U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

Measuring Educational Outcomes for At-Risk Children and Youth: Issues With the Validity of Self-Reported Data

NCJ Number
249659
Journal
Child & Youth Care Forum Volume: 44 Issue: 6 Dated: December 2015 Pages: 853-873
Author(s)
A. C. Teye; L. Peaslee
Date Published
December 2015
Length
21 pages
Annotation
This study examined the validity of academic and attendance self-reporting among children and youth.
Abstract
Youth programs often rely on self-reported data without clear evidence as to the accuracy of these reports. Although the validity of self-reporting has been confirmed among some high school and college age students, one area that is absent from extant literature is a serious investigation among younger children. Moreover, there is theoretical evidence suggesting limited generalizability in extending findings on older students to younger populations. The findings of the current study indicate that children and youth reported their individual grades and attendance poorly. Younger and lower performing children were more likely to report falsely; however, there is some evidence that a mean construct measure of major subjects GPA is a slightly more valid indicator of academic achievement. Based on these findings, the authors advise researchers and practitioners to exercise caution in using self-reported grades and attendance indicators from young and low-performing students. The study used original data collected from 288 children and youth using Big Brothers Big Sisters enrollment and assessment data, paired with school-records from two local school divisions. Initially, the study used percent agreement, validity coefficients, and average measures ICC scores to assess the response validity of self-reported academic and attendance measures. It then estimated the effects of several moderating factors on reporting agreement (using standardized difference scores). The study also accounted for cross-informant associations with child reported GPA using a moderated multiple regression model. (Publisher abstract modified)