An Inquiry into Pragmatic Data Collection Methods

While the significance of validation of data collection instruments in speech act research has been recognized and has attracted considerable interest, most validation studies employed a between-subjects design. In so doing, it is possible that the differences were caused by the group effects rather than different data collection techniques. This study attempts to resolve the above-mentioned problems by employing a within-subject design and comparing refusal responses obtained from the written DCT [discourse completion test/task], closed role play (oral DCT), and open role play in terms of response length, frequency of semantic formulas and the combination pattern of the semantic formulas. Sixty female college students participated in the study. It was found that the three methods elicit similar samples in terms of response length, mean number of semantic formulas, frequency of semantic formulas and the combination pattern of the semantic formula. Two refusal situations are appended. (Contains 12 tables.)