12.1.2 Assessment of personality

Explained

1. Self-report measures.

• The person answers fixed questions about himself (yes / no / cannot say, true / false, or a 5-point scale). Scoring is objective and quick, the test can be given to groups, and norms allow comparison. That is why inventories dominate selection and research.
• Three ways of building them: rational / theory-based (items written to fit a theory: EPPS from Murray's needs), empirical criterion keying (items kept if they separate a patient group from normals, whatever their content: MMPI), and factor-analytic (items grouped by factor analysis: 16 PF, EPQ, NEO).
• MMPI scales include hypochondriasis, depression, hysteria, psychopathic deviate, masculinity–femininity, paranoia, psychasthenia, schizophrenia, hypomania and social introversion. Validity scales (L = lie, F = infrequency, K = defensiveness) catch careless or faked answering.
• Limitations: answers can be faked; people may lack insight; social desirability and acquiescence response sets; items may not suit another culture or language without adaptation.

2. Projective techniques: how they work.

• The stimulus is ambiguous or unstructured, the task is open, and the person is not told what is being measured. Because there is no "right answer", the person's inner world (needs, conflicts, unconscious wishes) is projected onto the material.
• They come from the psychoanalytic tradition and aim at the whole personality and the unconscious.
• Scoring and interpretation need trained experts and are partly subjective.

3. The main projective tests.

• Rorschach: 10 symmetrical blots (5 black-grey, 2 black and red, 3 multicoloured). Two phases: free association ("what might this be?") and inquiry (where and why). Scored for location (whole, detail), determinants (form, colour, movement, shading), content (human, animal, object) and popular vs original answers. John Exner's Comprehensive System (1974) standardised scoring.
• TAT: pictures of people in unclear situations; the person tells a story: what led up to the scene, what is happening, what the people feel and think, how it ends. Based on Murray's need–press theory; the story's hero is taken to stand for the teller. Usually 10 to 20 cards are used, chosen by age and sex. Morgan and Murray described the method in 1935; the standard 31-card set was published in 1943.
• CAT: for children about 3 to 10 years; animals in human-like scenes, because children identify more easily with animals.
• Sentence completion: e.g. "My father ...", "I feel guilty when ...". Fairly structured, so easier to score; reveals attitudes, fears and conflicts.
• Word association: the examiner reads words one by one; the person says the first word that comes to mind; delay, odd answers or emotional signs point to "complexes" (Jung).
• Drawing tests: Draw-a-Person / HTP read size, placement, details and omissions as clues to self-image and conflicts. Goodenough's Draw-a-Man was first a non-verbal intelligence test for children.
• Picture-Frustration: shows how a person typically reacts to frustration (aggression turned outwards, inwards, or played down).

4. Self-report vs projective.

5. Behavioural assessment.

• Observation: naturalistic (in real settings) or controlled; the observer records what the person actually does, often using frequency, duration and intensity of target behaviours, or an A-B-C record (antecedent, behaviour, consequence). Needs a trained observer; the observer's presence can change behaviour.
• Interview: structured (fixed questions in fixed order, comparable answers) or unstructured (flexible, follows the person's leads, deeper but harder to compare). Used in clinics, counselling and selection; affected by interviewer bias and rapport.
• Behavioural ratings: people who know the person well (teachers, supervisors) rate specific traits on a rating scale (graphic, numerical or descriptive). Errors: halo, leniency/severity, central tendency, and the logical error (assuming two traits go together).
• Nominations: peers name who in the group is, for example, the most helpful or the best leader; useful in schools and training groups; may be biased by friendships.
• Situational tests: the person is put in a constructed situation and observed. Examples: Hartshorne and May's honesty tests; the leaderless group discussion and stress tasks used in officer selection (World War II, later SSB and assessment centres); in-basket exercises.
• Other sources: autobiography, case history, and self-monitoring diaries.

6. Limitations of personality assessment.

• No single test captures the whole person; good practice uses several methods together.
• Self-reports: response sets, faking, lack of insight.
• Projectives: subjective scoring, low inter-scorer reliability and weak validity, and dependence on the tester's skill.
• Behavioural methods: observer bias, halo effect, time and cost, and behaviour seen in one setting may not generalise.
• Culture: tests built in one culture may be unfair in another without adaptation and local norms.
• Ethics: informed consent, confidentiality and careful reporting of results.