Objective Personality Tests

Response faking

*impression* management: a person tries to create a certain impression by their responses *faking*: deliberate attempt to create a certain impression (fake good/fake bad-malingering) basic problem: disentangling true personality traits from response sets

Five-factor model

-*5* higher-order traits & *30* lower order facets -facets are highly correlated within each factor, but not across factors

Response sets/responses styles

-*set/style*:a person's tendency, conscious or unconscious, to respond to items in a *certain way*, independent of the person's true feelings about the items -*yea*-sayer vs. *nay*-sayer -*distortion*: a person's true feelings are distorted in some way by the response set

Criterion-Keying Approach

-Items are selected strictly in terms of their *ability to discriminate* between two well-defined groups of examinees

Theory-driven approach

-References a *specific* personality theory (items created to reflect the theory)

Scores of Piers-Harris

-Total score -*6* domain scores (derived by factor analysis) -*inconsistency* index: derived from comparing responses on 15 pairs of related items -*response bias* index: based on a simple count of all yes responses (range from 0 -> 60): low scores nay-saying, high scores yay-saying

Criterion-keying pros and cons

-advantages: direct and simple which encourages new research, focuses on what we want the test to do -disadvantages: theoretical orientation limits the *generalizability* of score interpretations, applicable when groups well-defined, always *overlap* in group distributions

Content method pros and cons

-advantages: simple, easy to generate items -disadvantage: high *face validity* -> subject to faking good/bad

Content method

-develop test items/scales based on simple straightforward understanding of what you want to measure

Four major uses of objective personality tests

1. *Clinical*- neuropsychological assessment 2. *Counseling*- couples therapy 3. *Personnel selection*- identify traits predictive success or to identify problem traits 4. *Research*- structure of personality, reliability/validity for different populations

Four strategies for dealing with response sets and faking

1. Checking responses to items with *extreme* empirical frequencies for normal groups (<10% of a certain answer means someone is faking good/bad) 2. Checking for response *consistency* on same or similar items 3. Using *forced-choice* method for items matched on relevant variables -resulting separate scores called validity indexes 4. Balancing the direction of items

5 Approaches to objective personality test development

1. Content method 2. Criterion-keying 3. Factor analysis 4. Theory-driven approach 5. Combinations

5 trends in objective personality tests

1. Many new tests are being published 2. Methods of development have matured (dealing with response sets/faking, demonstrating test validity empirically through the use of criterion-keying and research on group differences 3. Managed care *prefers* objective test due to convenience and lower cost 4. *Narrative* reports are now common 5. *Online* administration

Specific domain tests: 6 common characteristics

1. number of items: few (20-80) 2. administration time: brief (*10-15* minutes) 3. number of scores: few, often 1 4. applications: narrow, targeted 5. norm groups: very limited 6. scoring/reports: simple

Comprehensive inventories: 6 common characteristics

1. number of items: large 2. administration time: 30-90 minutes 3. number of scores: many 4. applications: wide variety 5. norm groups: good 6. reports: elaborate

Piers-Harris Statistics

Norms: representative of US population by sex, race, geographic region, and household Reliability: -internal consistency: hella, domain score: good Validity: -median intercorrelation among domains is .64- too high for domains purported to be relatively independent

Objective Personality Tests

Objective-> test items can be scored objectively (inventory, structured) 1. Selected response format: M/C, T/F, Likert, 2. Simple item stems: usually one-sentence statements

Factor Analysis pros and cons

Pros: bring *order* to undifferentiated mass of items Cons: final results depend on the content of the initial pool of items, does not yield a *definitive* set of factors

Piers-Harris Children's Self-concept scale

Self-concept: an organized view or ourselves (way of representing information about the self) -most widely used measure of self-concept for ages 7-18 years -structure: *60 item self-report scale* (examinees respond about themselves in yes-no format)

Factor Analysis

identify the *dimensions* underlying a large number of items in a personality measure - factors are identified by examining the *correlations* among all the items

Trait theories of personality

personality: emotional, cognitive, and behavioral tendencies that constitute underlying personality dimensions -can be measured by asking other to rate a person or self-report (most common) -Allport: 18,000 traits -Cattell: 16 traits -Big Five Factors found by factor analysis

theory-driven approach pros and cons

pros: provides an *operational* definition of the theory, a good test of a good theory can have great utility cons: the test's utility is limited by the theory's *validity*, how well does the test actually *reflect* the theory