Challenges and the Way Forward: Global issues related to the measurement of socioemotional skills among marginalized communities.

• In Event: Exploring Assessment Practices and Educational Equity in Diverse Contexts

Proposal

There is converging evidence that adolescents’ socioemotional (SE) skills, or the beliefs, attitudes and behaviors that allow youth to manage self and social relationships, predict academic success and long-term life outcomes (e.g., Almlund, Duckworth, Heckman, & Kautz, 2011; Paunesku et al, 2015). However, this evidence is drawn primarily from Western societies (Puerta, Valiero, & Bernal, 2016), and measures of SE skills have been validated largely for European and U.S. contexts (Jones et al. 2019). The purpose of this roundtable is to initiate a discussion on the challenges regarding the measurement of SE skills cross culturally and the solutions to some of them.
This builds on the research that conducted psychometric testing of measures of several prominent SE skills among highly marginalized rural adolescent Indian girls who are Dalits and Tribals. This was crucial to understand the consequence of growing up in marginalized groups with limited opportunities and the developmental outcomes, as a result.
Support was received from the Indian government to collect data from 663 Dalit and Tribal girls studying in Grades 6 to 9 residing in 8 “KGBV” schools in Haryana, India. KGBVs are special government schools established for the most ‘underprivileged’ sections of the country. Most participants were first-generation learners. During school hours, participants completed a 37-item questionnaire that was double-back translated from English to Hindi. For limited literacy participants, a trained member of the research team assisted in completing the questionnaire.
In addition to administrative records of student grades, data was collected on 3 SE skills using adapted existing measures. All items are arrayed on a 1(very inaccurate) - 5(very accurate) Likert-type scale. The measures were:
(1) Self-esteem, or positive self-evaluations of the student's own worth (Harris, Donnellan, & Trzesniewski, 2017);
(2) Sense of belonging, or student’s feeling that they are accepted members of the school community (Good, Rattan, & Dweck, 2012);
(3) Perseverance, or the degree to which students keep up effort to achieve their goals despite difficulty (Duckworth & Quinn, 2009);
The psychometrics using latent variable techniques was assessed by examining the factor structure and reliability for each construct. Then the validity was tested through correlations across constructs and predictive relations between constructs and student grades. Results provided evidence that each construct could be modeled using a single latent factor; however, reliability for grit was low. Additionally, two items did not work in the given context and hence were dropped, even though these were standardized measures. Construct bias and method bias were also faced by the research team during the study. In response to the association with the school grades, grit and sense of belonging were associated with higher school-reported grades.
This research builds on the experiences of conducting assessments in low literacy environments and analyzing the psychometric results. This also ties to the theme of ‘Protests in education, by challenging the majoritarian approach to evaluation and design, where the educational and social justice programs are often designed, devoid of acknowledgment of the context and culture; and lays the foundation for culturally sensitive design process.

Authors

• Vaibhav Kumar, SwaTaleem Foundation

• Ananya Tiwari, Texas A&M University