Tests of measurement

Tests of reliability: if the measuring instrument provides consistent results, it is considered to be a reliable instrument. A reliable instrument need not necessarily be a valid instrument. For example if a scale always over measures by six kgs, it is a reliable instrument scale but one does not get a valid measure of weight. Reliability is not as crucial as validity but it becomes easier to assess reliability if it is compared to validity. If the instrument satisfies the quality of reliability, then during its usage the researcher could be confident that the transient and situational factors would not interfere.

Reliability could be improved in two ways:

If the conditions are standardized under which the measurement takes place, the researcher must ensure that the sources of variation that are external like boredom, exhaustion are lessened to the maximum extent possible, whereby improving stability feature.

By careful designing directions for measurement with no group to group variation, by making use of trained and motivated individuals for the purpose of conducting the research and also by broadening the sample of used items. This would enhance the quality of equivalence aspect.