Self-report assessments, although is the preferred method to measure many psychological scales, <u>tend to suffer from reliability problems. </u>Reliability, in statistical measurement terms, <em>refers to the level of consistency displayed by a scale measuring a variable. </em>
Self-report assessments are at risk of social desirability bias, where respondents tend to report answers that would be in line with the outcome preferred by others around them. Underreporting and exaggeration of certain traits are also a commonplace issue with self-report assessments.
However, ways to lessen this have been conceived, both in survey design and statistical control, such as for example, implementing the lying scale, designing reverse items, avoiding the use of even-numbered Likert scales, and many more. The more sophisticated the personality indicator, the more would they implement such mitigations.
To know whether the self-report assessments you are about to use are reliable or not, you can look at its external reliability-testing scores. Scores that are above 0.7 tend to have good reliability. It is also possible to use internal reliability-testing scores (such as Cronbach's alpha), but the measurement might suffer from issues described above.
