Employment Test, Employment Testing and Pre-employment Testing with CraftSystems Human Capital Management   "Talent Assessment Specialists
for Over 25 Years."  
Article Archive
    Library

Product Info
Employment Test Resources
    Personality (Job Fit) Employment Test
    Learning Aptitude Employment Test
    Interview Guides
    Sales Aptitude Employment Test

Employee Development
    Personality (Coaching)
    360 Feedback Dimensions

Employee Satisfaction Surveys
    Culture Survey
a member of the PreVisor family
 

Are Your Applicants Honest With You? Understanding the CPQ Accuracy Index

By Dr. Larry Craft


In the late 1970's, I conducted research to determine whether applicants would admit what every manager already knew: Applicants Lie. In my previous management position I had conducted hundreds of employment interviews and listened intently as applicants tried to say exactly what I wanted to hear. If I asked them to "Describe your worst fault," I would receive a response such as, "I try too hard." I would quickly lose my motivation to ask probing questions because I knew what was coming. The majority of the time they would spend their time crafting the perfect response that gave them the best possible impression and told me nothing of any substance. I decided I'd do something about it.

I created a personality questionnaire that tried to "pin them down" to a realistic and honest response that I could use to help them make a career decision and help me avoid costly hiring mistakes. While constructing the questionnaire, I worked backwards from the goal that this employment test would be "fake proof." I began with a review of the most recent research and found very little being done in the field of psychology. It seemed that most psychologists didn't have to worry about "fakers" because most of their test takers were volunteers who were looking for solutions to their mental health problems.

I quickly realized I'd have to go outside the box and develop a new strategy to not only discourage faking but measure it so that we could either adjust their scores or invalidate the test results. My initial research focused on determining which items in the questionnaire were the easiest to fake. Once they were analyzed and modified (or eliminated), I tried to come up with a way to make applicants more honest when they responded to each item. I simply added another set of responses and asked the test takers to describe the most Favorable response (mFr) that would most impress their employer before they responded to the most Honest response (mHr) which honestly and accurately described themselves.

Over the next ten years, research confirmed the success of this unique "mFr/mHr" response system. Not only did it significantly reduce applicant faking, it became an integral part of the Accuracy/Validity Index that measured the impact of the applicant's faking. Data on over 40,000 applicants confirmed the fact that our questionnaire had become the first in the industry to reduce and measure applicant faking, significantly increasing the accuracy and predictability of this unique personality test.

But how can you use this unique Accuracy/Validity Index in your selection process? First and foremost, this Index tells you whether you can depend upon the scores to either select or coach your new employee.

High Accuracy
Even when describing potentially negative behaviors, your applicant has been frank and forthright. If the individual's Compatibility Rating is High, this applicant's personality and motivation are compatible to the job description. In addition, be sure to assess his or her ability (skills, knowledge and experience) and reliability (work ethic/self-discipline) through the use of CraftSystems' Interviewing Guide. When it comes to coaching or training your new employee, there is evidence that suggests employees with High Accuracy (especially when coupled with low Good Impression) have fewer ego defenses, causing them to be more open and receptive to instruction.

Moderate Accuracy
There were times when your applicant unintentionally attempted to leave an unrealistically favorable impression. Some of the scores will be less accurate than others. Cross-validate these test results using our follow-up questions that are available on the Internet. Generally, these results can still be used for training and development purposes. You should know that "compliant" personalities who are highly motivated to please others often have Moderate Accuracy due to their need to be socially desirable. Unlike applicants who intentionally "fake" the questionnaire, these applicants generally avoid all negative behaviors and avoid offending others at all costs. Note: Compliant personalities can be recognized by their low Need for Control (Nc) Scores and high scores on Need to Nurture (Nn).

Invalid
This Invalid rating is based on extensive research from candidates who attempted to "sell themselves" to leave an unrealistically favorable impression when responding to the CPQ. Though many of your candidate's individual responses may be accurate, the overall Compatibility Score cannot be calculated. As a result, CraftSystems recommends that you DO NOT use these CPQ results in your hiring/selection decision. Instead, rely on your candidate's Employment Interview Guide responses, background history, and other information. It is important to understand that there is no research that indicates Invalid results are related to honesty or dishonesty. This candidate's results/reports should only be used for employee development or training.

Important Note: You should know that there are literally hundreds of personality questionnaires in the marketplace. Psychologists developed the great majority to use in a mental health (not employment) environment where faking was seldom a problem. A host of other personality tests were recently developed by marketing specialists who have very little knowledge of test validity or reliability. The test items are so transparent that any "test-wise" applicant can see-through the questionnaire and respond in a way that generates the highest possible score. Very few of the existing personality tests have the sophistication and research required to reduce and measure applicant faking. Whichever questionnaire you're presently using, you can test its transparency by giving it to a few of your employees and asking your employees to fake the test to get the highest score. If the test developer's scoring routine does not warn you of "invalid" or "questionable" scores, you should reconsider your use of the test.

 

Assess the Assessment!

If you are a manager with 10 or more employees, business owner, or a consultant, we invite you to complete our complimentary 10-15 minute CPQ personality profile and receive its personalized reports as well as a copy of Dr. Craft's "Developing HorsePower" audio cd. All information will be kept confidential and will not be distributed without your written permission.