Reliability & Validity of
Selection Tests
-Anjali Maurya and Neha Prajapati
Selection Tests
 Individuals differ in many respects including job-related abilities and skills.
 In order to select a right person for the job, their abilities and skills need to be
equally measured. That is done through selection tests.
 A selection test is a device that uncovers the information about the candidate,
which are not known through application blank.
 They can measure certain abilities, aptitudes and skills that provide objective
information how well the applicant can be expected to perform the job.
Selection Test Types
 Aptitude Test: Learning capacity
 Intelligence Tests: Reasoning, verbal, number, memory
 Achievement Tests: Theoretical and practical knowledge
 Situational Tests: Situational problems
 Interest Tests: Likes and dislikes of candidates
 Personality Tests: Relationship, Talkative, Reliability, Self-confident
 Honesty Tests: The "Polygraph" a lie detector mechanical device that
measure honesty of the candidate. Reduce losses and Employee theft
The Basic of Testing and selecting
employee
 Reliability
 Validity
Criterion validity
Content validity
Construct validity
Test Reliability
 Reliability is a tests first requirement and refers to its consistency.
 Reliability is the consistency of scores obtained by the same person when retested
with the identical tests or with alternate forms of the same test.
 So therefore, “A reliable test is one that yields consistent scores when a person takes
two alternate forms of the test or when he or she takes the same test on two or more
different occasions”.
How to measure Reliability
 Retest Estimate
Same test to same people at different point in time
 Equivalent form estimate
Administer a test and administer what believes to be an equivalent test.
 Inter-rater reliability
Panel of interviewer jugde , analyse candidate.
What Can Cause a Test to be Unreliable?
These include the following:
 Physical conditions -(quiet one day, noisy the next),
 Differences in the test taker (healthy one day, sick the next), and
 Differences in test administration (courteous one day, rude the next).
 Or the questions may do a poor job of sampling the material;
for example, test one focuses more on Chapters 1 and 3, while
test two focuses more on Chapters 2 and 4.
Test Validity
 Validity: is a selection tool’s defined as the accuracy with which a test, interview, and so on,
measure what is purports to measure or fulfills the function it was designed to fill.
 A test should be job related
 Performance on a test should be a valid predictor of subsequent performance on the job.
 Test validity answers the question Does the test actually measure what we need for it to
measure?
 Validity refers to the accuracy of the inferences that we can make based on the test.
For example, if Janes scores on mechanical comprehension tests are higher than Jims, can we
be sure that Jane possesses more mechanical comprehension than Jim?
 How to demonstrate Validity?
A. Criterion Validity
B. Content Validity
C. Construct Validity
 Criterion validity
A type of validity based on showing that scores on the test (predictors) are related to job performance
(criterion).
• Are test scores in this class related to students’
knowledge of human resource management?
 Content validity
A test that is content valid is one that contains a fair sample of the tasks and skills actually needed for the
job in question.
• Do the test questions in this course relate to human
resource management topics?
• Is taking an HR course the same as doing HR?
 Construct Validity
The extent to which the test measure the psychological quality or quantifies the psychological aspect of an
individual.
How to validate a Test
 Step 1: Analyze the job
– Predictors: job specification (KSAOs)
– Criterion: quantitative and qualitative measures of job success
 Step 2: Choose the tests
– Test battery or single test?
 Step 3: Administer the test
– Concurrent validation
• Current employees’ scores with current performance
– Predictive validation
• Later-measured performance with prior scores
 Step 4: Relate Test Scores and Criteria
– Correlation analysis
• Actual scores on the test with actual performance
 Step 5: Cross-Validate and Revalidate
– Repeat Step 3 and Step 4 with a different sample
of employees.
Reliability & Validity of Selection Tests....pptx

Reliability & Validity of Selection Tests....pptx

  • 1.
    Reliability & Validityof Selection Tests -Anjali Maurya and Neha Prajapati
  • 2.
    Selection Tests  Individualsdiffer in many respects including job-related abilities and skills.  In order to select a right person for the job, their abilities and skills need to be equally measured. That is done through selection tests.  A selection test is a device that uncovers the information about the candidate, which are not known through application blank.  They can measure certain abilities, aptitudes and skills that provide objective information how well the applicant can be expected to perform the job.
  • 3.
    Selection Test Types Aptitude Test: Learning capacity  Intelligence Tests: Reasoning, verbal, number, memory  Achievement Tests: Theoretical and practical knowledge  Situational Tests: Situational problems  Interest Tests: Likes and dislikes of candidates  Personality Tests: Relationship, Talkative, Reliability, Self-confident  Honesty Tests: The "Polygraph" a lie detector mechanical device that measure honesty of the candidate. Reduce losses and Employee theft
  • 4.
    The Basic ofTesting and selecting employee  Reliability  Validity Criterion validity Content validity Construct validity
  • 5.
    Test Reliability  Reliabilityis a tests first requirement and refers to its consistency.  Reliability is the consistency of scores obtained by the same person when retested with the identical tests or with alternate forms of the same test.  So therefore, “A reliable test is one that yields consistent scores when a person takes two alternate forms of the test or when he or she takes the same test on two or more different occasions”.
  • 6.
    How to measureReliability  Retest Estimate Same test to same people at different point in time  Equivalent form estimate Administer a test and administer what believes to be an equivalent test.  Inter-rater reliability Panel of interviewer jugde , analyse candidate.
  • 7.
    What Can Causea Test to be Unreliable? These include the following:  Physical conditions -(quiet one day, noisy the next),  Differences in the test taker (healthy one day, sick the next), and  Differences in test administration (courteous one day, rude the next).  Or the questions may do a poor job of sampling the material; for example, test one focuses more on Chapters 1 and 3, while test two focuses more on Chapters 2 and 4.
  • 8.
    Test Validity  Validity:is a selection tool’s defined as the accuracy with which a test, interview, and so on, measure what is purports to measure or fulfills the function it was designed to fill.  A test should be job related  Performance on a test should be a valid predictor of subsequent performance on the job.  Test validity answers the question Does the test actually measure what we need for it to measure?  Validity refers to the accuracy of the inferences that we can make based on the test. For example, if Janes scores on mechanical comprehension tests are higher than Jims, can we be sure that Jane possesses more mechanical comprehension than Jim?
  • 9.
     How todemonstrate Validity? A. Criterion Validity B. Content Validity C. Construct Validity
  • 10.
     Criterion validity Atype of validity based on showing that scores on the test (predictors) are related to job performance (criterion). • Are test scores in this class related to students’ knowledge of human resource management?  Content validity A test that is content valid is one that contains a fair sample of the tasks and skills actually needed for the job in question. • Do the test questions in this course relate to human resource management topics? • Is taking an HR course the same as doing HR?  Construct Validity The extent to which the test measure the psychological quality or quantifies the psychological aspect of an individual.
  • 11.
    How to validatea Test  Step 1: Analyze the job – Predictors: job specification (KSAOs) – Criterion: quantitative and qualitative measures of job success  Step 2: Choose the tests – Test battery or single test?  Step 3: Administer the test – Concurrent validation • Current employees’ scores with current performance – Predictive validation • Later-measured performance with prior scores
  • 12.
     Step 4:Relate Test Scores and Criteria – Correlation analysis • Actual scores on the test with actual performance  Step 5: Cross-Validate and Revalidate – Repeat Step 3 and Step 4 with a different sample of employees.