Validated 360 surveys measure behaviors that meet the following criteria:
The behaviors are observable
The behaviors can be trained.
The behaviors do not overlap.
They are based on skills that research has shown are correlated with actual job performance. This approach also provides a hierarchy of skills, which helps feedback recipients prioritize their training. Again, the hierarchy is based on research about which skills impact performance the most.
There are many kinds of validity in psychological testing and surveys. All aim to determine whether a test actually measures what it was designed to measure. All approaches to validity have a role to play. In our field, 360 degree feedback surveys for training and development, the key type is construct validity.
This refers to the underlying qualities the instrument measures. Construct validity is evaluated by determining the degree to which the explanatory concepts behind the instrument account for an individual's performance on the instrument. For example, the Survey of Management Practices measures the qualities of management. A demonstrated strong manager should have high scores on the survey whereas a known weak manager should have low scores.
The aim of a 360 feedback research program should be to test the instruments and their underlying constructs. The research should show how the instrument meets or exceeds technical standards for each of the following techniques:
Application of judgmental and logical analysis
Correlational analysis
Process analysis
Analysis of group differences
Changes over time
Responsiveness of scores to experimental treatment.
