Should You Use Psychometric Assessments to Hire?
Given a choice between having well-designed, validated psychometric assessments as a balanced part of my selection process or not, I’d choose to have them, every time. In the now-famous words of Jim Collins, there are few things as important as getting the right people in the right seats on the bus.
Educate, aka Hit the Abusers with a Book
When people raise concerns about assessments, you generally hear concerns about misuse and abuse. Education is the best defense against misuse – which I’m sure does happen. Picking the right assessments is the best start. Are the assessments ipsative or normative? Is the validation predictive or another type(s)? I prefer normative and predictive. All of this might seem daunting at first, but it’s not that difficult to understand the basics.
For those interested in exploring assessments or just learning more, here are some decent non-partisan educational resources:
- http://www.uniformguidelines.com/index.html
- http://wdr.doleta.gov/opr/FULLTEXT/99-testassess.pdf
- http://www.shrm.org/about/foundation/research/Documents/assessment_methods.pdf
Mix Candidate with One-Third Assessment and Stir Vigorously
The other way to avoid misuse is to implement assessments intelligently. In my experience, most reputable vendors suggest using assessments as one-third of the decision process. Many now generate multiple reports, including an interview guide to help you dig into certain areas where the candidate didn’t assess well, in comparison to the predictive model for your position. In addition, the information can also be used developmentally, as well as for selection. That’s a big bonus to me.
As Frankie Said, I Do it My Way
The selection process I’ve used whenever possible is:
- Resume screen, mostly as a “fog the mirror test” and ensuring they have the sense not to use “sexydude-at-email.com” as their resume email and can represent themselves well in writing.
- Phone screen, to ensure coherent speaking ability and that I’d want them representing my product and company, as well as covering any knock-out factors (they want 25% travel and the position is 80%; they want full base + bonus and it’s small base + mostly commission)
- Conduct the Assessment
- Review the Assessment review, and if all of the above indicate we have a decent or possible match…
- Conduct a full-fledged interview (some behavioral elements, some problem-solving, some hypothetical, all judging different things) and likely, multiple interviews by different parties
- As relevant, some sort of audition or other non-psychometric assessment (for instructors a mock-classroom audition, for sales folks a sales roleplay, for instructional designers a writing assignment, etc.).
- A review and calibration of results.
- Final decision and offer.
- Background, criminal and reference checks.
Don’t Pull Grass Out with the Weeds
Now, while I strongly believe in assessments, to those who prefer to weed out candidates based on an assessment only, delivered first, I remain cautious about that.
There are several reasons why I review resumes and conduct brief phone screens prior to assessing.
Validity and Predictive Reliability
- I’ve heard many wild claims, but I still haven’t personally seen a detailed validation study, conducted with generally-accepted statistical measurement and assessment validation methods, that shows more than 75-85% predictive reliability for selection. That means that the best are still not 100% reliable for predictive validity (statistically predicting success in a certain role), and in fact, may be inaccurate between 15-25% of the time (in the above example, which is common). I’m sure I don’t need to remind anyone that 25% is 1 out of 4 and 15% is 1 out of 6.67. Much better odds than without an assessment, but for my taste, still not a standalone tool.
From a logic perspective, if I have just added assessments to improve my hiring success, performance and retention rates, why would I then purposefully deflate my potential success rates but using only the assessment?
Cost Factors
- After receiving the resume (because there are costs leading up to that point), reviewing a resume is inexpensive (just my time and lost opportunity costs, and I can scan them quickly, for the things I’m weeding for).
- The screen conversation is relatively inexpensive.
- Good assessments cost money. Real, hard dollars – an actual expense line item. (I find them worth it and believe there is significant ROI, but they’re not free. If you are hiring enough people can cost-justify and arrange an unlimited enterprise license, this is the way to go, in my opinion. But if not, you pay by the drink, each time.)
Weed-out Factors
- If the candidate uses bad judgment in their resume (aka sexydude-at-email.com) or exhibits poor communication and print-presentations skills (and assuming those are required for the job), I might weed them out without an assessment.
- In the screen, if they want a large base and bonus and the position is 100% commission, and they aren’t willing to consider the upsides and move forward, I don’t need an assessment to weed them out. Buh-bye. (And yes, I would be clear about that in postings, but people don’t always read closely, and they submit their resume anyway.)
If I didn’t do the weed out factors, it also connects back to cost. I might waste time and an assessment (and the associated costs, if you don’t have an unlimited enterprise license) on someone that I absolutely wouldn’t hire anyway. Why do that?
Putting Screens in Your Window
Now, if you are absolutely besieged with resumes and overwhelmed in the hiring process, it gets more palatable to screen with the assessment to pare down the number of viable candidates to a manageable level – but that’s using assessments as screening tool. (Of course, then you have the data for selection).
Whichever you do, do it consistently, legally and ensure you’ve got HR in the loop (preferably from the beginning). Reputable assessment companies comply with EEOC and legal guidelines, but you also need to comply with your organizational policies and practices.
Why the Risk-Averse Curse
By the way, if you do what I recommend and don’t use the assessment as a knock-out factor by itself, there is one inherent risk. Managers may still develop a preference for candidates based on old methods – because they like the candidate based on gut feel (or halo effect, or a variety of other interview biases) – and then they may resist the assessment results. That’s why I like to assess early, but after only a knock-out resume review and quick phone screen. Then you assess, before face to face interviews. It doesn’t give people as much of a chance to form those bonds, before getting the assessment data and corresponding interview guides, to prepare for the interview. In many organizations, the resume review and phone screen are done by a recruiter, who administers the assessments, and passes the candidate to a hiring manager for a decision. In those cases, this process is even more effective, because the hiring manager hasn’t had an opportunity to form a bias.
If You Could Be Any Animal…
So, those are some quick thoughts on using assessments. I use ‘em, whenever possible. Coming back to sales effectiveness, which is my playground, I always remember this:
- Pigs aren’t birds. They don’t fly.
- Turkeys are birds, but can’t achieve lift off.
- Sparrows are birds. They fly, but they’re just not as strong or big as eagles.
- Hawks are closer, but they’re still not quite eagles either.
- If you want something that looks, sounds, flies, hunts and gets results like an eagle, why not go find an eagle?
With the right people in the right seats on your bus, your training, compensation, support, policies, processes and other performance tools and resources can really help move the needle. With great performance support, you can help great people excel even more. But you still can’t make pigs fly.
Be safe out there.
Mike
____________________________________
Mike Kunkle
Contact me:
mike_kunkle at mindspring dot-com
214.494.9950 Google Voice
Connect with me:
http://www.linkedin.com/in/mikekunkle
http://twitter.com/mike_kunkle
