A Study on the Reliability and Validity of the Turkish Free Association Norms

Abstract

Pre-existing semantic associations of words influence performance in various episodic tasks (Nelson, Dyrdal, & Good-mon, 2005). The free association task indexes strong associative relations among words. In turn, this makes it possible to map associative structure of the words in semantic memory. Although Tekcan and Goz (2005) collected free association norms of Turkish words in an earlier study, the reliability and validity of their findings have not been empirically tested with a specific study. The goal of the present study was to determine the reliability and validity of the existing free association norms. In Study 1, 100 words were randomly selected from the existing normative database and were normed again by using the standard free association method as well as an online version of the same task in order to assess the reliability. The obtained findings were compared in regard to associative strength, associative set size, and heterogeneity values. The results indicated that the existing norms have high reliability. Study 2 assessed the predictive validity of the existing norms in an extra-list cued recall experiment. The cue-to-target strength and cue set size were manipulated in order to examine their effects on correct recall. In line with the previous findings, high cue-to-target strength led to higher recall rates compared to low cue-to-target strength; however, the effects of cue set size were not in the expected direction. The present findings imply that the existing free association norms have high reliability and validity overall, but there is a crucial need to overcome the shortcomings of the Turkish free association norms by norming additional words.