Psychology › Section K · Personality › Topic K.9
Personality assessment
Personality tests are used to select employees, support court decisions and guide clinical care, so it matters a great deal whether they actually work. This topic looks at the two main families of tests, self-report inventories and projective tests, and at the evidence on how good each one is.
Why measure personality?
Personality is assessed for research, career guidance, job selection, legal cases such as custody disputes, and as one part of clinical assessment. Each use demands a test that meets the standards from the research topic: reliability, meaning the test gives consistent results across time and between scorers, and validity, meaning it measures what it claims and predicts what it should. A test also needs norms, scores from a large, representative comparison group, so that one person's score can be interpreted. Some tests meet all three standards; others remain popular despite weak results.
Self-report inventories
A self-report inventory asks people to answer many statements about themselves, either true or false or on a rating scale from, say, 1 (strongly disagree) to 5 (strongly agree). This rating format is called a Likert scale, after Rensis Likert, who introduced it in 1932. Big Five questionnaires used in research are self-report inventories. Their strengths are clear: they are quick, cheap and objectively scored, and good ones have high reliability and useful validity. Their main weakness is that people may not answer honestly or accurately. Some present themselves in a socially desirable light, especially when a job is at stake, and some simply lack insight into themselves.
The MMPI
The Minnesota Multiphasic Personality Inventory, first published in 1943, is one of the most widely used clinical tests. Its authors chose items by empirical keying: they kept statements that people with particular diagnosed conditions answered differently from comparison participants, whatever the statement seemed to be about. The original comparison group was small and mostly white, rural Minnesotans, so the revised MMPI-2 (1989, 567 true-false items) used a larger, more representative national sample. Shorter versions followed: the MMPI-2-RF (2008, 338 items) and the MMPI-3 (2020, 335 items). It is designed to be given and interpreted by trained professionals as one part of a wider assessment.
Validity scales
The MMPI's most useful feature may be its validity scales, which check how the person approached the test rather than their personality. The L (lie) scale contains statements that almost nobody can honestly endorse, such as never having felt annoyed; agreeing with many of them suggests "faking good," presenting an unrealistically positive picture. Other scales flag answering at random, exaggerating problems ("faking bad") or being unusually defensive. If the validity scales look suspicious, the clinical scores are interpreted with great caution or not at all. This built-in check is one answer to the honesty problem that affects every self-report test.
Projective tests
Projective tests take a completely different route. They present ambiguous material and assume that, in interpreting it, people reveal hidden feelings, motives and conflicts, by projection in Freud's sense. The best known is the Rorschach Inkblot Test, published in 1921 by the Swiss psychiatrist Hermann Rorschach: ten symmetrical inkblots, some coloured, and the question "What might this be?" Another is the Thematic Apperception Test (TAT), developed in the 1930s by Christiana Morgan and Henry Murray, in which people tell stories about ambiguous pictures of people. A third is the Rotter Incomplete Sentences Blank (1950), in which people finish sentence openings such as "I worry about..."
The limits of projective tests
Projective tests are harder to fake than questionnaires, but the scientific case for most of them is weak. Their theory rests on Freud's idea of projection, which itself lacks good support. Scoring is often subjective, so two trained scorers may read the same answers differently, which lowers reliability. The TAT has no single standard way of giving or scoring it. Major reviews have found that most projective scores have weak validity for diagnosing conditions or predicting behaviour. Rorschach scoring systems such as Exner's improved consistency, and a few scores, mainly those measuring disordered thinking, show reasonable validity. But older norms tended to make ordinary people look disturbed.
Worked example: choosing a test for hiring
A company wants to use the Rorschach to choose its managers. Evaluate the plan. 1. Reliability: inkblot scoring depends heavily on the examiner's judgement, so different examiners may disagree about the same applicant. 2. Validity: there is little evidence that Rorschach scores predict how well someone will manage people. 3. Norms: older norms tend to make healthy people look disturbed, which could unfairly reject good candidates. 4. Alternative: a well-validated Big Five inventory, especially its conscientiousness scale, has evidence of predicting job performance, though applicants may try to fake good. 5. Verdict: the Rorschach is a poor choice here, and any test should be combined with interviews and work samples.
Culture and fairness in testing
A test built and normed on one group can misjudge people from another. Researchers found that the traditional TAT pictures, which show mostly white figures, tended to produce shorter, less engaged stories from African American test takers. In response, the psychologist Robert Williams developed a culturally specific picture test, now called the Contemporized-Themes Concerning Blacks Test, with scenes from African American life; it drew longer, more emotionally positive stories. The TEMAS test ("tell me a story," also Spanish for "themes"), developed by Giuseppe Costantino, does the same for Hispanic children and other minority groups. Fair testing also requires checking that questionnaire items mean the same thing across languages and cultures.
Worked example: the online personality report
After a five-minute online quiz, a website tells you: "You have a great need for others to like you, yet you are critical of yourself. You have unused potential." It feels uncannily accurate. Analyse why. 1. Check how specific it is: almost everyone would agree with every sentence. 2. Recall the evidence: in a classic 1949 demonstration, Bertram Forer gave students the same vague profile and they rated it as highly accurate for themselves. This is the Barnum, or Forer, effect. 3. Ask about the quiz's reliability, validity and norms; usually none are published. 4. Verdict: the feeling of accuracy is not evidence. Treat such results as entertainment, never as a diagnosis.
Never a self-diagnosis
Even a well-validated personality test describes where someone falls on normal dimensions; it does not, by itself, diagnose any condition. Clinical tests such as the MMPI belong with trained professionals, who combine them with interviews, history and careful use of validity scales and norms. A free online version of a clinical test, or a trait score read as a label, can mislead badly. If you or someone you know is struggling, the right step is to talk with a trusted adult, a doctor or a qualified mental health professional, not to rely on a quiz result.
Common mistakes to avoid
First, assuming a test that feels accurate must be accurate: the Barnum effect shows otherwise. Second, believing projective tests are scientifically strong because they are hard to fake; most have weak reliability and validity. Third, treating self-report scores as perfectly honest; validity scales exist because people fake good or bad. Fourth, confusing reliability (consistent results) with validity (measuring the right thing). Fifth, using a test normed on one population to judge people from another. Sixth, using any personality test, including online versions of clinical tests, to diagnose yourself or anyone else.
Key takeaways
Good tests need reliability, validity and representative norms. Self-report inventories are quick and objective but open to faking. The MMPI (1943; MMPI-2 1989; MMPI-2-RF 2008; MMPI-3 2020) uses empirical keying and validity scales such as the L scale. Projective tests, including the Rorschach (1921), TAT and Rotter sentences, rest on projection; most have weak reliability and validity, with a few Rorschach scores the main exception. Culturally specific tests such as the C-TCB and TEMAS address bias. The Barnum effect makes vague feedback feel accurate. Tests never diagnose the reader.
Flashcards
Flashcard
Consistent results across time and between scorers
Test your understanding
Question 1 of 5
A job applicant agrees with almost every MMPI statement such as "I have never felt annoyed with anyone." What does the L scale suggest?
Question 2 of 5
What is the main scientific weakness of most projective tests, such as the TAT?
Question 3 of 5
Most students in a class rate the same vague horoscope-style profile as an accurate description of themselves. This demonstrates:
Question 4 of 5
How were the items of the original MMPI chosen?
Question 5 of 5
A student takes a free online quiz and scores high on neuroticism. What is the most appropriate conclusion?
