Adjusting the Consensus Measure to Target Ordinal Scale Arguments
暂无分享,去创建一个
Much research data is collected and analyzed as ordinal scale data. The Likert scale is an example of such a data collection tool. Comparing different Likert scale data sets can be difficult, for no specific tool has been identified that is conceptually sound, though researchers typically apply ratio scale mathematics to the data. In this paper a modification of the consensus measure is examined and shown to be a valuable tool in the comparison and ranking of ordinal data. By identifying a specific target category, i.e., the "strongly agree" category of a Likert scale, the target consensus measure calculates a value between 0 and 1, the 0 identifying a Likert scale frequency distribution as being the furthest distance possible from the targeted value, and the 1 representing all values being assigned to the targeted category
[1] Leonard R. Sussman,et al. Nominal, Ordinal, Interval, and Ratio Typologies are Misleading , 1993 .
[2] Robert K. Leik,et al. A Measure of Ordinal Consensus , 1966 .
[3] Godfrey Pell,et al. Use and misuse of Likert scales , 2005, Medical education.
[4] S S Stevens,et al. On the Theory of Scales of Measurement. , 1946, Science.
[5] M. J. Wierman,et al. Consensus and dissention: theory and properties , 2005, NAFIPS 2005.