The Psychometric Properties of a Student Survey of Interpersonal and Intrapersonal Competencies

• In Event: Excellence in Education Research: Early-Career Scholars and Their Work

Abstract

This study uses data from the Study of Deeper Learning: Opportunities and Outcomes (Huberman, Bitter, Anthony, O’Day, 2014). In 2011–12, the Hewlett Foundation selected ten school networks within six school districts in California and New York to participate in what was known as a “Deeper Learning Community of Practice.” A set of 20 network schools were sampled from these ten networks for the purposes of the study. Additionally, a comparison (non-network) school was recruited for each network school. For a detailed description of the sampling process, see Huberman, Bitter, Rickles, Brown, Anthony, O’Day (2014).

This study primarily uses data from a student survey that was administered in the spring of 2013 to all sampled and consented students in Grade 11 and Grade 12. Scales were developed based on a combination of existing measures and items written for this study. Before the administration in 2013, the survey was piloted in six network schools in spring 2012. This analysis focuses on a set of 52 survey items that were designed to measure one interpersonal competency and intrapersonal competencies. Additionally, high school graduation status was obtained from the participating school districts in fall 2013.

In this study, the psychometric properties and predictive validity of the intrapersonal and interpersonal competencies were examined. Given the theoretical and interrelated nature of the intrapersonal and interpersonal competencies, I first empirically tested the factor structure of the seven competencies to assess whether these competencies are operating as unique factors or represent a smaller number of broad competencies. After the dimensionality of the competencies was established, I examined the reliability of the measures. The third step examined whether the measures of intrapersonal and interpersonal skills measure the constructs similarly among different student subgroups. A series of differential item functioning (DIF) studies were conducted to see if there is evidence that an item performs differently for members of different groups (e.g., low achieving vs. high achieving students, students in the network and non-network schools) who are matched on the skill being measured. Lastly, I examined the association between the domains from the exploratory models and on-time high school graduation and college enrollment.

Author

• Megan Kuhfeld, NWEA