earticle

영어

Song, Sanghoun and Oh, Eunjeong. 2016. 5 or 7: Is the Choice So Important in Acceptability Judgment Testing? Korean Journal of Linguistics, 41-3, 449-480. This article concerns whether two different numbers of points on a Likert scale task produces different results in acceptability judgment testing. The most popular numbers of points on response scales are 5 and 7, and there seems to be no clear consensus about which of the two is better and why. As the same goes for experimental syntax studies, the choice of the numbers of points on the scales still remains questionable, though the Likert scale task has been widely employed in acceptability judgment testing. The present study compares two experimental data sets using the same stimuli sentences but with different point scales (5 and 7). It includes 46,356 data points and 506 Korean native speakers participated in the study. The comparison between the 5- and 7-point scale data is made in terms of (a) variance of data points, (b) convergence between the linguists’ judgments and the participants’ judgments, and (c) response time. The comparative analysis reveals that the two different point scales do not yield significantly different results. Yet, it is also observed that there exist pros and cons to both sides. The 7-point scale is more demanding than the 5-point scale to the extent that the middle point is rather scarcely used. On the other hand, the 5-point scale is sloppier than the 7-point scale to the extent that the latter captures the variation in acceptability judgments across the participants slightly better. (Incheon National University & Sangmyung University)