Studying the Transitivity of Motivation Ratings

Abstract

Measurement in social sciences implies that the measured feature is quantitative, or in other words that it is possible not only to arrange the values of any given attribute, but also to express the difference between ordered magnitudes using a certain unit of measurement. However the need to verify this basic assumption is often ignored. And though there are a few possible excuses for this, but fundamentally this neglect distracts the social sciences from its main task of exploring reality. In this work, one of the requirements for the ordinal structure of motives was checked, namely the requirement of transitivity: if a > b and b > c, then a > c. If transitivity is not observed, then motives cannot be evaluated even on an ordinal scale (“more – less”, “stronger – weaker”), not to mention their quantitative measurement, which all methods that use Likert scales are supposedly tailored to. On a sample of 250 students, it was shown that about half of the respondents established transitivity when arranging their motives (internal, external and social ones), which justifies the use of ordinal scales for motivation assessment, at least for these motives and for two values: “more” and “less”; however, even in these cases, further validation of the assumptions about additivity when it comes to measuring motives is required to justify the use of Likert scales. The other part of the respondents (about 40%) could neither distinguish nor arrange their motives, therefore not only measuring, but even defining the order of their motives in these cases is impossible. It is concluded that the transitivity error is associated with the individual characteristics of the respondents and requires further study as a systematic error.