This research chapter examines the appropriate selection of measurement scales for academic research, specifically focusing on interval and ratio scales in dementia screening and quality of life studies. The analysis evaluates colleague responses regarding measurement scale choices and provides critical assessment of their methodological decisions. The discussion highlights the importance of understanding scale properties, including the presence or absence of true zero values, when designing quantitative research studies.
My first colleague responds by indicating that the interval scale would be the most ideal level of measurement in as far as dementia screening and quality of life are concerned. In some instances, it could be difficult to differentiate between interval scale and ratio scale. In the former instance, however, we do not have a true zero (Boslaugh, 2012). In his presentation, my colleague argues that change in dementia screening as well as quality of life within certain percentage points would be numerical. This is the basis used to reinforce the choice of level of measurement. I am in agreement. This is more so the case given that as Lavrakas (2008) indicates, “to identify whether a scale is interval or ordinal, consider whether it uses values with fixed measurement units, where the distances between any two points are of known size” (312).
Udenze indicates that the ratio scale will be used as a level of measurement for some of the variables highlighted. It would be prudent to note that the level of measurement for the quality of life, self-care, depression, and psychosocial intervention highlighted by Udenze may not necessarily align with the requirements of a ratio scale. This is more so the case given that it happens to be quantitative in nature (Indeed Editorial Team, 2021). For this reason, I am of the opinion that it would be more ideal in contexts that permit my colleague to come up with a comparison of differences or intervals. The ideal level of measurement on this front would perhaps be nominal scale. This is particularly the case given that as Illowsky & Dean (2017) observe, the expression of data in words is largely possible on this front. The kind of data my colleague would derive, given the choice of variables, will not have a true zero value or rank order.
You’re 90% through this paper. Sign up to read the full paper.
Sign Up Now — Instant Access Already a member? Log inAlways verify citation format against your institution’s current style guide requirements.