JobCannonAll Terms
Psychometrics & Testing

Construct Validity

Whether a test actually measures the theoretical construct it claims to measure. Does the test measure the real trait, or something else by coincidence?

Construct validity (APA Standards, 2014) addresses the core question: "Does this test measure what it says it measures?" A Big Five Extraversion test should measure extraversion, not just sociability or talkativeness.

Evidence for construct validity: (1) Factor structure matches theory (5 factors in Big Five), (2) Correlations with related measures are as expected (high Extraversion correlates with self-reported sociability, r=0.50+), (3) Known-groups validity (extraverts score higher than introverts), (4) Relationships with external criteria match theory (Extraversion predicts sales performance as expected).

Constructs can be narrow (Extraversion) or broad (General Intelligence). MBTI has low construct validity for some dimensions (Thinking-Feeling predicts gender better than thinking style). Personality tests with demonstrated construct validity give interpretable, meaningful scores.

Source: American Psychological Association (2014). Standards for Educational and Psychological Testing.

Related Terms

Test-Retest Reliability
A measure of how consistent a test's results are when the same person takes it again. High reliability (0.70+) means you'll get similar results each time.
Validity (Psychometric)
Whether a test measures what it claims to measure. A valid personality test actually predicts real-world outcomes like job performance, relationship satisfaction, or mental health.
Likert Scale
A response format used in personality tests where you rate agreement on a scale (e.g., "Strongly Disagree" to "Strongly Agree"). Named after psychologist Rensis Likert (1932).
Psychometric Test
A standardized, scientifically designed assessment that measures psychological attributes — personality traits, cognitive abilities, aptitudes, or attitudes — in a reliable and valid way.
Cronbach's Alpha
A statistic measuring internal consistency — how closely related a set of test items are as a group. Values above 0.70 indicate acceptable reliability for personality scales.
← Back to Glossary