Projective Tests:
s are special kinds of tests based on the idea that if you give people an ambiguous stimulus and ask them to describe it, their answer will tell you some thing about their personality.The most well-known example is the Rorschach test. Another example is the Thematic Apperception Test (the version used in research now is called the Picture.
Story Exercise:
Psychologists who use the Rorschach test believe people’s descriptions of what they see in ink blots tell something about their personality. The test is linked to psychoanalysis (see Lesson 3). ii. There is little evidence that these tests are valid (see section on validity below)
Objective tests:
Most personality tests are “objective tests,” so-named because they are not projective (not because they are especially objective). These include any tests with yes/no response options, numerical response options, etc. Examples: The Minnesota Multiphasic Personality Inventory (MMPI) and the Narcissistic Personality Inventory (NPI)
The rational method of creating objective tests: Re searchers write items that seem like they would be a good measure of the personality trait they want to measure. Example: Researchers involved in creating the NPI in cluded items they thought would provide clues about the respondent’s level of narcissism
The empirical method of creating objective tests is to write many items, try them all out, and keep the ones that are actually a good measure of the trait the researcher wants to measure. Example: The MMPI was created this way — this is why some of the items on the MMPI (e.g., “Do you prefer baths to showers?”) seem to have nothing to do with what the MMPI is measuring (personality disorders). The items pre dict who has a personality disorder or not, and even if we can’t understand why they do, they are included in the test because they work.
Get Lookh’s stories in your inbox
Join Medium for free to get updates from this writer.
Strength of objective tests:
Personality tests are less susceptible to faking than self-re port questionnaires. It’s harder to change your answers to get a specific score, because it’s not always obvious what the test is measuring or how you should answer to get the score you want.
Weakness of objective tests:
Objective tests require the researcher to interpret your answer, and that interpretation could be wrong (note that here, “interpret” means making empirical associations with outcomes of interest). Thus, interpreting an answer as a sign of narcissism (or anything else) is a little risky because it requires making some assumptions that may or may not be true. Example: Perhaps you agreed with the statement, “If I ruled the world, it would be a better place” not because you think you would be a particularly good world leader, but because you are very disappointed with the current world leaders.
Characteristics of a good personality measure:
Reliability: The consistency of a measurement 1. Reliability does not indicate how accurate or true the measure is. Reliability is about consistency, not about accuracy.
There are different ways to measure reliability:
Test-retest reliability is when you get similar results when you measure someone’s personality using measure X at time 1 and then again at time 2. Example: Imagine an average-sized adult weighing him self on a scale. The first time he steps on the scale, it indicates he weighs 25 pounds. He steps off and 5 min utes later steps on again. Again, it indicates he weighs 25 pounds. This scale is reliable. It’s just not accurate.
internal consistency applies to measures with multiple items (e.g., a self-report questionnaire with 10 items). Internal consistency is the degree to which the various items produce similar scores (i.e., measure the same thing). There are many ways to measure internal consistency, including “split-half reliability,” which refers to the consistency between one half of the items (e.g., all the odd-numbered items) and the other half of the items on the same measure (e.g., all the even-numbered items)
Parallel forms reliability refers to the consistency across the different forms when there are different versions, or “forms,” of a scale. Would people get the same score if given either Form A or Form B? The SAT is an example of a measure with parallel forms. It is important that the SAT have parallel forms reliability, because if a person’s scores were to change a lot depending on which form he or she got, that would be an inconsistent, unreliable measure.
