Do Data Characteristics Change According to the Number of Scale Points Used? An Experiment Using 5 Point, 7 Point and 10 Point Scales

TLDR: In this article, the authors investigated the impact of scale format on data characteristics such as the mean, coefficient of variation, skewness, and kurtosis on customer satisfaction.

Abstract: To what extent does the number of response categories in a Likert-type scale influence the resultant data? Surprisingly little attention has been paid to the issue of whether the response category format has any influence on data characteristics such as the mean, coefficient of variation, skewness and kurtosis. This issue is important for several reasons. The first is that decisions are made based on outcomes such as the mean score. For example, marketing organizations and research providers use Likert type scales to measure constructs such as customer satisfaction. In this situation a higher score is better. Could the score have been comparatively better if a different scale format had been used? There is an absence of evidence on this issue. The second reason is that scale formats that are used in on-going market research projects such as tracking studies occasionally change. Can the old results be re-scaled or transformed to be comparable to data from a new scale format? Again, little is known about this. The third reason concerns data characteristics such as variation about the mean, skewness and kurtosis. Analysis tools such as regression are often used on data of this type to explain the variation in certain variables. If there is little variance in the data, this is harder to do. How does scale format affect these characteristics? The answers would be useful to both market researchers as well as academics. A literature review found that little work has been done on this issue. Therefore, this study set out to investigate the impact of scale format on data characteristics. It examined how using Likert-type scales with varying numbers of response categories affects the resultant data in terms of mean scores, and measures of dispersion and shape. Three groups of respondents were administered a series of eight questions (group n’s = 300, 250,185). Respondents were randomly selected members of the general public. A different scale format was administered to each group – either a five-point, seven-point or ten-point scale. The surveys were conducted by a professional market research organization via telephone interview. Data characteristics of mean score, standard deviation, skewness and kurtosis were analyzed according to scale format. The five and seven-point scales were re-scaled to a comparable mean score out of ten. The study found that the five and seven-point scales produced the same mean score as each other, once they were re-scaled. However the ten-point format tended to produce slightly lower relative means than either the 5 or 7-point scales (after the latter were re-scaled). The overall mean score of the eight questions was 0.3 scale points lower for the 10-point format compared to the 5 and 7-point format. This difference was statistically significant at p=0.04. In terms of the other data characteristics, there was very little difference among the scale formats in terms of variation about the mean, skewness or kurtosis. Therefore each of the three formats appears comparable for the type of research project in which multiple-item scales are analyzed with multivariate statistical methods. This study is also ‘good news’ for research departments or agencies who ponder whether changing scale format will destroy the comparability of historical data. Five and seven-point scales can easily be re-scaled with the resultant data being quite comparable. In the case of comparing five or seven-point data to 10-point data, a straightforward re-scaling and arithmetic adjustment easily facilitates the comparison. Finally, it appears that indicators of customer sentiment – such as satisfaction surveys – may be partially dependent on the choice of scale format. A five or seven-point scale is likely to produce slightly higher mean scores relative to the highest possible attainable score, compared to that produced from a ten-point scale.