Reliability values for the practice

To work with sufficiently reliable data, the following values are recommended for practical use. The measurement error is then still within the acceptable range.

R?
. 50 for group comparison
R? .
70 (generally in research)
R? . 90 in single case diagnostics

Methods for determining reliability

The following methods are used in practice to determine the reliability:

Repeat test method (sample ends perform the same test twice under the same conditions)
Parallel test method (The raw values of two tests are correlated with each other)
(Test halving method (A test is divided into two equivalent halves. The two halves are correlated)
Consistency analysis (A test is performed once on a sample and broken down into as many parts as there are items in it. The items are then correlated with each other)

1. repeat test method

A test and its retest is performed at different times under identical conditions. A change of the test leader allows simultaneous determination of objectivity and reliability.

Question wording: How much time should elapse between the two tests?
Problem: A retest can include any learning experience from the first test. (e.g. learning effects, exercise effect, but also fatigue effects, motivation effects)

2. parallel test method

Two different tests with the identical target (identical validity range) are performed on the same sample. (parallel test reliability) examples: Note: Not all tests can be considered as parallel tests.

Deep Start – Fleigender Start
Medicine ball throw – Medicine ball shot

3. test halving method

The prerequisite for the test halving method is that the test can be divided into two equal halves. (Ex. 20 free throws from the free throw line in basketball).

Some tests cannot be halved (e.g. squats). Procedure: Both test halves are added together and correlated. Possibilities of halving the test:

Halving after even and odd numbers
Halving at random