×

Probabilistic model of school-achievement test with triple choice response. (English) Zbl 0844.62092

Summary: In connection with the analysis of achievement tests with triple choice response it is possible to construct various mathematical models describing the probability structure of such tests. The present variant proceeds from the assumption of the existence of four possible levels of the real familiarity with the one question tested topic. The proportion of the whole topic to be examined which the tested person really does not know then can be expressed in the form of a weighted sum \(\tau = \nu + \delta/3 + 2\omega/3\), where \(\omega\) is the proportion of “one-third-familiarity”, \(\delta\) is the proportion of “two-thirds-familiarity”, and \(\nu\) is the proportion of “total unfamiliarity” with the tested topic.

MSC:

62P15 Applications of statistics to psychology

Citations:

Zbl 0844.62093
PDFBibTeX XMLCite
Full Text: EuDML

References:

[1] Tesaříková E.: Probabilistic model of school-achievement test with double-choice response - variant I. Acta UP Olomucensis, Fac. rer. nat., 105 (1992), 119-126. · Zbl 0825.62747
[2] Tesaříková E.: Statistical analysis of probabilistic model - variant I of the school-achievement test with double choice response. Acta UP Olomucensis, Fac. rer. nat., 105 (1992), 127-146. · Zbl 0825.62748
[3] Tesaříková E.: Variant II of probabilistic model of the school-achievement test with double-choice response. Acta UP Olomucensis, Fac. rer. nat., 110 (1993), 159-164. · Zbl 0806.62092
[4] Tesaříková E.: Statistical analysis of variant II of the probabilistic model of the school-achievement test with double choice response. Acta UP Olomucensis, Fac. rer. nat., (1993), 165-175. · Zbl 0806.62092
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.