Overview and Objective Methods
Personality: an individual's unique constellation of psychological traits and states.
Trait versus State
Any distinguishable, relatively enduring way in which one individual varies from another.
Labeling of a trait depends on both the behavior itself and the context in which the behavior appeared.
State versus Trait
Personality states involve the transitory exhibition of some personality trait; a relatively temporary predisposition.
Measuring personality states amounts to a search for and an assessment of the strength of traits that are relatively transitory in nature and/or fairly situation-specific.
A personality type involves a constellation of traits and states that is similar in pattern to one identified category of personality within a classification of personalities.
Personality types are only descriptions of people -- not something presumed inherent in them
Personality assessment involves the measurement of traits and states.
Why measure personality?
- To identify clients' problem areas.
- To identify strengths and weaknesses for occupational decisions.
- To identify non-academic factors that may be acting to impede academic performance.
- To identify psychopathology.
Why Measure Personality?
- In the area of psychological research:
- studies of individual psychological variables.
- studies of psychological theories.
Bases of Personality Assessment
Construction based on theories of personality or empirical methods.
Basis in self-report, behavioral observation, measure of verbal, nonverbal, and physiological responses.
Paper-and-pencil-type tests may be objective or projective.
Advantages of Objective Measures of Personality:
- Answers can be answered rather quickly.
- The items require little explanation.
- The items are quick and reliable to score by machine, by hand, or by computer.
Approaches to the Development of Objective Measures of Personality
Logical or Content Test Construction
Personality inventory comprises items that logically, intuitively, or rationally seem to belong in the test.
Appeals to test takers, since content is straightforward and directly related to the objective of the test.
Test takers feel more in control of the information they are revealing.
Woodworth Psychoneurotic Inventory (WPI) was the first effort to measure personality.
- The WPI screened World War I recruits for personality and adjustment problems.
- Test items included: self-report of fears, sleep disorders, and other "neurotic" disorders.
Mooney Problem Checklist items were taken from 4,000 high school students, counseling interviews, and reviews of clinical records.
- The MPC items relate to emotional functioning in some of the following areas: home & family, boy/girl relations, courtship and marriage, religion, school, health and so on.
- Students were asked to underline items that were important to them and circle those of most concern.
Factor Analysis Test Construction
Factor analysis is a data reduction method.
Factor analysis is used as an aid in personality research designed to identify major traits, cognitive styles, and other personality factors.
Also, factor analysis is used to create tests to measure personality factors.
The 16PF (i.e., the Sixteen Personality Factor Questionnaire) was developed by Raymond Bernard Cattell--using factor analysis.
- He wanted to conduct a personality test that was designed to assess all personality traits.
- He found that 18,000 traits were previously known and out of all of these only a few were "real traits."
- After many test tryouts, the list was reduced to the present day 16. The actual test has 185 items.
- The Sixteen Personality Factor Questionnaire is constantly being revised to fit the needs of new test takers.
Many researchers argue that less factors can be used to determine personality.
The NEO -PI-R is a measure of the Big Five personality traits; it was developed by Paul Costa Jr. and Robert McCrae.
- The NEO-PI-R is a factor analytically derived self-report scale designed to test five personality factors.
- Studies of children, workers, and spouses show that the NEO -PI-R is useful in describing personality.
- NEO-PI Personality factors
- Neuroticism-- includes "worrying vs calm".
- Extroversion-- includes dimensions of "loner vs joiner".
- Openness to Experience-- includes "original vs conventional".
- Conscientiousness-- includes "laziness vs hardworking".
- Agreeableness-- includes "lenient vs critical".
Test Construction by Empirical Criterion Keying
In this method, one administers the test items to at least two groups of people:
- a criterion group composed of people that possess the trait.
- a control group of people who are assumed to not possess the trait.
Items that discriminate between the criterion and control groups are retained.
Items that fail to discriminate between the two are discarded.
This technique is referred to as empirical, because only those items that demonstrate an actual empirical relationship between the test item and the trait are retained.
This technique is called criterion keying, because each item of the test is keyed to a criterion.
The original MMPI - a "Multiphasic" personality inventory.
- Starke Hathaway & John McKinley
- The MMPI is an aid in assessing the mental health of patients seen in medical practice
- The MMPI consists of 550 statements (True/false response).
- The original MMPI statements are printed on cards (the examinee sorts them into true/false and cannot say).
- Eight MMPI clinical scales were developed from eight clinical groups of psychiatric inpatients.
- These include: hypochondriasis, depression, hysteria, paranoia, schizophrenia, social introversion, psychasthenia, and hypomania.
- The MMPI can be used with people 16 or older, with at least a 6th grade education.
- Research for the MMPI involved the study of psychiatric textbooks, reports and previously published personality test items
- Test items reflected 26 content categories including: general health, family issues, religious attitudes, and sexual identity
- The MMPI items were presented to control and criterion groups.
- The MMPI norm control group consisted of 1500 subjects taken from hospital visitors, normal clients and local WPA workers.
- There are 3 validity scales:
- The L scale or lie scale: if the score on this scale does not fall within certain limits, questions will be raised about the test taker.
- The F scale or Infrequency scale: 64 items endorsed by members of non-psychiatric.
An elevated F score may mean the person did not take the test seriously (a high F scale may mean "faking bad").
- The K scale or correction scale is a reflection of the frankness of the test taker's self-report.
Elevated K scores may be associated with defensiveness and a favorable impression (a low score may be excessive criticism).
- "Can not say" scale is also used with the MMPI: this is simply a frequency count of the number of items that were answered with a "can not say "response.
- The tests validity comes into question when this score is above 30.
- Test can be computer scored or computer interpreted.
- MMPI Raw scores are converted into T scores, with a mean of 50 and SD of 10.
- Scores of 70 or above indicate a problem for more in-depth studies.
- Over 400 new MMPI scales have been devised.
- Researchers have examined and compared not only the responses of normals but also members of more offbeat groups:
- Examples: snake handling religious cults, castrated males, and submarine school dropouts.
There are many criticism of the MMPI:
- One is the limitations of its construction and use.
- The normative sample is said to have been deficit in size and in the representatives of the general population.
- The age of the norms is a problem.
In 1983 a new set of MMPI norms were set for normal adult responses.
These norms were taken from 1,408 adults ages 18-99, living in the same area as the original sample.
The results of the revision include: people in the 80s have more elevated MMPI profiles then of those in the 40s.
There are two explanations of this: people in their 80s may be under more psychological and physical stress; and changes in societal mores and perceptions.
Also, the same items are used for different scales, resulting in some scales being highly correlated.
Other MMPI problems include:
- Computer scoring which is limited due to its availability.
- The offensiveness of some of the questions.
- The length of the test.
The "Minnesota Multiphasic Personality Inventory-2" (MMPI-2)
A revised edition of the MMPI took place in August of 1989.
This new edition changed approximately 14% of the original items.
- Grammatical errors.
- Language to be more contemporary, nonsexist, and readable.
The newly added items addressed such topics as
- drug abuse.
- suicide potential.
- "Type A" behavior patterns.
- marital adjustment.
- attitudes toward work.
394 items identical with the original MMPI.
66 items that were modified or rewritten, 107 new items.
The MMPI-2 also contained the original ten clinical scales from the MMPI.
These scales included:
- The original 10 scales:
- Psychopathic deviate.
- Social introversion.
- The MMPI-2 added a new set of content scales. These include:
- Health concerns.
- Social discomfort.
- Family Problems.
- Work interference.
- Antisocial practices.
Comparison of MMPI and MMPI-2
566 total items
16 repeated items
4 validity scales
10 clinical scales
Standardized on 724 individuals with mean educational of 8th grade.
Hand or computer
567 total items
no repeated items
7 validity scales
10 clinical scales
Standardized on 2,600 individuals with mean education of 13 years
Representative of gender, ethnicity, and SOE level
Hand or computer
The Standardization Sample of the MMPI-2
- 1,462 females and 1,138 males
- They were matched to 1980 U.S. census data variables of age, gender, minority status, social class, and education.
- 81% were white and 19% non-white.
- The age range: 18 to 85 years.
- Formal education ranged from 3 to 20+ years.
- Median family income for females was $25,000 to $30,000.
- Median family income for males was $30,000 to $35,000.
- The Standardization Sample also included airline pilot applicants, people in chronic pain, college students, psychiatric inpatients, and couples in marital counseling.
Psychometric properties of the MMPI-2
No alternate form of the MMPI-2 (as opposed to the MMPI) was published.
Objective scoring minimized the impact of error variance due to rater differences.
Therefore, inter-rater reliability is not a concern.
MMPI-2 test-retest reliability
- Over a one-week period, test-retest reliability ranged from .58 to .92 (with a median value of .81).
- Test-retest for content scales over a 9 day period were .78 to .91, with a median value of .85.
- Test-retest for supplementary scales from 3 weeks to 4 months produced mostly coefficients in the .80s.
- Most of the MMPI-2 scales have test-retest reliability estimates in acceptable ranges.
Internal consistency reliability
- Was not something the original developers of the MMPI attempted to build into the test.
- Internal consistency reliability tends to be low for the MMPI-2 scales (with a median of .63).
- In the supplementary scales, internal consistency was relatively low.
An evaluation of the MMPI-2
- Reliability is acceptable for the majority of the scales.
- The validity of the original MMPI is impressive and this evidence is generalizable to the MMPI-2.
- Much remains to be done to demonstrate the merits of the MMPI-2 with criterion-related and construct validity.
- The original MMPI explored the test's efficacy with regard to populations at both ends of the normal-abnormal continuum.
- The trend in MMPI-2 research, with some exceptions, seems to be more focused on the abnormal end of the spectrum.
- The original MMPI was standardized on Caucasians, but the MMPI-2 used a broader normative sample.
- Research supports the adequacy of the MMPI-2 and its new norms for African American and Hispanic American individuals.
- Work has begun on translating the test for use among Spanish-speaking Mexicans.
- The validity of the new content scales.
- The utility of treatment recommendations made on the basis of MMPI-2 scores.
- Heterogeneity of item content within the scales remains a problem.
- Overlapping items across the different scales.
- Atheoretical nature of its development.
There was interest in using the MMPI with adolescents.
Separate adolescent norms for the original MMPI had been developed, and the developers of the MMPI-2 would have had to gather normative data on adolescents for the revision as well.
Instead of doing that, a version of the MMPI for adolescents was developed.
This new test, the MMPI-A ("A" for Adolescent) was published in 1992.
Developed for individuals 14 to 18 years of age who could read at the 6th grade level.
Contains 478 true-false items, making it somewhat shorter than the MMPI and the MMPI-2.
The normative data are based on 1,620 adolescents in several geographic regions, selected specifically to include ethnically diverse areas.
Contains all of the same clinical and validity scales as the MMPI-2.
The Theoretical Approach to Test Construction
In reaction to the widespread popularity of computerized personality testing, there is a need for more clinical and more theoretically based approaches to personality assessment.
It has been argued that personality theory as applied in personality assessment serves an organizing function that allows for prediction.
Instruments used in personality testing and assessment range from what we might term "theory-saturated" to relatively atheoretical (i.e., no theoretical foundation).
An example of a theory-saturated instrument is "The Blacky Pictures Test."
Clinical Versus Actuarial Prediction
Two approaches to interpreting data from personality tests
Actuarial Approach(Statistical Approach):
- evaluation of data according to preset rules.
- reliance on statistical/empirical procedures.
- no preset rules.
- interpretation of test data is made from the evaluator's judgement.
- reliance on evaluator's experience.
Actuarial vs. Clinical Approach
The actuarial approach (or statistical approach) tends to be superior to the clinical approach in prediction.
However, few studies focused on the role of the clinician in the data-gathering process.
In practice, clinicians seldom gather data using a rigorous set of rules.
Data tends to be gathered with clinician's own perception.
It is possible that in some types of situations, the clinical data-gathering process employed by some clinicians--as well as the subsequent predictions made as a result of that process--may prove superior to any comparable actuarial process.