Predicting Adult Under- and Over-Reporting of Self-Reported Arrests From Discrepancies in Adolescent Self-Reports of Arrests

Abstract

Building on previous research, this article investigates whether discrepancies between official and self-reported measures of arrests as an adult can be predicted from such discrepancies as an adolescent. We use longitudinal data from the Rochester Youth Development Study to assess whether a pattern exists in adolescent and adult under-and over-reporting of arrests. We find consistency in under-and over-reporting throughout the adolescent-young adult life course. In other words, when respondents misreport the number of arrests they have experienced, they do so consistently regardless of age. This is reassuring for scholars using self-report data, as under-and overreporting behaviors remain stable over this span of the life course. Finally, our models predicting discrepancies in official and self-reported arrests during the combined period of adolescence and young adulthood are both extremely strong. Our findings support the continued use of self-report measures as a valid indicator of delinquency.

Predicting Consistency Between Officially Recorded and Self-Reported Records of Arrest

“…This result generally agreed with Krohn et al (2013) finding that a higher number of officially reported arrests was correlated to a higher error in self-reported arrests (due to a systematic relationship); however, Emmert et al (2015) noted that young adults tend to be more predictable even though under-and over-reporting of arrests exist on a continuum from adolescence to young adulthood.…”

“…The consistency between self-reported arrests and officially-recorded arrest data has been given some attention (Emmert, Carlock, Lizotte, & Krohn, 2015;Hindelang et al, 1981;Kirk, 2006;Krohn, Lizotte, Phillips, Thornberry, & Bell, 2013;Maxfield et al, 2000;Piquero et al, 2014;Thornberry & Krohn, 2003); however, those studies have primarily looked at the consistency and validity between self-reported arrest and officially-recorded arrests. Generally, the results have shown moderate to strong consistency between self-reported arrest and officially-recorded arrests with some differential validity across gender and/or race.…”

“…In their study of respondents from the Rochester Youth Development Study (RDYS), examining the same respondents as adolescents and again as young adults, Emmert et al (2015) found that, Bwhen respondents misreport the number of arrests they have experienced, they do so consistently regardless of age^(p. 13). This result generally agreed with Krohn et al (2013) finding that a higher number of officially reported arrests was correlated to a higher error in self-reported arrests (due to a systematic relationship); however, Emmert et al (2015) noted that young adults tend to be more predictable even though under-and over-reporting of arrests exist on a continuum from adolescence to young adulthood.…”

The School-To-Prison Pipeline: An Examination of the Association Between Suspension and Justice System Involvement

“…Finally, this study is limited by its use of self-report data to measure justice system involvement. Although self-reports of justice system involvement are generally accurate, responses may be less accurate for individuals with frequent justice system contact (Emmert, Carlock, Lizotte, & Krohn, 2017; Piquero, Schubert, & Brame, 2014).…”

The dark figure of sexual offending: new evidence from federal sex offenders

“…Both sources of data are valid proxies of an individual's antisocial history, however. Despite the potential differences in these sources of data, there nevertheless is evidence that they substantially overlap and the overlap is often moderate to strong (Emmert et al, 2015;Hindelang et al, 1979;Huizinga and Elliott, 1986;Maxfield et al, 2000;Piquero et al, 2014;Sutton et al, 2011;Thornberry and Krohn, 2000). For instance, a recent study of self-reported vs official recorded arrests using data from the National Youth Survey Family Study found that self-reported and official records of arrest were concordant for more than 80 percent of offenders (Pollock et al, 2015).…”