Reasons to avoid z-scores in graphs displaying profile or group differences
This article summarizes reasons why z-standardized scores should be avoided in graphs that display profiles or mean score differences between groups. It provides examples showing why and how the use of z-scores in such group comparisons can be misleading and concludes with proposing alternative methods that avoid such risks of misinterpretation.The reasons why z-standardized scores should be avoided when displaying differences between groups of profiles are: (1) The ratio of the difference between two groups is distorted in z-scores.(2) The ratio of the difference between two variables is distorted in z-scores.(3) Information about item endorsement and item rejection is lost.(4) The psychological meaning a given z-score cannot be compared across samples and variables.(5) People may end up in a group they do not belong to if z-scores are used to assign individuals to groups.(6) The group size and group frequency may be affected if z-scores instead of raw scores are used to assign individuals to groups.(7) Group differences in further outcome variables can change if z-scores instead of raw scores are used to assign individuals to groups.(8) Alternative transformations perform better than z-standardization in cluster analyses.Alternatives to using z-scores in graphs displaying profiles and group differences are using raw scores or using scale transformations that do not lead to the aforementioned problems, such as the POMS transformation. The latter can be combined with the formula used to calculate the item difficulty to relate to larger audiences, or with outlier removal techniques to make it less susceptible to the influence of outliers.