Personnel Selection: Tips and FAQs: Test Administration

CHAPTER 8
Issues and Concerns with Assessment

It is important to remember that an assessment instrument, like any tool, is most effective when used properly and can be very counterproductive when used inappropriately. In previous chapters you have read about the advantages of using tests and procedures as part of your personnel assessment program. You have also read about the limitations of tests in providing a consistently accurate and complete picture of an individual's employment-related qualifications and potential. This chapter highlights some important issues and concerns surrounding these limitations. Careful attention to these issues and concerns will help you produce a fair and effective assessment program.

Chapter Highlights
1. Deciding whether to test or not to test
2. Viewing tests as threats and invasions of privacy
3. Fallibility of test scores
4. Appeals process and retesting
5. Qualifications of assessment staff
6. Misuse or overuse of tests
7. Ensuring both efficiency and diversity
8. Ethnic, linguistic, and cultural differences and biases
9. Testing people with disabilities

1. Deciding whether to test or not to test

How successful is your current assessment program? Is it in need of improvement? The decision to use a test is an important one. You need to carefully consider several technical, administrative, and practical matters.

Sometimes a more vigorous employee training program will help to improve individual and organizational performance without expanding your current selection procedures. Sometimes a careful review of each candidate's educational background and work history will help you to select better workers, and sometimes using additional tests will be beneficial.

Consider how much additional time and effort will be involved in expanding your assessment program. As in every business decision, you will want to determine whether the potential benefits outweigh the expenditure of time and effort. Be sure to factor in all the costs, such as purchase of tests and staff time, and balance these against all the benefits, including potential increases in productivity.

In summary, before expanding your assessment program, it is important to have a clear picture of your organization's needs, the benefits you can expect, and the costs you will incur.

2. Viewing tests as threats and invasions of privacy

Many people are intimidated at the mere thought of taking a test. Some may fear that testing will expose their weaknesses, and some may fear that tests will not measure what they really can do on the job. Also, some people may view certain tests as an invasion of privacy. This is especially true of personality tests, honesty tests, medical tests, and tests that screen for drug use.

Fear or mistrust of tests can lower the scores of some otherwise qualified candidates. To reduce these feelings, it is important to take the time to explain a few things about the testing program before administering a test. Any explanation should, at a minimum, cover the following topics:

! why the test is being administered
! fairness of the test
! confidentiality of test results
! how the test results will be used in the assessment process.

3. Fallibility of test scores

All assessment tools and procedures are subject to measurement errors. This means that a test neither measures a characteristic with perfect accuracy for all people, nor fully accounts for their job performance. Thus, there will always be some errors in employment decisions made based on assessment results. This is true of all assessment procedures, regardless of how objective or standardized they might be.

It is, therefore, important not to rely entirely on any one assessment instrument in making employment decisions. Using a variety of assessment tools will help you obtain a fuller and more accurate picture of an individual. Consider such information as an evaluation of a person's education, work experience and other job-relevant factors in addition to standardized test results.

4. Appeals process and retesting

Every test taker should have a fair chance to demonstrate his or her best performance on an assessment procedure. However, at times this might not occur. If the results may not be valid for an individual, consider retesting or using alternative assessment procedures before screening the individual.

There are external circumstances or conditions that could invalidate the test results. These may include the test taker's state of mind or health at the time of the test, the conditions under which the test is given, and his or her familiarity with particular questions on the test. To give some specific examples, a person who has a child at home with the measles may not be able to concentrate on taking a vocabulary test. Someone sitting next to a noisy air conditioner may also not be able to concentrate on the test questions. On another day, under different circumstances, these individuals might obtain a different score.

If you believe that the test was not valid for an individual, you should consider a retest. If other versions of the test are not available, consider alternative means of assessment. Check the test manual for advice from the publisher regarding retesting. It is advisable to develop a policy on handling complaints regarding testing and appeals for retesting, so that these concerns can be resolved fairly and consistently.

5. Qualifications of assessment staff

Test results may not be accurate if the tests have not been administered and scored properly, or if the results are not interpreted appropriately. The usefulness of test results depends on proper administration, scoring and interpretation. Qualified individuals must be chosen to administer and score tests and interpret test results. These individuals must be trained appropriately. Test manuals will usually specify the qualifications and training needed to administer and score the tests and interpret results.

6. Misuse or overuse of tests

A single test cannot be expected to be valid in all situations and for all groups of people. A test generally is developed to measure specific characteristics and to predict specific performance criteria for a particular group. For example, a test with items designed to select salespersons may not be valid for identifying good sales managers.

In addition, test results usually provide specific information that is valid for a specific amount of time. Therefore, it is unlikely to be appropriate to consider an employee for a promotion based on his or her test scores on a proficiency test taken 5 years earlier.

The test manual and independent reviews of the test remain your best guides on administering, scoring, and interpreting the test.

7. Ensuring both efficiency and diversity

Use of reliable and valid assessment tools can result in improved performance of your workforce. However, when designing an assessment system, it is also important to consider how to ensure a diverse workforce that can help your organization be successful in todays diverse marketplace. To encourage diversity in your organization, consider how different types of people perform on different types of tests. Some research has indicated that older workers and members of a variety of racial and ethnic groups do not do as well on certain types of tests as members of other groups. For example, older people and women tend to do less well on physical ability and endurance tests. Members of some ethnic and racial groups, on average, may do less well on ability tests. Older people tend not to score as high as younger people on timed tests. Even though these groups perform less well on certain tests, they may still perform on the job successfully. Thus by using certain types of assessments, or relying heavily on one type of test, you may limit the diversity of your workforce and miss out on some very productive potential employees (e.g., if you used only physical ability tests, you may unnecessarily exclude older workers). You might also be violating federal, state, and local equal employment opportunity laws.

To help ensure both efficiency and diversity in your workforce, apply the whole-person approach to assessment. Use a variety of assessment tools to obtain a comprehensive picture of the skills and capabilities of applicants and employees. This approach to assessment will help you make sure you don't miss out on some very qualified individuals who could enhance your organization's success.

8. Ethnic, linguistic, and cultural differences and biases

The American workforce is made up of a diverse array of ethnic and cultural groups, including many persons for whom English is not the primary language. Some of these individuals may experience difficulty on standardized tests due to cultural differences or lack of mastery of the English language. Depending on the nature of the job for which they are applying, this could mean that their test scores will not accurately predict their true job potential.

Before selecting new tests, consider the composition of your potential candidate population. Are the tests appropriate for all of them? The test manuals may provide assistance in determining this. If you need further clarification, contact the test publisher.

There may be cases where appropriate standardized tests are not available for certain groups. You may have to rely on other assessment techniques, such as interviews and evaluations of education and work experience, to make your employment decisions.

9. Testing people with disabilities

Many people with disabilities are productive workers. The ADA protects qualified individuals with disabilities from discrimination in all aspects of employment, including personnel assessment. Your staff should be trained to evaluate requests for reasonable accommodation and provide these accommodations if they are necessary and would not cause "undue hardship." These situations must be handled with professionalism and sensitivity. Properly handled, this can be accomplished without compromising the integrity of the assessment process.

Accommodation may involve ensuring physical accessibility to the test site, modifying test equipment or tests, or providing other forms of assistance. Giving extra time for certain kinds of tests to test takers with dyslexia or other learning disabilities and administering a braille version of a test for the blind may be examples of reasonable accommodation. See Chapters 2 and 6 for further discussions on testing people with disabilities.

A document by the:
U.S. Department of Labor
Employment and Training Administration
1999