Validity and reliability study of the attitude scale towards second foreign language learning

Abstract

In this study it is aimed to develop a measurement tool to determine the attitudes of foreign language teacher candidates towards a second foreign language. The item pool was prepared by literature review and teacher candidates’ views and finalized by taking expert opinions. The 61 item preliminary form was applied to 504 teacher candidates studying in different grades of Foreign Language Teaching Departments of Gazi University Gazi Faculty of Education in 2018-2019 academic year Spring semester. Analyses towards determining the psychometric features of the scale were made on the data set obtained from the application. For construct validity exploratory factor analysis and afterwards confirmatory factor analysis was made. The data set was applied first Exploratory Factor Analysis (EFA) for construct validity and then Confirmatory Factor Analysis (CFA). The exploratory and confirmatory analyses were made on two groups by dividing the total student number into two. EFA was conducted on the first group (n1=252) whereas CFA was conducted on the second group (n2=252). At the end of exploratory factor analysis, the scale was observed to consist of 43 items and 3 sub-dimensions. The sub-dimensions were named according to the judgments the scale items reflected as “positive belief towards learning a second foreign language”, “resistance to learning a second foreign language” and “attaching importance to learning a second foreign language”. The three factor structure about the scale explains the 49.72% of the total variance. At the end of the reliability analysis Cronbach alpha reliability coefficient for the overall scale is .95, for sub-dimensions respectively as .95 and .90 and .76. The model fit indices for the scale at the end of the confirmatory factor analysis are determined as RMSEA, .076; I‡2/df=2.4; SRMR=.81; NFI=.94; NNFI=.96; IFI=96; CFI=.96; RFI=94. Values obtained from confirmatory factor analysis are in acceptable reference range and shows that 3 factor scale construct is validated.