The Impact of Cognitive and Noncognitive Factors on EFL Achievement in Europe

• In Event: Conceptualizing and Assessing English as a Foreign Language: The European Perspective

Abstract

Student characteristics need to be considered when investigating academic achievement and evaluating educational effectiveness (Creemers & Kyriakides, 2008). Recently, attention to non-cognitive skills (e.g. personality, self-concept) has increased as they predict achievement beyond cognitive abilities. Investigating how cognitive and non-cognitive skills are related to English as a foreign language (EFL) achievement can help improve learning outcomes. This is especially important considering the high relevance of English language competencies in a globalized world. Most European educational systems use different measures to evaluate EFL achievement at the end of upper secondary school: grades, standardized tests and final examinations. These measures are associated with different aspects of EFL performance: commonly, cognitive abilities predict standardized test scores, whereas non-cognitive skills predict grades (Lechner et al., 2017). Domain-specific differences can be obtained in mathematics and languages (e.g. Spengler et al., 2013). However, there are only few studies (e.g. Spinath et al., 2006) focusing on EFL and domain-specific interest in this context. Further, as far as we know, no previous research has examined final exams as an additional outcome next to grades and test scores.
In this study we investigated the predictive value of cognitive abilities and interest on three EFL achievement measures. By showing how results vary with the particular achievement measure we aimed to gain a more detailed understanding of foreign language learning and the importance of cognitive and non-cognitive skills in the European context.

Methods
Our sample consisted of N=1317 students in the academic track of upper secondary school (13th grade, 57.3% female, M=19.86 years) in Germany (Schleswig-Holstein). EFL grades and final exam results were collected via school administration lists. To measure test performance we used the ITP TOEFL (ETS, 2015). Cognitive abilities were assessed with the Cognitive Abilities Test (Heller & Perleth, 2000). Interest in EFL was measured with five items adapted from Trautwein et al. (2012). Reliability was sufficient (α=.79). To obtain reliable proficiency scores for EFL and cognitive abilities, we computed five plausible values for each student and combined them following Rubin (1987).
Following Lechner et al. (2017) we decomposed the variance of the achievement measures by regressing them on interest and cognitive abilities both separately and jointly. We applied structural equation modeling using Mplus.

Results and discussion
The findings were in line with our expectations: Both cognitive ability and interest explain a significant amount of variance in EFL achievement. Differential results can be obtained for achievement measures: interest explains more variance in grades than cognitive abilities. In test scores cognitive ability explains more variance than interest. In final exams cognitive ability and interest explain a similar amount of variance.
We conclude that both cognitive and non-cognitive skills are important in foreign language learning. This indicates the value of fostering EFL interest and creating stimulating learning environments, especially in light of a potentially greater malleability of non-cognitive skills compared to cognitive ability (e.g. Hulleman & Harackiewicz, 2009). Further, results show the importance of considering different aspects of achievement and corresponding measures when analyzing EFL achievement in Europe.

Authors

• Jennifer Meyer, Leibniz Institute for Science and Mathematics Education

• Johanna Fleckenstein, Pädagogische Hochschule FHNW

• Jan Retelsdorf, University of Hamburg

• Olaf Koeller, Leibniz Institute for Science and Math Education