Methodological approaches reducing heterogeneity in reporting behavior in different groups of students
Abstract
In contemporary heterogeneous student population there is an increased risk that the data collected in questionnaire surveys may be inaccurate due to the differences in scale usage among different students or groups of students. For example, two students with the same level of motivation for learning can rate their motivation differently, one as high motivation and another one as low motivation.
Such data may lead researchers to inaccurate conclusions about the actual level of the measured concept and make comparison of different students or schools invalid. Researchers can consider some of the following methodological approaches for reduction of heterogeneity in reporting behavior with respect to the research goal and the data structure.
These are anchoring vignettes method, adjusting for tendency to choose certain scale categories regardless of item content (e.g., tendency to agree with items regardless of their content), overclaiming technique, forced choice assessment, and situational judgement tests. The aim of this paper is to present these (new) methodological approaches together with the examples of their application in educational research.