Fairness

P. Schaap
D.Com.

RESPONSIBILITIES OF TEST USER

According to the APA [American Psychological Association, 1974] almost any test is of use in some circumstances, but even the best test can damage a subject if used incorrectly. In this regard, the APA (1974) prescribes that test users should be knowledgeable on the reasons for using a particular test, its results and the procedures required to maximize the test’s effectiveness and to limit unfairness. It is a prerequisite for a test user to know the underlying principles of test development as well as the related research done on that particular test. It is the test user’s responsibility to ensure that the test-based interpretations are fair and that the test is used correctly.

The test user should know the psychometric qualities of a test such as its reliability and its validity indicators to determine the quality of the test according to certain standards. Ethical and social values should also be considered when choosing a test and considering what the consequences of the test will be on the individual as well as on his\her human rights. The responsibilities of the test user are clearly stated by the APA (1985) and includes the following professional standards for test users:

• The available information about the validity and reliability of a test should be evaluated against the planned use of the test (Primary).
• Comment: If insufficient reliability and validity information is available, the test user should refrain from implying the decisions and conclusions are based on known reliability and validity indices.
• If the test is applied for other uses than validated for, it is the responsibility of the test user to validate the test accordingly or to provide proof of these validation particulars (Primary).
• Comment: Which is in fact a quest for one or the other form of situation-specific application of the instrument.
• If the test is merely for screening purposes prior to further evaluation, the test results should not be used for classification and decision-making objectives before proof of sufficient reliability and validity has been provided (Primary).

1.2 Standards for Employment Tests:

Tests are seen as the most valid and non-discriminatory decision-making aid for employment if used correctly. For this purpose, the APA (1985) prescribes the following standards:

Criteria-related validity determinations can only be generalized from one situation to another if it has originally been conducted in ideal circumstances and if the various situations are similar.

• Comment: This implies that many similarities should exist in work duties and requirements, underlying psychological constructs and predictors (tests).

If tests are used for job classification decisions, proof of the differential prediction value between jobs or job groups should be documented.

It is important that the rationale for the applicability of a criterion should be stated explicitly. It should include a job description as well as the base of appraisal for the applicability of the criterion.

• Comment: The applicability of a criterion is judged according to the relation between behaviour and other elements evaluated by the criterion and the goals of the organization. It should reflect substantial aspects of the work, meaning the main factors reflecting the individual’s contribution to the organization. The comprehensiveness of the criteria should however, not be of such an extent that it harms the reliability of the measurement or cause substantial bias.

The validation of test contents should be based on an explicit definition of contents field. Regarding selection, classification and promotion purposes, the contents of the test should be based on job analysis data.

• Comment: The job should be described in term of characteristics representing test contents. It should not change substantially with time and contain characteristics required for the job.

If the contents-related validity evidence has to justify the use of a test for selection and promotion purposes, a close relation between test contents and job contents should be demonstrated in terms of task contents and the skills required of a successful work performance.

If the contents-related validity evidence is used, the rationale for describing and defining a particular job content field (for example in terms of tasks to be executed or knowledge required, skills, abilities or personal characteristics required) should be clearly stated. The rationale should reflect the knowledge, skills and abilities defined as the main determinants of achievement in the specified contents field.

• Comment: If contents validity evidence is to be provided, a description of the job characteristics representing the test should be provided. This description should include relative frequencies and importance of work elements.

If a test is validated for the selection of candidates for a specific post, it can only be generalized to similar positions if the critical job contents factors are substantially similar.

• Comment: This could also be seen as a quest for a situation-specific (applicant-in-relation-to-the-particular-post) approach.

If construct-related evidence is used to determine the validity of a personnel selection test, the following two aspects should be considered:

Firstly, proof of validity should be provided that the test does measure construct; and 

Secondly, proof should be submitted that the construct is the main determinant of work achievement.

There should be a clear conceptual rationale in the above relation. Both the construct and the job-related elements linked to each other, should be clearly defined. The proposed linkage should be confirmed by empirical results following a constant pattern. The judgement of experts alone should never be used as confirmation of construct-related validity.

The technical basis valid for cut-off points of personnel decisions should be clearly presented. Cut-off points should not be solely based on the recommendation of the test manual.

DIMENSION DESCRIPTIONS

INDEX 4: CALCULATIONS [SPEEX 301 AND 302] GENERAL DESCRIPTION OF THE DIMENSION

Numerical potential: The ability to work with figures and numbers; to add, subtract and divide.

INDEX 5: MENTAL ALERTNESS [SPEEX 400: OBSERVANCE] GENERAL DESCRIPTION OF THE DIMENSION

The ability to recognize/observe deviations from and differences between seemingly unrelated items and materials - this ability is usually associated with the ability to classify objects correctly and is of special interest in administrative and clerical fields. It is, however, also strongly associated with cognitive ability in general and intellectual sharpness in particular.

INDEX 7: ABSTRACT REASONING [SPEEX 100: CONCEPTUALZATION] GENERAL DESCRIPTION OF THE DIMENSION

The ability to reason in terms of concepts and ideas; it is associated with imaginativeness. The concept is possibly also related to mental alertness and environmental awareness.

INDEX 8: FLEXIBILITY / ADAPTABILITY [SPEEX 1300: ADAPTABILITY] GENERAL DESCRIPTION OF THE DIMENSION

Flexibility in PIB terms refers to the person’s ability and willingness to adapt to change; to new conditions and circumstances; to break with routine and the old ways; in short, the general ability to accept and change.

INDEX 9: INTERPERSONAL RELATIONS [SPEEX 1200: SOCIALIZATION] GENERAL DESCRIPTION OF THE DIMENSION

This index screens the respondent’s ability, need and willingness to relate (effectively and with the required level of empathy, when called for) to people (in the workplace).

INDEX 10: SELF-IMAGE [SPEEX 1100: SELF-ACTUALIZATION] GENERAL DESCRIPTION OF THE DIMENSION

The person’s view and appreciation of himself/herself, the extent to which he/she feels accepted by other people; to which he/she feels self-confident; able to assert himself/herself; to set an example to others in some respects; to act securely; to lead and to be followed.

INDEX 15: MOTIVATION [SPEEX 1401 AND 1402: ACTUALIZATION] GENERAL DESCRIPTION OF THE DIMENSION

The index screens whether the person is motivated by internal factors (internal locus of control) or by external factors (external locus of control). The internally motivated person is self-motivated and often not in search of encouragement from external sources to act or perform. The externally motivated person needs continuous encouragement and support from others.

INDEX 24: COPING SKILLS [NO SPEEX EQUIVALENT] GENERAL DESCRIPTION OF THE DIMENSION

The index screens for pessimism versus optimism; negativism versus positivism; feelings of rejection and dejection versus feelings of being accepted by others; the feeling of being able to cope versus feelings of being unable to cope with the demands made by life in general.

CONCLUSION

There is a clear construct and predictive validity evidence concerning the PIB in a financial institution. There are indications that the PIB is a valid predictor of performance in a financial institution. All the Indices, except Index 16.3 seem to relate significantly to performance, although some to a lesser degree. Indices 4, 11 and 16.5 seem to be the best predictors in this case.

There is strong construct related evidence for all the Indices, except Index 16.3. Additional and on-going research into the validity and reliability of the PIB in financial and other institutions is of great importance. Although much has been attained through the research done according to the standards set by the APA [American Psychological Association] further analysis needs to be done concerning reliability and the effect of pre-selection on validity coefficients. Cross validation and refinement of the selection battery need to be done on a continual basis.