The following is some detailed information about the test construction of the California Psychological Inventory (CPI), a very robust and well validated psychometric assessment. If you would like more general information about the CPI and its scales click here
Niche Consulting has developed its own NZ norms for the CPI from well over 10,000 business people, which allows a comparison of the individual against an appropriate business group on this tool.
A major strength of the CPI concerns the way in which it was developed. Most of the scales were developed using an “empirical keying” strategy.
What this means is that the scales were devised based on a comparison of what the scales suggested about an individual, with the way observers described them, and a measurement of their performance.
The CPI has been designed based on an instrumental standpoint, meaning evaluation of whether a scale is useful or valid has more to do with the outcome of the test than the “look” or “face validity” of the items being asked.
An instrumentally constructed test uses the score on the test to classify people in a way others classify them and the accuracy with which scores in the test predict or confirm the behaviour that is relevant to the test.
Scales developed by instrumental criteria often contain a mix of both obvious and subtle items.
When interpreting the CPI we are never looking at individual answers but looking for the trends on each scale and what that means in terms of how the individual may be described by the people who know them.
The CPI also has many more items relating to any one scale (usually between 28-40 items) than most personality tools (typically only 8 items).
The complexity of the tool comes from the way the test is constructed and that the 20 personality scales are very broad and overlap.
This means the whole profile should be interpreted rather than single scales in isolation. Tools such as OPQ, HPI or 15FQ on the other hand have very narrow scales, which are meant to have no overlap.
Buros Institute of Mental Measurement does a peer review of assessments where the scientific rigor of test development is reviewed, this is where experts “test” the test makers.
“Over the nearly five decades since the creation of the CPI, an extensive body of research has formed that examines its performance in diverse assessment populations and age groups. This body of knowledge provides a wealth of comparative reference materials."
"The CPI provides a substantiated method to aid in the consensual description of differences between individuals and groups across many substantiated dimensions of personality. Since its inception, the CPI has been quite successful in its groundbreaking attempt to describe a broad array of fairly robust personality characteristics across a wide cross section of society.”