Selectivity of an item

This item has been on the quality assurance side of the portal mathematics entered. This is done in order to bring the quality of the mathematics articles to an acceptable level . Please help fix the shortcomings in this article and please join the discussion !  ( Enter article )

This article deals with the term selectivity as a description of the correlation of the item with the overall result of a test. For the selectivity as a counterpart to the error of the second kind (1-β; test strength ) see selectivity of a test .

In classical test theory, the selectivity of an item , also called item selectivity , formula symbol , is understood as the correlation of an item  i with the overall result of a test . ${\displaystyle r_{i(t-i)}}$

The selectivity is intended to enable an assessment of how well an item distinguishes between people with low and high characteristics. A highly selective item allows a prediction that persons with a high (low) test result or a high (low) characteristic score will achieve a high (low) value for a given item. For this purpose, the correlative relationship between an individual item and the overall test is determined.

For the determination of the selectivity i. d. Usually a "part-whole correction" is carried out, i. H. the item is related to the test value, which is determined from all other items in the test:

${\displaystyle r_{i(t-i)}={\frac {\sigma (x_{i},x_{t-i})}{\sigma (x_{i})\;\sigma (x_{t-i})}}}$

With

• Values for the item i ${\displaystyle x_{i}}$
• Values for the overall test without the item i. ${\displaystyle x_{t-i}}$

The range of values ​​for the selectivity is between –1 and +1:

• With a high positive selectivity, the item records something similar to the overall test.
• A selectivity close to 0 indicates that an item has little in common with the rest of the test.
• A negative selectivity can indicate that an item was not understood as intended, but the other way around or that it was not correctly polarized during the evaluation.

Rule of thumb for the item selection: the items should have at least better selectivity than +0.30.

See also

• Item analysis

literature

• Hermann-Josef Fisseni: textbook of psychological diagnostics . 3rd, revised and expanded edition. Hogrefe, Göttingen 2004, ISBN 3-8017-1756-9 , p. 36-40 .