A word on standardization in longitudinal studies: don't

Frontiers in Psychology

Julia Moeller¹ 

关键词: standardization;    ipsatization;    longitudinal data;    profiles;    experience sampling method (ESM);   

DOI  :  10.3389/fpsyg.2015.01389

学科分类：心理学（综合）

来源: Frontiers

【 摘 要 】

Z-standardization and ipsatization are procedures to transform absolute values, or ratings (e.g., 1 = don't agree at all to 7 = totally agree) to relative scores that reflect each answer's rank in comparison to the ranks of all responses in that sample. In z-standardization, the sample mean score is subtracted from each single observation, and this difference is then divided by the sample's standard deviation. The result is a scale where a score of 0 means that this observation was at the sample's mean level, and a z-score of 1 reflects an observation one standard deviation above the sample mean. Ipsatization also converts absolute ratings into relative ranks, but relates each answer to the individual's own mean, not the sample mean. An individual's responses are ipsatized by subtracting the individual's mean score from each response the individual gave in a questionnaire. A positive ipsatized score means that the individual rated this item higher (which often means: affirmed more strongly), than the average of other items in that questionnaire.

【 授权许可】

CC BY