The Fallibility of Psychological Testing
Psychological Testing has become rampant across industries, more so inthe case of Information Technology, BPOs and ITES companies. Thesetests are used to 'throw up' personality profiles and competencydescriptions that would help companies recruit the 'right' candidate.The Human Resources department in most organizations is responsible forthe administering of Psychometric tests.
The International body that sets guidelines for testing is theInternational Test Commission ( ITC ) which stipulates guidelines foradaptation and usage of tests. ITC has issued guidelines to cover thefollowing -
Professional and ethical standards in testing
Rights of the test candidate and other parties involved in the testingprocess
Choice and evaluation of alternative tests
Test administration, scoring and interpretation
Report writing and feedback.
ITC has defined competence in test usage as, " A Competent test userwill use tests appropriately, professionally and in an ethical manner,paying due regard to the needs and rights of those involved in thetesting process, the reasons for testing, and the broader context inwhich the testing takes place."
There are many reasons why a test can be fallible -
1. Inappropriate usage of Tests
The norms for which the tests have been designed have to be consideredcarefully. They cannot be administered to a population that isdifferent from the norm population (Norm Population for example couldbe, adolescents, Senior Executives; norms could be based on age,occupation, gender, economic status etc.). That is, a test designedfor adolescents cannot be administered on Senior Executives and viceversa because the results of such administration would be faulty.
2. Culture Fair Testing
Another common flaw is in the usage of tests which are not culture fair.Many psychological tests have been designed for the western populationand can be used appropriately only in that culture. Unfortunately manyof these tests are used on the Indian population giving rise to faultyconclusions.
3. Design of Test
The psychological test can only be designed by professionals qualifiedin psychometrics. Many a time lay people design tests using the help ofinformation accessed through the internet or sometimes even from theirown common sense. Such tests are not valid.
4. Validation of tests
Every test that has been designed has to be validated. The process ofvalidation would involve administering the test on a large populationover a period of time before it is certified as 'ready for use'. Thiscannot be done from tests that are fabricated overnight.
5. Downloaded Tests
Often tests are downloaded from the Internet and are used for purposesother than for which they have been designed. For example, apersonality test being used to test the Emotional Intelligence or TeamSkills of a person.
6. Wrong Customization
Sometimes the items of a test are changed to suit the user. In suchinstances the test cannot give valid results. For example an item ( aquestion / statement in the test ) would state - " Do you usually date on weekends ?" , could be changed to " Do you socialize on weekends? " to fit the Indian scenario. The results of the test with such changed'items'will not be not valid.
7. Test Administration
The test administration should be systematically standardized, i.e., thetest should be administered under exactly the same conditions to allusers. This means that the actual test environment, the instructions,the timing and the materials are the same on every test occasion.Before administering the test the user must consider the followingguidelines -
• What is the purpose of testing?
• What outcomes will be achieved through testing?
• Why are these specific tests being used?
• Why are these tests relevant to the outcomes being sought?
• What evidence is there that these tests are appropriate for the peoplewho are to be assessed?
8. Confidentiality of Test Results
All results of tests must be maintained by the administrator in strictconfidence. Revealing the results could lead to unfair discriminationin the workplace. This happens when a test used for selection isdisadvantageous to certain 'groups' within the organization. In the US,the use of IQ tests in employee selection has been prohibited. This wasdue to the lack of confidentiality maintained by the administratorswhich resulted in employees being unfairly discriminated. At presentany psychometric test has to be administered with utmost care and afterthe requisite permissions is obtained.
To ensure confidentiality, the following guidelines should beconsidered:
1. Ensure that test materials are kept securely
2. Ensure secure storage of and control access to test materials.
3. Respect copyright law and agreements that exist with respect to atest including any prohibitions on the copying or transmission ofmaterials in electronic or other forms to other people, whetherqualified or otherwise.
4. Protect the integrity of the test by not coaching individuals onactual test materials or other practice materials that might unfairlyinfluence their test performance.
5. Ensure that test techniques are not described publicly in such a waythat their usefulness is impaired.
9. Competence of Test Administrator and Interpreter
The personnel administering the tests have to be strict in following theinstructions stipulated by the test designers. The interpretations toohave to be done by qualified personnel who are psychologists.
10. Limitation of tests
The tests are 'limited' for the purposes for which they have beendesigned. For a holistic assessment of an individual, it is requiredthat the test results are supplemented by information from interviewsand group discussions. In addition the test results are not valid aftereighteen months of its administration.
11. Test Copyrights
It is an offence to copy tests that have been copyrighted. They can onlybe used after the necessary permission is obtained from the designers.Otherwise this illegal copying may lead to lack of standardization inTest conditions and poor control of materials.
12. Test feedback
Often times the guidelines for feedback are not followed by the Users.The British Psychological Society has set out the following guidelinesfor written and oral feedback:
• Ensure that the technical and linguistic levels of any reports areappropriate for the level of understanding of the recipients.
• Make clear that the test data represent just one source of informationand should always be considered in conjunction with other information.
• Explain how the importance of the test results should be weighted inrelation to other information about the people being assessed.
• Use a form and structure for a report that is appropriate to thecontext of the assessment.
• When appropriate, provide third parties with information on howresults may be used to inform their decisions.
• Explain and support the use of test results used to classify peopleinto categories (e.g. for diagnostic purposes or for job selection).
• Include within written reports a clear summary, and when relevant,specific recommendations.
• Present oral feedback to test takers in a constructive and supportivemanner.