MANA 4328 Dennis C. Veit dveit@uta.edu Measurement MANA 4328 Dennis C. Veit dveit@uta.edu 1
Basic Concepts HOW TO USE TESTING Common reasons for testing Job Change Skills necessary have changed New Position created Poor quality of output from new employees How do you know if your tests are accurate Many people taking the test One person taking the test many times 95% Confidence Intervals Variability and comparing test scores Mean / Standard Deviation Z scores and Percentiles Correlation coefficients What type of test are there Test vs. interviews Bang for the buck: getting the most from your tests
Variability Sample selection portfolio 30% 40% 20% 70% 80% Personality Inventory Identify Difference Test Critical Thinking Situational Judgment Test Interview Teamwork 30% 40% Problem Solving 20% Responsibility 70% Attention to detail Flexibility 80%
Validity Tests Why perform a Validity Test? We want an answer to a simple question: Do people who do better on this test perform better on the job? A simple question, but how do you answer it? THREE TYPES OF VALIDITY TEST: Content Validity Construct Validity Criterion Validity
Types of Validity Criterion-Related Content-Related Job Performance Job Duties Selection Tests KSA’s
Variability Content Validity – A measure of how well the content of the test corresponds to the content and requirement s of the job. Construct Validity – A measure of how well a test assesses intelligence or extroversion or shyness. Criterion Validity – A measure of the link between test results and job performance.
Understanding Correlation and Statistical Significance - Z Score or “Standard” Score Score – Mean Z Score = Std. Dev Test 1 Test 2 Bob Jim Sue Linda Raw Score 49 47 Mean 48 46 Std. Dev 2.5 .80 Z score .4 -.4 3.75 1.25
Z scores and Percentiles Look up z scores on a “standard normal table” Corresponds to proportion of area under normal curve Linda has z score of 1.25 Standard normal table = .9265 Percentile score of 92.65% Linda scored better than 92.65% of test takers Z score Percentile 3.0 99.9% 2.0 97.7% 1.0 84.1% 0.0 50.0% -1.0 15.9% -2.0 2.3% -3.0 .1%
What Types of Test Are There? Skill or Ability Tests - These measure some basic talent, such as a general mathematical ability Personality or Interest Inventories - Called inventories rather than tests because they identify characteristics, rather than measure according to a standard. Knowledge Tests - Measure a level of understanding in a specific area – (usually certification tests)
Returning “ROI” Utility - Is based on three factors: ROI - “Return on Investment” Often described as Utility How do you track? Utility - Is based on three factors: % of people in the general population (applicant pool) Validity of test to predict performance Level you use for cutoff
Why use Test over Interviews? Tests tend to be more reliable – Since tests are standardized, all candidates answer the same questions and their responses are scored consistently Tests allow us to evaluate complex competencies without formal training – A well designed test allows us to gather reliable and accurate information that might be difficult to gather effectively through an interview Test can help reduce time you spend in interviews - Test can “screen out” unacceptable candidates
Advantages of Interviews vs. Tests Interviews are often perceived to be more fair than tests– Allows for a fairer interview than just numbers Interviews profed face-to-face opportunity– Good idea to meet people Provides for “follow-up” questions - Test do not give you this opportunity
Factors to consider when determining cutoff score and evaluation of the ROI test Cost of recruitment – will raise the cost the more you add. Quality of the applicant pool - must have a valid number of applicants in our pool, which may require more time Advers Impact - On specific cutoff scores on protected groups Difference in performance between good and poor employees