Assessing to Select the Best Students Holds Back Low-Social-Class Students

• In Event: 52.032 - The Social Class Achievement Gap: Interventions to Promote Performance for First-Generation College Students

Abstract

Objectives
International surveys show that students’ socio-economic status (SES) still strongly predicts their scholastic outcomes. Lower social class students underperform, are more likely to repeat grades, to drop out, and attain a lower level of education (OECD, 2010, 2013). There is growing research in social psychology on the impact of social class on educational attainment but most focuses on the psychological consequences for students. We propose to investigate structural obstacles by focusing on the use by evaluators of selection practices and more specifically, assessment practices.

Theoretical framework
The usual assessment tool used in educational systems is the normative assessment based on grades. It is usually perceived as an objective measure of students’ qualities and thus a reliable basis for a merit-based selection. We propose that ironically the use of assessment practices for selective purposes lead to the production of performance differences that correspond to the social hierarchy, thus holding back students from lower social class.

Methods and data
In two experimental studies (Ns = 203 and 259) participants were asked to assess a dictation test. The test was supposedly produced by a low or a high-SES pupil, which was introduced by information about the parental occupation and hobbies of the pupil. Participants had to assess the test using either a normative method (i.e., grades) known to emphasize on ranking and selective purposes, or a formative method (i.e., comments) known to focus on improving learning.

Results
Results showed that participants using a normative assessment found more mistakes in the test of a low-SES pupil than in the test of a high-SES pupil. The production of this social-class performance gap was not observed for participants who used a formative assessment that is less focused on selective purposes. In the second study, we more directly manipulated the selective function of assessment. Participants were told that their evaluation of the test would help them to either decide whether the child should move to the next grade (i.e., selective purpose) or provide improvement strategies (i.e., educational purpose). The results showed that the selective purpose of assessment drives the creation of a social-class performance gap.

Significance
These results show that low social class students’ success is restrained by assessment practices that emphasize the need to select the best students. This research suggest that the ideal of a meritocratic selection based on objective measure of individuals’ qualities is not only a fallacy but actually participate in the reproduction of social class inequalities. This work contribute to the growing literature in social psychology that investigates how the functioning of the educational system in itself contribute to maintaining social class inequalities by pointing out structural barriers (i.e., selection practices) and by taking the perspective of the agents of the educational system.

Authors

• Frédérique Autin, Universite de Poitiers

• Anatolia Batruch, University of Lausanne

• Fabrizio Butera, University of Lausanne