evaluation of construct validity and reliability of language learning styles instruments

educators have been preoccupied with individual differences in learning for quite a long time. one way of addressing these differences has been a focus on learning styles. understanding students’ learning styles is important for the quality of instructional design. a major problem in measuring the learning styles of learners of english as a foreign language (efl) is the development of a valid and reliable instrument. this paper reports the results of administering two learning styles questionnaires, namely reid’s (1987) perceptual learning style preference questionnaire ( plspq) and wintergerst and de capua’s (2002) learning styles indicator (lsi) originally developed for the esl learners. over 1000 iranian bs and ba students took the questionnaires. the results reveal that the plspq is weak in terms of construct validity and reliability while the lsi enjoys acceptable construct validity and reliability. since the lsi has fewer items than the plspq, it tends to be more practical. all in all, it can be claimed that the lsi will better assess the learning styles of the iranian students of english. most readers readily agree that there are few questionnaires that entail all aspects of a given construct. in the lsi, for example, the mode of reasoning (inductive vs. deductive) is missing. hence, some items geared towards this aspect of learning are proposed to be added to the lsi to render it more valid in terms of content (content validity).