Abstract
Theories of educational and psychological measurements have been heavily, even too much, relied on statistical theories as a model for describing the human abilities and behavioral traits. Perspectives of the current researches as well as the historical roots of the present thoughts in mental testing are briefly sketched out for the readers in adjacent fields. The list of section headings is as follows:
1. The root of educational measurements
2. Problems of measurement scale
3. On absolute scaling
4. Ipsative score
5. Criterion-referenced measurement and item sampling
6. Sequential testing procedures and tailored testing
7. Errors of measurement and the reliability problems
8. The latent trait models
9. Other Problems
Multivariate treatments including factor and regression analyses are excluded here, as they are too much to be covered in a short article like this.
