Unit 2: Bias in validity studies

In this talk, you will learn about common sources of bias in validity studies, how these biases can affect correlation results, and the impact they have on interpreting test manuals. Dr Graham Tyler explains the importance of criterion reliability and validity, the effects of restriction of range, and how contamination can artificially inflate correlations. You will also discover ways to evaluate and avoid these biases when assessing the usefulness of psychometric tests in the workplace.

After this talk, you will:

• Understand what bias in validity studies means and why it matters for interpreting test manuals.

• Know the four main types of bias in validity studies.

• Be able to explain how issues with the reliability of the criterion can reduce correlation results.

• Recognise how validity problems with the criterion used can lead to lower correlations.

• Understand the effect of restriction of range and how it can artificially reduce the strength of correlations.

• Know what criterion contamination is and how it can lead to an inflated correlation.

• Be able to identify when corrected correlation coefficients are reported and know why publishers might use them.

• Understand why proper training in rating and defining criteria is important for reliability and validity.

• Know which type of validity is most relevant depending on whether you are using a norm-referenced or criterion-referenced approach.

• Be prepared to critically evaluate correlation results in test manuals, considering possible biases and corrections.