THE ROUGH GUIDE TO PSYCHOLOGYto explore the extremes of personality, many people will of course
score moderately on one or more factors reflecting subtle differences
in character.
Measuring personality
Personality is usually measured with the use of self-report question-
naires, in which a respondent rates their agreement with a series of
descriptive statements, known as test items (see box on p.184). When
psychologists devise these tests, they make sure that the different items
really do measure what they’re supposed to be measuring – a quality
known as validity. To do this, they will check that an affirmative answer
to one item that’s supposed to measure, say, extroversion, tends to go
hand in hand with affirmative answers to the other items intended to
measure that same factor. The psychologist might also check that scores
on the new test agree with scores on an already established measure.
Finally, they might consider ensuring that the test correlates with other
corroborating evidence, such as the respondent’s diary records or the
verdicts of their friends and family.
Another quality expected of an established personality test is that
it should have reliability. That is, the same or similar score should be
achieved when the same person is tested repeatedly, or when the test
is completed by different people who ought to achieve the same score
because they have the same or similar personality-type. It’s also impor-
tant to establish “norms” for a new personality test, which requires
getting as many people as possible to complete the measure, so that
some sense can be built up as to what constitutes a “normal” or common
score and what constitutes a more extreme or unusual score.
A major drawback of self-report questionnaires of the kind used to
measure personality is many people’s concern to make a good impression
- a phenomenon known as social desirability. After all, who wants to reveal
to an unfamiliar researcher that they are lazy, obnoxious or neurotic?
Of course, where possible, making a test anonymous helps. Also, some
personality tests include “too good to be true” items designed to detect
when social desirability is likely to be distorting the results (see p.264).
More old-fashioned approaches to measuring personality are today
considered to be lacking in reliability and validity. This includes the
Thematic Apperception Test devised by Harvard psychologist Henry
Murray in the 1930s. The TAT requires participants to generate stories
in response to ambiguous picture cards, for example of a woman lying