Affectivity    17  than shared impressions of Negative Affectivity. Notably, the fear and discomfort scales, which were the least related to the consensus factor, loaded moderately on the scale factors. This would ared overarching view of Negative Affectivity is unrelated to responses on these two scales, there is some scale specific consensus.     Standardized regression coefficients and 95% confidence intervals for the conditional trifactor model paths can be found in Table 6. Maternal Negative Emotionality significantly predicted the consensus factor to a moderGeneral distress, Anhedonic depression, total distress, and Negative Emotionality were all significantly, positively Positive Emotionality was als-.19).  The strength of the associations between internalizing symptomatology, personality traits, and the three factors did not significantly differ across mothers and fathers. Similarly, the effects described here were not significantly different when comparing boys and girls as described above.      Surgency The Surgency dimension is made up of 5 CBQ scales: impulsivity, high intensity pleasure, smiling and laughter, shyness, and positive anticipation (Clark et al., 2016; Rothbart et al., 2001). Specification of the unconditional trifactor model for Surgency was similar to the procedure described for the other two dimensions. As was the case with Negative Affectivity, triv were specified to all load on the same scale factor rather than loading on distinct scale factors. These two scale factors were combined    18  because they were correlated (r > .90) to the point where it caused problems in estimation, however, unlike the impulsivity and anger/frustration factors, the individual scales did not load negligibly on their respective factors. Factor loadings and intercepts were all freely estimated initially. Systematically applying and evaluating equality constraints across parents resulted in a final model in which factor loadings on the consensus factor, the unique perspective factors, and the scale factors, were equal across parents. Moreover, all intercepts except for smiling and laughter were equal across mothers and fathers. The final unconditional trifactor model for Surgency fit the data at: 2 = 74.99, df = 33, p < .01, df = 33; RMSEA = .08; SRMR = .09; CFI = .94; TLI = .92. Unstandardized and standardized parameter estimates can be found in Table 5. The impulsivity, high intensity pleasure, and shyness scales all loaded moderately to strongly on the consensus factor. smiling and laughter, and positive anticipation, scales loaded negligibly on the consensus factor, suggesting that these scales largely fail to rloaded moderately to strongly on the unique perspective factors, indicating that mothers and ratings on each of the Surgency subscales. With the exception of the impulsivity scale, all scales loaded non-trivially on their scale specific factor. Standardized regression coefficients and 95% confidence intervals for conditional trifactor model paths can be found in Table 6. ss and constraint were both -.28) associated with lower shared views of Surgency. On the other hand, maternal Positive Emotionality was    19  significantly, positively associated  with the maternal unique perspective factor, and paternal anhedonic depression -.19) and anxious arousal -.18) were significantly, negatively associated with the father perspective factor.  The associations between the parental characteristics and the major factors did not significantly differ across mothers and fathers. Furthermore, there were no significant differences in the effects described here when comparing boys and girls.                Discussion  The current study evaluated how internalizing symptomatology and personality traits are related to inter-parent agreement and disagreement when rating child temperament. All told, there was evidence that both maternal and paternal reports of child temperament were colored by their internalizing symptomatology and personality traits.  Summary  The conditional Effortful Control trifactor models revealed that anhedonic depression negatively predicted the father factor, and negative emotionality negatively predicted the mother factor. These findings indicate that anhedonic fathers, and dispositionally negative mothers, hold lower Effortful Control, impressions that are unique from that of the other parent. Additionally, maternal distress, anxious arousal, and constraint, and parental positive emotionality, all significantly predicted the consensus factor.  The conditional trifactor models for Negative Affectivity suggested Negative Affectivity are related to distress, anhedonic depression, and Negative Emotionality.  Negative Affectivity on the other hand were predicted by distress, anhedonic depression, and both Negative and Positive Emotionality. With the exception of the Positive Emotionality path, all effects were positive. This indicates that heightened levels of internalizing symptomatology and Negative Emotionality    20  correspond to idiosyncratically higher levels of perceived Negative Affectivity. There was also one significant pathway from a parental attribute to the consensus factor; specifically, maternal Negative Emotionality was positively related to gative Affectivity.  The conditional trifactor models for Surgency suggested mwere positively associated with Positive Emotionality. This implies that the more positive mothers are in general, the more likely they are to uniquely report that their children are high in Surgency. unique perspectives were negatively predicted by anhedonic depression and anxious arousal. Thus fathers who are more depressed and anxious are more likely to think their children are lower in Surgency compared to mothers. In addition to these paths, maternal distress and Constraint significantly predicted the consensus factor.  Higher levels of maternal distress were associated with higher shared views of Surgency, whereas higher levels of Constraint were associated with lower shared views of Surgency. Notably, the path from maternal distress to the consensus factor appears non-intuitive, implying that mothers with more internalizing symptomatology (the other internalizing paths, though not significant, ran in the same direction) have children that are seen as more positive by both parents. This may partly be a result of Surgency dimension. If impulsivity is viewed as a negative trait (hinted at by the negative paths from internalizing symptomatology to the Effortful Control factors), then negative affect may increase ratings on that scale, thus increasing scores on the Surgency dimension as a whole. Follow-up analyses supported this notion. When impulsivity was removed from the Surgency model the path from maternal distress to the consensus factor -.36; 95% CI: -.63, -.10), implying that heightened distress was associated with a decreased shared view of child Surgency.    21  Overall, mointernalizing symptomology and personality traits. In other words, disagreement between parents regarding where their children fall on some temperamental trait is partially attributable to the psychological characteristics of each of the parents. Importantly, although the pattern of attributes to the unique perspective factors were consistent for mothers and fathers. This implies psychological characteristics examined here. In other words, the intrapsychic processes the bias maternal reports appear to operate in a similar fashion for paternal reports.    In addition to these associations with the unique perspective factors, there were several significant pathways pointing to an association between parental attributes and the consensus factor, most notably for the dimension of Effortful Control. Although most of these paths did not significantly differ across mothers and fathers, some did, suggesting that the attributes of a single parent can influence how both parents perceive and report on their child. These paths could represent parent-offspring transmission effects, either through environmental or genetic pathways. For example, genetically informed research suggests that maternal internalizing symptomatology can act as a purely environmental risk for unconstrained behavior (e.g., Kim-Cohen, Moffitt, Taylor, Pawlby, & Caspi, 2005; Leve et al., 2010). And given that paths from shared genetic makeup between parents and offspring. On the other hand, parents may influence how the other parent perceives the child on their basis of their personality or symptomatology. It    22  internalizing symptomatology or personality, then such perceptions might also influence how that other parent views the child. Given that mothers traditionally spend more time with young children (e.g., Barnett & Baruch, 1987), this could account for why most attributes that were larger than the paternal effects, in the case of any differences. Of course, these possible explanations need not be mutually exclusive. Implications  The current findings have implications for the measurement of child temperament and behavior. Specifically,  disagreement on temperament questionnaires  reports may be colored by a wide array of their own symptoms and traits. In other words, parental reports capture some information about the parental informant in addition to information about the child. The effects were not especially large in size, but considering that the aim of parent report questionnaires is to collect information about the child, such biases might prove important. Broadly speaking, these findings are consistent with past studies. For instance, there has been evidence that depressed mothers may exaggerate the presence of problematic child behavior (e.g. Fergusson et al., 1993). The findings here extend reports of temperament. Likewise, the current report suggests that reports by fathers may also suffer from these biases.  Practically speaking, this study should inspire investigators to think about parent reports in a more strategic manner, and to collect more information about the psychological characteristics of the parents.  The current findings highlight the need to collect data from    23  multiple sources/methods, as any one source will provide data that may conflate child temperament with a number of irrelevant personal factors. The present pattern of findings especially emphasize that care is needed in soliciting parental reports when there are likely to be elevated levels of parental internalizing symptomatology in a sample. For example, researchers interested in studying the influence of parental internalizing psychopathology on child development would do well to rely on methods of child assessment beyond parent report. Even when there is no a priori reason to expect widespread pathology in a sample, it would be beneficial to include brief assessments of internalizing symptomatology and parent personality along with assessments of child temperament.   Limitations and Future Directions There are several limitations to the present study that are worth acknowledging. First and foremost is the issue of generalizability. The current findings are based on a single measure of parent reported temperament. However, the CBQ is widely used and similar in structure and content to other parent report scales. Most parent report scales, no matter the overarching construct of interest, usually operate by asking parents to report on specific child behaviors. Second, the vast majority of participants were European American and the target ages of the children were young. Future investigations should test the extent to which child age and family ethnicity moderate the findings reported here. Also, parent symptomatology and personality was assessed via self-report. Self-reports provide meaningful information, but it would be helpful to know if and how the effects reported here change as a function of the way parent characteristics are measured. For example, it might be the case that certain effects are stronger when parental symptomatology is assessed via clinical interview.     24  Future work, in addition to generalizing the present findings as recommended above, should attempt to evaluate the processes underlying the effects reported here. For example, scale (as opposed to dimension) level analyses could shed light on what particular items are more likely to be influenced by parental attributes. This could reveal specific child behaviors that mothers and fathers are especially unlikely to agree with each other over, and some the factors that contribute to this disagreement. A more fine grained analysis of the role of personality is also warranted. In the current study, personality was kept at the level of superordinate dimensions. To better understand how parent personality influences ratings of child behavior it is worth examining personality at the facet level. This provides a more detailed view of the relationship between parent personality and parent perceptions of child temperament.  Subsequent research could also attempt to extend the findings reported here to other common informants of child temperament. For example, teachers and clinicians often provide reports of child behavior. It is just as useful to know about how their attributes affect reports of child behavior as it is to know about parents. Indeed, it could even be worth assessing the extent to which behavioral coders (of laboratory assessments and home observations) are influenced by certain characteristics. Informants such as teachers and clinicians have a wealth of experience regarding child development to draw on when filling out temperament questionnaires, and behavioral coders frequently base their ratings on clearly defined coding schemes and observed behaviors. Despite this, it may be misguided to simply assume that all their ratings are immune from the potentially distorting effects of certain personality traits and pathologies.      25             REFERENCES     26  REFERENCES   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.   Barnet, R. C., & Baruch, G. K.  (1987).  Determinants of father  Journal of Marriage and the Family, 49, 29-40.  Bates, J. E., & Bayles, K. (1982).   of their children from age 6 months to 3 years. Merrill-Palmer Quarterly, 30(2),   111-130.   Bauer, D. J., Howard, A. L., Baldasaro, R.E., Curran, P. J., Hussong, A. M., Chassin, L., &   Zucker, R. A. (2013). A trifactor model for integrating ratings across multiple  informants. Psychological Methods, 18(4), 475-493. doi: 10.1037/a0032475  Boyle, M. H., & Pickles, A. R. (1997). Influence of maternal depressive symptoms on ratings of  childhood behavior. Journal of Abnormal Child Psychology, 25(5), 399-412.   Briggs-Gowan, M. J., Carter, A.S., & Schwab-Stone, M. (1996). Discrepancies among mother,   child, and teacher reports: Examining the contributions of maternal depression and   anxiety. Journal of Abnormal Child Psychology, 24(6), 749-765.   Carey, W. B. (1983). Some pitfalls in infant temperament research. Infant Behavior and   Development, 6, 247-254.   Casillas, A., & Clark, L. A. (2000, May). The mini-mood and anxiety symptom questionnaire  (Mini-MASQ). Poster presented at the 72nd Annual Meeting of the Midwestern Psychological Association, Chicago, IL.  Chan, S. W. Y., Goodwin, G. M., & Harmer, C. J. (2007). Highly neurotic never-depressed   students have negative biases in information processing. Psychological Medicine, 37,   1281-1291. doi: 10.1017/S0033291707000669  Chen, F. F., Hayes, A., Carver, C. S., Laurenceau, J-P., & Zhang, Z. (2012). Modeling general   and specific variance in multifaceted constructs: A comparison of the bifactor model to   other approaches. Journal of Personality, 80 (1), 220-250.  doi: 10.1111/j.1467-6494.2011.00739.x  Chi, T. C., & Hinshaw, S. P. (2002). Mother-child relationships of children with ADHD: The   role of maternal depressive symptoms and depression-related distortions. Journal of   Abnormal Child Psychology, 30(4), 387-400.      27  Church, A. T. (1994). Relating the Tellegen and five-factor models of personality structure.   Personality Processes and Individual Differences, 67(5), 898-909.   Clark, D. A., Donnellan, M. B., Robins, R. W., & Conger, R, D. (2015). Early adolescent  temperament, parental monitoring, and substance use in Mexican-origin adolescents.  Journal of Adolescence, 41, 121-131. doi:10.1016/j.adolescence.2015.02.010  Clark, D. A., Listro, C. J., Lo, S. L., Durbin, C. E., Donnellan, M. B., & Neppl, T. K. (2016).  Measurement invariance and child temperament: An evaluation of sex and informant   differences on the Child Behavior Questionnaire. Psychological Assessment.  doi: 10.1037/pas0000299  Connelly, B. S., & Ones, D. S. (2010). An other perspective on personality: Meta-analytic   edictive validity. Psychological Bulletin,  136(6), 1092-1122. doi: 10.1037/a0021212  Conrad, M., & Hammen, C. (1989). Role of maternal depression in perceptions of child  maladjustment. Journal of Consulting and Clinical Psychology, 57(5), 663-667.    Crockenberg, S., & Acredolo, C. (1983). Infant temperament ratings: A function of infants, of   mothers, or both?  Infant Behavior and Development, 6, 61-72.   De Los Reyes, A., & Kazdin, A. E. (2005). Informant discrepancies in the assessment of   childhood psychopathology: A critical review, theoretical framework, and   recommendations for further study. Psychological Bulletin, 131(4), 483-509.  doi: 10.1037/0033-2909.131.4.483  Donnellan, M. B., Conger, R. D., & Burzette, R. G. (2007). Personality development from late  adolescence to young adulthood: Differential stability, normative maturity, and evidence  for the maturity-stability hypothesis. Journal of Personality, 75(2), 237-264.  doi: 10.1111/j.1467-6494.2007.00438.x   Duhig, A. M., Renk, K., Epstein, M. K., & Phares, V. (2000). Interparental agreement on  internalizing, externalizing, and total behavior problems: A meta-analysis. Clinical  Psychology: Science and Practice, 7(4), 435-453.   Durbin, C. E., & Wilson, S. (2012). Convergent validity of and bias in maternal reports of child  emotion. Psychological assessment, 24(3), 647-660. doi: 10.1037/a0026607  Elder, G. H., & Conger, R. D. (2000). Children of the Land: Adversity and success in rural   America. Chicago, IL: University of Chicago Press.   Fergusson, D. M., Lynskey, M. T., & Horwood, L. J. (1993). The effect of maternal depression  on maternal ratings of child behavior. Journal of Abnormal Child Psychology, 21(3),   245-269.       28  Funder, D. C. (2012). Accurate personality judgment. Current Directions in Psychological   Science, 21(3), 177-182. doi: 10.1177/0963721412445309  Gartstein, M. A., Bridgett, D. J., Dishion, T. J., & Kaufman, N. K. (2009). Depressed mood and  maternal report of child behavior problems: Another look at the depression-distortion   hypothesis. Journal of Applied Developmental Psychology, 30, 149-160.  doi: 10.1016/j.appdev.2008.12.001   Gartstein, M. A., & Marmion, J. (2008). Fear and positive affectivity in infancy:   Convergence/discrepancy between parent-report and laboratory-based indicators. Infant  Behavior and Development, 31, 227-238. doi: 10.0716/j.infbeh.2007.10.012  Gelfand, D. M., & Teti, D. M. (1990). The effects of maternal depression on children. Clinical   Psychology Review, 10, 329-353.   Goldsmith, H. H., & Gagne, J. R. (2012). Behavioral Assessment of Temperament. In M.  Zentner, & R. L. Shiner (Eds.), Handbook of Temperament (pp. 209-228).  New York: The Guilford Press  Goodman, S. H., Rouse, M. H., Connell, A. M., Broth, M. R., Hall, C. M., & Heyward, D.   (2011). Maternal depression and child psychopathology: A meta-analytic review.  Clinical Child and Family Psychology Review, 14, 1-27.  doi: 10.1007/s10567-010-0080-1   Kenny, D. A. (1994). Consensus. In Interpersonal perception: A social relations analysis (pp.   50-81). New York, NY: The Guilford Press.     Kim-Cohen, J., Moffitt, T. E., Taylor, A., Pawlby, S. J., & Caspi, A.  (2005).  Maternal   Archives of General Psychiatry, 62,   173-181.    Leerkes, E. M., & Crockenberg, S. C. (2003). The impact of maternal characteristics and   sensitivity on the concordance between maternal reports and laboratory observations of   infant negative emotionality. Infancy, 4(4), 517-539.   Leve, L. D., Kerr, D. C. R., S  (2010).  Infant pathways to externalizing behavior: Evidence of genotype x environment   interaction.  Child Development, 81(1), 340-356.   Lo, S. L., Vroman, L. N., & Durbin, C. E. (2015). Ecological validity of laboratory assessments  of child temperament: Evidence from parent perspectives. Psychological Assessment,  27(1), 280-290. doi: 10.1037/pas0000033     Mathews, A., & MacLeod, C. (2005). Cognitive vulnerability to emotional disorders. Annual  Review of Clinical Psychology, 1, 167-195.  doi: 10.1146/annurev.clinpsy.1.102803.143916     29    (2011). A gradient of childhood self-control predicts health, wealth, and public safety. PNAS, 108(7), 2693-2698.  Muller, J. M., Achtergarde, S., & Furniss, T. (2011). The influence of maternal psychopathology  on ratings of child psychiatric symptoms: An SEM analysis on cross-informant   agreement. European Child & Adolescent Psychiatry, 20, 241-252.   doi: 10.1007/s00787-011-0168-2  Muthen, L. K., & Muthen, B. O. (1998-2012). . Seventh Edition.   Los Angeles, CA: Muthen & Muthen.   Najman, J. M., Williams, G. M., Nikles, J., Spence, S., Bor,   M. J. (2000). -effect   association or observation bias?  Journal of the American Academy of Child and  Adolescent Psychiatry, 39(5), 592-602.    Neppl, T. K., Donnellan, M. B., Scaramella, L. V., Widaman, K. F., Spilman, S. K., Ontai, L. L.,   & Conger, R. D. (2010). Differential stability of temperament and personality from  toddlerhood to middle childhood. Journal of Research in Personality, 44, 386-396.  doi: 10.1016/j.jrp.2010.04.004  Patrick, C. J., Curtin, J. J., & Tellegen, A. (2002). Development and validation of a brief form   of the Multidimensional Personality Questionnaire. Psychological Assessment, 14,  150-163.   Phares, V. (1992). lative lack of attention to the role of fathers in child  and adolescent psychopathology. American Psychologist, 47(5), 656-664.   Phares, V., & Compras, B. E. (1992). The role of fathers in child and adolescent   psychopathology: Make room for daddy. Psychological Bulletin, 111(3), 387-412.  Phares, V., Fields, S., Kamboukos, D., & Lopez, E. (2005). Still looking for poppa. American   Psychologist, 60(7), 735-736. doi: 10.1037/0003-066X.60.7.735   Reber, A. S., Allen, R., & Reber, E. S. (2009). Penguin dictionary of psychology(4th ed.).  Strand, London: Penguin Books.   Richters, J. E. (1992). Depressed mothers as informants about their children: A critical review of  of the evidence for distortion. Psychological Bulletin, 112(3), 485-499.   Robins, R. W., Donnellan, M. B., Widaman, K. F., & Conger, R. D. (2010). Evaluating the link   between self-esteem and temperament in Mexican origin early adolescents. Journal of   Adolescence, 33, 403-410. doi: 10.1016/j.adolescence.2009.07.009      30  Rothbart, M. K. (2011a). Definition and historical roots. In Becoming who we are:  Temperament and personality in development (pp. 9-32). New York, NY: The   Guilford Press.  Rothbart, M. K. (2011b). The structure of temperament. In Becoming who we are:  Temperament and personality in development (pp. 33-61). New York, NY: The   Guilford Press.  Rothbart, M. K. (2011c). Temperament, environment, and psychopathology. In Becoming who  we are: Temperament and personality in development (pp. 224-238). New York, NY:  The Guilford Press.  Rothbart, M. K., Ahadi, S. A., Hershey, K. L., & Fisher, P. (2001). Investigations of   . Child  Development, 72(5), 1394-1408.   Rothbart, M. K., & Goldsmith, H. H. (1985). Three approaches to the study of infant   temperament. Developmental Review, 5, 237-260.   Sameroff, A. J., Seifer, & Elias, P. K. (1982). Sociocultural variability in infant temperament   ratings. Child Development, 53(1), 164-173.   Saudino, K. J. (2003). Parent ratings of infant temperament: Lessons from twin studies. Infant   Behavior and Development, 26, 100-107.   Saudino, K. J., & Eaton, W. O. (1991). Temperament and genetics: An objective twin study of  motor activity level. Child Development, 62(5), 1167-1174.   Saudino, K. J., McGuire, S., Reiss, D., Hetherington, E. M., & Plomin, R. (1995). Parent ratings   of EAS temperaments in twins, full siblings, half siblings, and step siblings. Journal of   Personality and Social Psychology, 68(4), 723-733.   Seifer, R. (2002).   . Infant   Behavior and Development, 25, 117-120.   Seifer, R., Sameroff, A., Dickstein, S., Schiller, M., & Hayden, L. C. (2004). Your own  children are special: Clues to the sources of reporting bias in temperament assessments.  Infant Behavior and Development, 27, 323-341. doi: 10.1016/j.infbeh.2003.12.005  Shiner, R. L., & DeYoung, C. G. (2013). The structure of temperament and personality traits:  A developmental perspective. In P. Zelazo (Ed.), Oxford handbook of developmental   psychology (pp. 113-141). New York: Oxford University Press    Spinath, F. M., & Angleitner, A. (1998). Contrast effects   questionnaire: A behavioral-genetic study on early developing personality traits    31   assessed through parental ratings. Personality and Individual Differences, 25, 947-963.  Tellegen, A., & Waller, N. G. (2008). Exploring personality through test construction:  development of the Multidimensional Personality Questionnaire. In The SAGE handbook of personality theory and assessment 2 (pp. 261-292). Thousand Oaks, CA: SAGE Publishing Inc.   Vaugh, B. E., Bradley, C. F., Joffe, L. S., Seifer, R., & Barglow, P. (1987). Maternal     on the Carey Infant Temperament Questionnaire. Developmental Psychology, 23(1),   152-161.    Vaughn, B. E., Taraldson, B. J., Crichton, L., & Egeland, B. (1981). The assessment of infant   temperament: A critique of the Care Infant Temperament Questionnaire. Infant Behavior  and Development, 4, 1-17.   Weissman, M. M., & Klerman, G. L. (1977). Sex differences and the epidemiology of   depression. Archives of General Psychiatry, 34, 98-111.   West, S. G., Taylor, A. B., & Wu, W.  (2012).  Model fit and model selection in structural   equation modeling.  In Handbook of Structural Equation Modeling (pp. 209-231).  New   York, NY: The Guilford Press.  White, C., & Barrowclough, C. (1998). Depressed and non-depressed mothers with problematic  preschoolers: Attributions for child behaviors. British Journal of Clinical Psychology,   37, 385-398.   Youngstrom, E., Izard, C., & Ackerman, B. (1999). Dysphoria-related bias in maternal ratings   of children. Journal of Consulting and Clinical Psychology, 67(6), 905-916.   Youngstrom, E., Loeber, R., & Stouthamer-Loeber, M. (2000). Patterns and correlates of  agreement between parent, teacher, and male adolescent ratings of externalizing and   internalizing problems. Journal of Consulting and Clinical Psychology, 68(6), 1038-  1050. doi: 10.1037//0022-006X.68.6.1038             32             APPENDICES      33               Figure 1. The Trifactor Model. M1 - M4 = -  CON = consensus factor; MP = maternal unique perspective factor; FP = paternal unique perspective factor; S1 - S4 = scale specific factor; MC = maternal characteristic; FC = paternal characteristic.       34                            Table 1 CBQ Means, Standard Deviations, and Mother-Father Scale Correlations    Mothers  Fathers     Mean SD  Mean SD  d r Activity Level 5.21 .62  5.18 .66  .04 .36 Smiling and Laughter 5.89 .50  5.55 .59  .50 .23 High Intensity Pleasure 5.13 .71  5.08 .73  .07 .53 Approach 5.27 .57  5.07 .56  .28 .22 Soothability 4.84 .68  4.70 .66  .18 .35 Impulsivity 4.81 .61  4.72 .60  .14 .41 Anger/Frustration 4.62 .70  4.62 .67  0 .37 Fear 3.74 .88  3.65 .77  .10 .37 Discomfort 4.11 .73  4.04 .64  .08 .27 Sadness 3.84 .58  3.85 .61  .01 .20 Shyness 3.33 1.15  3.39 .96  .06 .62 Attentional Focusing 4.40 .73  4.29 .66  .16 .51 Inhibitory Control 4.23 .72  4.20 .62  .05 .49 Low Intensity Pleasure 5.53 .62  5.22 .59  .42 .25 Perceptual Sensitivity  4.83 .78  4.67 .71  .18 .31  Note. SD = Standard deviation; d d for mean difference; r = correlation between mother and father scales     35                           Table 2 Parent Internalizing Symptomatology and Personality Means, Standard Deviations, and Mother-Father Correlations    Mothers  Fathers     Mean SD  Mean SD  d r General Distress 1.56 .52  1.55 .57  .01 .23 Anhedonic Depression 2.49 .61  2.45 .62  .05 .18 Anxious Arousal 1.12 .26  1.13 .33  .03 .39 Total Depression 2.02 .46  2.00 .48  .03 .28 Depression and Anxious Arousal 1.72 .35  1.71 .39  .02 .32 Negative Emotionality .24 .14  .25 .17  .05 .19 Constraint .76 .14  .65 .14  .61 .17 Positive Emotionality .64 .14  .66 .17  .10 .09  Note. SD = Standard deviation;  d d for mean difference; r = correlation between mother and father scales      36      Table 3 Unconditional Trifactor Model for Effortful Control  Consensus Factor  Unique Factor  Scale  Factor  Scale Intercepts Scale  STD   STD   STD   STD Attentional Focusing                 Mother .39 .55  .23 .33  .32 .46  4.43 6.26      Father .39 .58  .23 .38  .32 .48  4.43 6.56 Inhibitory Control                 Mother .39 .55  .31 .44  .25 .35  4.25 5.95      Father .39 .64  .08 .14  .25 .40  4.25 6.87 Low Intensity Pleasure                 Mother .19 .31  .41 .65  .23 .37  5.52 8.89      Father .19 .32  .41 .76  .23 .39  5.52 9.25 Perceptual Sensitivity                 Mother .28 .37  .28 .36  .33 .43  4.85 6.35      Father .28 .38  .28 .42  .33 .45  4.85 6.60 Note. Unique Factor = unique perspective factor; = un - standardized factor loading;  = unstandardized intercepts; STD = standardized parameter estimates. N = 222. df = 17. Father unique perspective factor mean and variance freely varied, all other factor means and variances set to 0 and 1, respectively.  Fit information: 2 = 16.90, p = .46; RMSEA = .00; SRMR = .07; CFI = 1.00; TLI = 1.00.     37      Table 4 Unconditional Trifactor Model for Negative Affectivity  Consensus Factor  Unique Factor  Scale  Factor  Scale Intercepts Scale  STD   STD   STD   STD Sadness                 Mother .13 .23  .42 .73  .12 .21  3.85 6.63      Father .13 .22  .42 .63  .12 .20  3.85 6.33 Anger                 Mother .41 .59  .32 .46  .00 .00  4.63 6.65      Father .41 .61  .32 .43  .00 .00  4.63 6.84 Discomfort                 Mother -.10 -.14  .50 .68  .25 .34  4.09 5.55      Father -.10 -.16  .50 .70  .25 .39  4.09 6.44 Fear                 Mother -.01 -.01  .47 .55  .45 .53  3.72 4.36      Father -.01 -.01  .47 .53  .45 .57  3.72 4.72 Soothability                 Mother .27 .40  .30 .45  .29 .43  3.16 4.68      Father .27 .41  .30 .41  .29 .44  3.33 5.03 Note.  Unique Factor = unique perspective factor; = un - standardized factor loading; N = 222. df = 33. Father unique perspective factor mean and variance freely varied, all other factor means and variances set to 0 and 1, respectively.  Fit information: 2 = 56.47, p = .01; RMSEA = .06; SRMR = .06; CFI = .95; TLI = .93.     38  Table 5 Unconditional Trifactor Model for Surgency  Consensus Factor  Unique Factor  Scale  Factor  Scale Intercepts Scale  STD   STD   STD   STD Impulsivity                 Mother .37 .62  .31 .51  .00 .00  4.82 7.98      Father .37 .61  .31 .65  .00 .00  4.82 7.95 High Intensity Pleasure                 Mother .27 .38  .30 .43  .43 .61  5.16 7.29      Father .27 .36  .30 .53  .43 .59  5.16 7.07 Smiling and Laughter                 Mother -.10 -.20  .34 .67  .16 .31  5.89 11.53      Father -.10 -.17  .34 .75  .16 .27  5.69 9.76 Shyness                 Mother .60 .53  .24 .22  .57 .51  4.69 4.19      Father .60 .61  .24 .31  .57 .58  4.69 4.77 Positive Anticipation                 Mother .02 .04  .33 .59  .21 .37  5.23 9.37      Father .02 .04  .33 .73  .21 .36  5.23 9.11 Note.  Unique Factor = unique perspective factor; = un - standardized factor loading; unstandardized intercepts; STD = standardized parameter estimates. Smiling and laughter and shyness scales loaded on the same scale factor. N = 222. df = 33. Father unique perspective factor mean and variance freely varied, all other factor means and variances set to 0 and 1, respectively.  Fit information: 2 = 74.99, p < .01; RMSEA = .08; SRMR = .09; CFI = .94; TLI = .92.     39  Table 6        Paths From Parent Characteristics to Consensus and Unique Perspective Factors Dimension/Factor General Distress Anhedonic Depression Anxious Arousal Total Distress Negative Emotionality Constraint Positive Emotionality Effortful Control             Consensus - Mother -.451 [-.62, -.29] -.151 [-.35, .05] -.211 [-.42, -.01] -.391 [-.57, -.21] -.131 [-.33, .07] .221 [.03, .42] .231 [.03, .43]      Consensus - Father -.182 [-.37, .01] -.131 [-.35, .08] -.091 [-.31, .13] -.141 [-.35, .06] -.141 [-.35, .06] -.152 [-.35, .06] .261 [.07, .46]      Mother Perspective -.103 [-.24, .05] -.083 [-.23, .07] -.043 [-.15, .07] -.073 [-.20, .06] -.303 [-.48, -.11] .393 [.18, .59] -.012 [-.12, .09]      Father Perspective .043 [-.19, .26] -.253 [-.44, -.06] -.043 [-.24, .16] -.143 [-.35, .07] -.103 [-.31, .11] .253 [.02, .47] .162 [-.05, .37] Negative Affectivity              Consensus - Mother .181 [-.02, .38] .131 [-.06, .32] .051 [-.17, .26] .181 [-.03, .38] .251 [.08, .42] -.161 [-.34, .02] .011 [-.17, .19]      Consensus - Father .101 [-.11, .30] .081 [-.14, .29] .131 [-.08, .35] .101 [-.11, .31] .131 [-.06, .32] -.031 [-.22, .17] -.071 [-.26, .13]      Mother Perspective .323 [.16, .48] .183 [.01, .34] .063 [-.12, .23] .293 [.13, .45] .263 [.10, .41] .083 [-.09, .25] -.153 [-.31, .01]      Father Perspective .253 [.07, .42] .373 [.10, .53] .143 [-.04, .31] .373 [.21, .53] .283 [.11, .44] -.023 [-.21, .16] -.193 [-.36, -.01] Surgency             Consensus - Mother .161 [-.01, .34] .131 [-.04, .30] .151 [-.03, .33] .211 [.04, .38] .111 [-.29, .05] -.281 [-.43, -.12] .101 [-.06, .27]      Consensus - Father -.011 [-.18, .17] .021 [-.16, .19] .071 [-.11, .25] .011 [-.17, .19] .151 [-.02, .32] -.071 [-.23, .10] -.131 [-.30, .04]      Mother Perspective -.143 [-.32, .04] -.053 [-.22, .13] -.113 [-.29, .06] -.133 [-.31, .05] -.123 [-.29, .05] .133 [-.04, .31] .223 [.05, .38]      Father Perspective -.013 [-.19, .16] -.193 [-.36, -.03] -.183 [-.35, -.02] -.173 [-.34, .00] -.073 [-.24, .11] .023 [-.17, .20] .173 [-.01, .35] Note. Consensus - Mother = path from maternal characteristic to consensus factor; Consensus - Father = path from paternal      40  )  characteristic to consensus factor; Mother Perspective = path from maternal characteristic to maternal unique perspective factor; Father Perspective = path from paternal characteristic to paternal unique perspective factor. Standardized regression coefficients presented. 95% confidence intervals appear in brackets underneath estimate. include zero. Corresponding parental effects tested for equality; corresponding paths that share superscripts equal, corresponding paths that do not share superscripts significantly different.