شواهدی که زیرگروههای پرسشنامه رفتار ضداجتماعی (STAB) گزارش لحظه ای از رفتارهای لوده بازی را پیش بینی می کند

Abstract There is growing recognition that substantively important distinctions exist across physically aggressive, rule-breaking, and socially aggressive forms of antisocial behavior. Even so, one limitation to accumulating additional scientific insights into the correlates and origins of these three varieties of antisocial behavior has been the lack of an efficient self-report assessment in the public domain. The Subtypes of the Antisocial Behavior Questionnaire (STAB) was developed to address this need. Although there is already a good deal of psychometric support for the STAB, prior research has yet to examine its “ecological” validity. In other words, it remains unclear whether the STAB scales would predict the frequency of acting-out behaviors in daily life. The current study sought to examine this question via an electronic diary study, in which participants reported on their momentary behaviors on multiple occasions in their natural environments. Analyses revealed that each STAB scale uniquely predicted only the momentary acting-out behaviors characteristic of that scale. Such findings provide further support for the STAB as a promising self-report measure of physically aggressive, rule-breaking, and socially aggressive forms of antisocial behavior.

Introduction There is converging evidence that physical aggression (e.g., assaulting others, bullying), rule-breaking (e.g., lying, stealing, vandalism), and social aggression (i.e., gossiping, ostracism, “stealing” friends) constitute meaningfully distinct, albeit overlapping, components of the broader construct of antisocial behavior. As reviewed in Burt and Donnellan (2009), these sub-types demonstrate distinctive developmental trajectories, have different demographic correlates and personological underpinnings, and evidence important etiological distinctions. For example, deficits in affective regulation are particularly characteristic of physical aggression, whereas impulsivity is more strongly associated with rule-breaking (Burt & Donnellan, 2008). Physical aggression is also more heritable than is rule-breaking (i.e., 65% versus 48%), whereas rule-breaking is influenced more by the shared environment (i.e., 5% versus 18%) (Burt, 2009). Nonetheless, more work is needed to firmly ground these constructs within the antisocial behavior literature. For example, longitudinal work examining differential outcomes for physical aggression, rule-breaking, and social aggression (e.g., conventional adult life versus prison) would more firmly cement these three sub-types of antisocial behavior within the literature. Such work has been difficult to conduct however, as researchers have historically needed to collect multiple or very long measures to assess all three forms of antisocial behavior. The 32-item Subtypes of the Antisocial Behavior questionnaire (STAB; Burt & Donnellan, 2009) was developed to address this need. The factor structure of the STAB was initially established in a sample of college students, and was then confirmed in a second sample of college students, a sample of community adults, and two samples of adjudicated adults. We also found consistent support for the criterion-related validity of the STAB scales when comparing them to related measures, as well as to frequently used conceptualizations of personality. Finally, the STAB demonstrated expected mean differences across the various sample types and across clinical treatment groups. In short, we have already marshaled a considerable amount of initial psychometric support for the STAB (Burt & Donnellan, 2009). Even so, establishing the validity of a measure is an ongoing process involving the accumulation of different kinds of evidence (Simms, 2008). One potential avenue for such research is to evaluate how well the subscales of the STAB predict momentary reports of behavior in the context of an experience sampling (or ESM) study (Conner et al., 2007 and Reis and Gable, 2000). In this approach, research participants carry an electronic device for a discrete period of time during which it periodically “beeps” and prompts them to report on their thoughts, feelings, and/or behaviors in that moment. However, this approach is rarely used to validate measures of antisocial behavior despite the clear advantages of ESM studies for providing reports that are acquired “in vivo (in life) and in situ (in place)” ( Conner et al., 2007, p. 82). Proponents of these approaches (e.g., Conner et al., 2007 and Trull and Ebner-Priemer, 2009) argue that such studies can provide important insights into constructs like antisocial actions which are likely to be contingent on the behaviors of others and to be expressed in a social context. Likewise, ESM reports are thought to better approximate actual thoughts, feelings, and behaviors because they do not rely on retrospective recall ( Conner et al., 2007). In light of these virtues, we used an ESM study to provide additional data on the validity of the STAB. In particular, we tested whether the STAB subscales predicted momentary reports of physical aggression, social aggression, and rule-breaking, respectively.

Results STAB scale means are reported in Table 2, along with the means for the momentary reports of specific acting-out behaviors/feelings. Means for the STAB correspond closely to those of other normative college samples (Burt & Donnellan, 2009). The STAB scales were at least moderately intercorrelated with one another (all ps < .01). Specifically, STAB rule-breaking was correlated .39 and .48 with social and physical aggression, respectively. STAB social and physical aggression were correlated .44. Table 2. Descriptives. Variable Mean (SD) Minimum Maximum STAB–AGG 21.5 (6.4) 11 44 STAB–RB 15.0 (4.6) 11 38 STAB–SA 25.8 (4.9) 14 39 Physically aggressive feelings .04 (.08) .00 .50 Rule-breaking behaviors .10 (.10) .00 .50 Socially aggressive behaviors .08 (.13) .00 .61 Table options The frequency of specific acting-out behaviors/feelings ranged considerably across the sample, with some participants reporting no such behaviors, whereas others endorsed them in approximately half (i.e., 50–61%) of their electronic diary responses. Despite this broad range, the average frequency of endorsed acting-out behaviors was (not surprisingly) rather low. Physically aggressive feelings were endorsed in an average of 4% of electronic diary interviews, whereas rule-breaking and socially aggressive behaviors were endorsed in an average of 10% and 8% of interviews. Similar to the STAB scales, the acting-out behaviors were intercorrelated. Rule-breaking behaviors over the course of the week were correlated .26 and .35 with socially aggressive and physically aggressive behaviors/feelings (both ps < .05). Socially and physically aggressive behaviors were correlated .42 (p < .01). Associations with the STAB scales are reported in Table 3. Physically aggressive feelings had significant zero-order associations with the AGG and RB scales, but not with the SA scale. Rule-breaking behaviors evidenced zero-order associations with all three STAB scales. Socially aggressive behaviors had significant zero-order associations with the AGG and SA scales. Importantly, however, a series of regression equations (with the specific behaviors as the outcome variables) revealed that, once we controlled for overlap across the STAB scales, each scale uniquely predicted only the momentary behaviors/feelings characteristic of that particular scale (see beta weights in Table 3). Such findings highlight both the convergent and discriminant validity of the STAB scales when predicting the frequency of specific forms of acting-out behaviors. Table 3. Associations between the STAB scales and the frequency of specific acting-out behaviors/feelings over the course of a week. Full-size image (20 K) Note: “r” indexes the zero-order correlations between the STAB scales and the specific acting-out behaviors/feelings. “Beta” represents the unstandardized and standardized beta weights (before and after the slash, respectively) from the three regression equations (in which each behavior was regressed onto the three STAB scales). The percent of variance accounted for by each regression equation is also indicated (i.e., R2). Concordant behavior-STAB associations (e.g., between social aggressive behaviors and the STAB social aggression scale) are highlighted with gray shading. ** and * indicate estimate is significantly greater than zero at p < .01 and p < .05, respectively. Table options As a second way of evaluating these associations, we compared the predicted values of specific acting-out behaviors for individuals high versus low on the corresponding STAB scale using the following unstandardized regression equations (based on log-transformed variables): equation(1) View the MathML sourcePhysically aggressive feelings=-.99+.63(AGG)+.40(RB)-.51(SA). Turn MathJax on equation(2) View the MathML sourceRule breaking behaviors=-2.25-.03(AGG)+.57(RB)+.43(SA). Turn MathJax on equation(3) View the MathML sourceSocially aggressive behaviors=-1.97+.23(AGG)-.07(RB)+.58(SA). Turn MathJax on Specifically, we used these equations to compute d-metric effect sizes for individuals uniquely high (versus) low on particular STAB scales. For example, we first used Eq. (1) to predict physically aggressive feelings for individuals with high scores on the AGG scale (i.e. 1 SD above the mean) but average scores on the RB and SA scales. The predicted value for this case was .55. We then used Eq. (1) to predict physically aggressive feelings for individuals low on AGG (i.e., 1 SD below the mean) and average on the other two scales. The predicted value for this case was .19. The difference between these two predicted values (.36) was divided by the (log-transformed) standard deviation for physically aggressive feelings (.68) to yield an effect size in the d-metric. This standardized difference of .53 was equivalent to a medium effect size ( Cohen, 1988). We then carried out parallel computations for rule-breaking and social aggression. The Cohen’s d-metric effect size for the rule-breaking comparison was .70 whereas the effect size for the social aggression comparison was .47, effects that were at least medium in magnitude by current conventions. Highly similar effect sizes (i.e., .46–.79) were also obtained when we simply dichotomized each STAB scale and compared differences in the frequencies of their corresponding acting-out behaviors/feelings. In short, the STAB scales predicted non-trivial differences in their associated feelings and behaviors in the context of daily life.