Personality of Psychology

RUNNING HEAD: Measures of Personality Research

Measures of Personality Research

If psychologists had a test that measured personality accurately and reliably, it would be on the most valuable tools. Clinical psychologists could quickly analyze their clients’ strengths and weaknesses, pinpoint sources of stress and anxiety, and determine the most effective way of helping in the struggle to cope. Guidance counselors would have a sure guide to jobs and careers, for personality is a major factor in determining whether a person will be happy and successful as a salesperson, teacher, police officer, or accountant. Research psychologists would have an invaluable aid in studying the conditions that foster or inhibit the flowering of personality.

Psychologists have spent a great deal of time, effort, and ingenuity on the creation of personality tests. They have devised several hundred tests that are useful in many ways, but they have yet to find the perfect test. Perhaps they never will because personality is such a complex matter-the product of a tangled and endless weaving together of experience beginning at birth, continuing throughout life, and unique for each individual that the difficulties in measuring it are staggering. In this paper I will attempt to show the development and use of the following assessment: The Hogan Personality Inventory and Myer Briggs.

The Hogan Personality Inventory

The Hogan Personality Inventory (HPI) is a well recognized, widely used personality inventory that was initially developed in the USA for industrial-organizational and vocational applications (Hogan & Hogan, 1995). It is based on the Five Factor Model (FFM) of personality (Costa & McCrae, 1992; Goldberg, 1990). The HPI makes use of homogeneous item clusters (HlCs) to construct seven primary scales that reflect aspects of the Big Five dimensions. The British version of the HPI was recently made available for use in personnel selection. Although there are only minor differences between the British and US versions, the test is supported by the US manual (Hogan & Hogan. 1995), with a number of items having been anglicized to British spellings and phraseology to be more acceptable to the British cultural context. It includes 206

dichotomous (true-false) items. The British HPI has 41 homogeneous item clusters which in turn load on to the seven primary scales.

The primary scales of the HPI are: Adjustment (Emotional Stability in the Five Factor model), Ambition and Sociability (Extraversion in the Five Factor model), Likeability (Agreeableness in the Five Factor model). Prudence (Conscientiousness in the Five Factor model), and Intellectance and School Success (Openness to Experience in the Five Factor model). In addition, there are six occupational scales that combine homogeneous item clusters from multiple primary scales. These occupational scales are constructed to predict performance in particular settings (sales, clerical, managerial, etc.). The occupational scales of the HPI are: Service Orientation, Stress Tolerance, Reliability, Clerical Potential, Sales Potential, and Managerial Potential. Further details of these scales, and those for the OPQ and BPl, can be

found in Anderson and Ones (1998) and Ones and Anderson (1999).

Occupational Personality Questionnaire

The British standardized student version FS5.2 of the Occupational Personality Questionnaire (OPQ) was used. OPQ is a widely used measure of normal adult personality designed specifically for work psychology applications. The FS5.2 version of the OPQ was developed specifically for student testees and comprises 136 normative items using a 5-point response scale (strongly disagree to strongly agree). Of these items, 128 contributed to content scales, and eight items made up the social desirability scale. Initially, the OPQ was designed around a four-factor model of personality comprising the Feeling, Relating, Thinking, and Energies domains (SHL, 1999).

There are primary content scales on the OPQ, each containing eight items. The scales of the OPQ are: Influential, Relaxed, Sociable, Emotionally Controlled, Empathic, Active, Innovative, Decisive, Social Desirability, Traditional, Competitive, Data Rational, Achieving, Conceptual, Outspoken, Optimistic, and Methodical.

Business Personality Indicator

The Business Personality Indicator (BPO, now updated and marketed by ASE under die title 'Pin-Point', is also a measure of adult normal personality (Feltham & Woods, 1995). The BPl uses 128 trichotomous items (yes-?-no) to measure 11 primary content scales. These scales are: Change Oriented, Risk Taking, Competitive, Limelight Seeking, Work Oriented, Stamina, Perfectionist, Time Managed, Warm, Outgoing, and Worrying. The BPl makes use of these 11 scales in forming five higher second-order dimensions that are similar to some of the FFM: Worrying, Extravert, Dynamic, Work Stamina, and Controlled.

The major endeavor in personnel selection has been the search for predictors with high levels of criterion-related validity and low adverse impact. From an applied point of view, gender and ethnic group differences on personality scales are relevant to the extent that they influence selection outcomes in organizations. Selection practitioners can be encouraged by the lack of subgroup differences found across the three tests included, particularly with regard to male-female differences. Furthermore, on all three measures in practice individual scales will often be interpreted in combination or even summed to produce second-order factor scores. Thus, any minor differences which may have been apparent at the individual scale level may cancel out at the second-order factor level. Given these mostly negligible to moderate group differences and the similarity in standard deviations between the groups, calls for gender-specific or ethnic-group specific norm tables are of dubious veracity for Britain (they are already illegal for utilization in personnel selection in the USA). Indeed, in the absence of significant subgroup scale differences, we would argue that minority group norms appear to be unnecessary; rather, it is preferable to retain the larger sample sizes of combined sex and ethnic group norm groups. Finally, without further empirical data, it would be unwise to attempt to generalize