We know the quality and accuracy of information in resumes is very suspect, often full of outright lies. So why is it that we put so much emphasis on the resume for evaluating candidates? The quality of information used for candidate evaluation has a direct relationship to the quality of a hiring decision.
4 \& q5 K  [2 r8 dStaffing professionals often inquire about the validity of an assessment before considering its use in a recruiting process. Why is it that the same question is not asked about using resumes in the recruiting process? A resume is a data source that's used to make decisions about a candidate. Assessment results are a data source used to make decisions about a candidate. Which data is more reliable, more objective, and more accurate at predicting success on the job?
+ T  v, V8 G7 ?! Y/ x. S1 hIn the tradition of assessments, the candidate often completes the assessment under supervision, works without the assistance of others, and often performs with limited resources under time pressure. The outcome is an objective description of one or more characteristics about an individual. In the tradition of resumes, quite the opposite is true.5 R2 N1 u: G' f5 m3 V2 }
Present business culture encourages the use of assistance in drafting a resume. Paid professionals are engaged to craft the message in an oh-so-appropriate manner. The document is completed in an unsupervised setting and evolves over lengthy periods of time based upon receiving the editorial assistance from many.
6 `* J- n2 l4 DAnd with the high probability of being scrutinized by a word-search engine, keywords carefully punctuate each work experience. After all this effort, who and what does the resume actually describe?
* X. K2 V6 h( @; f% H( u4 J, G/ ZResume-based candidate evaluation relies heavily upon a document that, by design, misrepresents the individual we seek to know. Assessment-based candidate evaluation, on the other hand, relies heavily upon obtaining data about an individual in a manner that reduces or eliminates misrepresentation.
  k4 S' v& ]6 _; e- Q! j, [Research ResultsThere's 85-plus years of research documenting both best practices and the value assessments can bring to the recruiting process. Recent research from a variety of sources has documented the degree of inaccuracies, misrepresentations, and bold-faced lies individuals weave into their resumes.
" x2 z$ K# |$ @6 X# ^: lJude Werra of Jude M. Werra and Associates has been publishing a Liars Index for over 10 years, based upon his firm’s personal experience verifying resume assertions. Yet in spite of all this, the resume is still at the core of parsing candidates into yes and no piles. Why?
) E* G/ H2 ~7 ~6 B' IAt a recent conference presentation on using assessments, a participant asked, "How do you tell a candidate that they didn't advance because of their assessment results?" The presenter gave a sound response, yet it lacked a simple truth.
; ?4 y& f5 b: d' @& _The answer should be the same as how you inform a candidate she or he did not advance due to their resume results.
8 X* e7 t& c: Y7 ?. T( X, I! T"After a review of the information submitted by you and all of the other candidates, we advanced those candidates identified to be the most qualified for the job requirements."+ Z/ K, o( H( \
If there's concern about openly discussing assessment results with a candidate, then the quality and job relevance of the assessment comes into question. Why do we feel squeamish about explaining the results of objective evaluation, yet have no qualms about dealing with the results of the subjective evaluation?

Request vs. SubmissionAn assessment is a standardized request for information from the candidate. The request is exactly the same for each candidate: fair, consistent, and reliable. You get what you ask for.
' C2 \6 ?, o0 W, k9 [A resume is an unpredictable submission of information (some fact, some fiction) by the candidate: biased, inconsistent, and unreliable. You get what they want to share.
. C2 i) |" W1 E4 mThe use of assessments requires education, discipline, and process controls to achieve meaningful results. The use of resumes is implemented with few standards, little-to-no education or training, and typically produces variable evaluation results. The heavy reliance on candidate evaluation methods that require little discipline may be one of the reasons some people find it easy to hate HR.
- K5 N/ U8 L. ~% PMeasurement Discipline: J8 l2 l9 K4 J0 N- ^
Validation is a measurement and analysis procedure that documents the relationships between evaluation methods, quality of hire, and the fairness of the hiring process. This disciplined and analytical process isn't possible with resume-based decision-making. The return on investment from using objective candidate evaluation can be calculated. Only those who engage in the disciplined approach of measurement-based candidate evaluation, however, have the data to calculate ROI. A look at the evidence is compelling. ( O% C8 a8 g& @$ P% |$ N" \
Six Sigma is a measurement discipline used to reduce variation in a process and increase yield. At its essence, assessment is a measurement system for a business process called staffing. Assessment can be implemented much like Six Sigma. The first step in considering the value of investing in a measurement discipline for your staffing process is to collect some data and examine the scope of your current variation and yield.
& B* P& l, @0 M8 oIn other words, your current process hired your superstars and your bottom performers. A simple analysis will show you the impact that bottom performers have on your overall results.
  Y* J- B6 K! R4 [: _9 |Pareto AnalysisThe 80/20 view of performance is a fast and effective method of exploring performance variation. Here are some suggestions to complete this analysis on one of your hiring populations.& J) x! }: H* K8 \: r, @1 Z9 x: v, I" }
Identify a job or job family with the following criteria:
3 U: I8 ^& w8 \! O2 ?

• One of the largest job families in the company.
• Some form of objective performance data is available on people in the job (productivity, sales, accuracy, attendance, first-time repair, first-call resolution, safety, training results, etc.).

Obtain the performance data. Set up the data in a spreadsheet. Names are not needed. Place each performance metric into its own column and sort it into a top-down rank order.
6 x# c* _; U* |/ g. DConduct the following calculations.& S/ b2 B0 ]. f% Z$ l4 f

• Calculate the average of each performance metric.
• Calculate the average of the top 80% for each performance metric.
• Calculate the average of the bottom 20% for each performance metric.

Examine the results and ask:
6 p6 o5 {) F+ b" a

• Is the variation between the top 80% and bottom 20% having a significant impact on overall productivity?
• How would hiring more candidates like your top 80% impact the achievement of your performance objectives?
• What part of your current candidate evaluation process should be preventing or minimizing the likelihood of hiring people like your bottom 20% performers?

Your response to these questions describes the variation, yield, and impact from your current candidate evaluation process. Your opportunity is to reduce low-end variation or to stop hiring candidates who become your bottom 20%. This can be done with more objective candidate evaluation methods.& `, G. t2 ?* a, _! C
Reducing the range of performance variation and improving the yield of your staffing process is one of the highest ROI projects you can initiate.
; H4 q$ [+ f/ G: P/ u, \& T* e$ d2 \
9 h1 p" Y) p3 l" q6 d  LBegin by exploring your options for a method of predicting job performance that is backed up with a validation analysis. An excellent first step in this direction is to go to your nearest college or university that has a psychology department offering courses in tests and measures. Speak to the professor and consider taking a few of the courses.
! u5 g# J9 V: V7 G/ LAnother option is to contract with or hire an I/O psychologist. Learn what a validation analysis really is and put yourself in a good position to say, "Our candidate evaluation method is valid."