Briefly describe some methods that have been devised to reduce response sets in personality tests. ...

(Answer to Q. 1)  Answer: A good answer would address at least two of the following:
o A test can structured in such as way that half of the items rely on a high rating to indicate that the responder has the trait and the other half the items rely on a low rating does.
o Some tests include an infrequency scale that would include items so odd or rare that picking even a few of them would indicate that the respondent is likely lying or not taking the test seriously.
o Q-sort is a method that has individual place cards with test items on them into piles that indicate, for example, strongly applies to me etc. It limits the number of times a given response can be used.
o Another solution the student could mention is the construction of a scale in which all the choices look good so there is no self-image advantage in picking items that dont really apply to you.
o Overall, self-image tests are reliant on the honesty and motivation of the subject and generally assume that respondents are willing to be honest about how they perceive themselves, have a certain level of accurate self-awareness, and are motivated to participate in the procedure.

(Answer to Q. 2)  Answer: Students may choose to cover the ease of administration and scoring and the good general reliability and validity as pros of self-report tests. With regard to cons, the emphasis will likely be on problems of faking or response problems. In general, self-report tests are easy to administer, but there is no opportunity to directly observe any behavior on the part of the respondent. They may be ashamed to admit certain behaviors or attitudes and censor themselves with socially desirable responding, acquiescent responding (yea-saying) or reactant responding (nay-saying). Other respondents are lazy or not so conscientious and fall into some pattern such as only checking one extreme or adopting an actual pattern such as 1-2-3, etc. The terms for this are response set or noncontent responding. The student may also mention cultural patterns, such as the tendency of North American students to use extremes and for Asian students to stick to the middle of scales.

(Answer to Q. 3)  Answer: Again, the student has a variety of valid topics that he/or she may decide to apply to this question. The basic issues have to do with the unstructured somewhat open-ended nature of performance-based tests, for example the TAT, and the more structured, limited-response nature of the self-report style, for example the Adjective Check List. It could be noted that self-report used to be called objective and performance-based used to be called projective. It should be mentioned that there is considerable evidence of low validity for performance-based methods. Examples of the structured nature of self-report tests could refer to Likert-type scales and forced-choice scales. The main example of a self-report scale in the textbook is the Machiavellianism Scale.

Examples of performance-based tests mentioned in the textbook include the Rorschach and TAT. The textbook gives examples of Association techniques (Rorschach); Construction techniques (TAT); Completion techniques (sentence completion); Arrangement (pick your favorite color); and Expression techniques (puppet play).

(Answer to Q. 4)  Answer: It is not necessary for the student to cover all the points made in the textbook about this test, but a good answer will cover several of the following points:
o It is reliable as demonstrated by a high Cronbachs alpha (.56-.81) for each of its factors and facets.
o It has a high test-retest reliability over a three-month period.
o With regard to validity, its main factors account for close to 60 of variability.
o Criterion validity was tested when, for example, the test was shown to produce low scores on Agreeableness when given to drug abusers.
o Convergent and discriminant validity was demonstrated by correlating scores on the NEO-PI-R with other valid personality tests such as the Extraversion scales of the Eysenck Personality Inventory.
o Test generalizability was indicated by its usefulness in various populations and clinical settings.

(Answer to Q. 5)  Answer: The student has a fair amount of latitude with this question but necessary elements for a good answer would include definitions of each concept and at least one example of types of reliability and validity and some mention of methods to assess them. The main point of similarity is that they are both necessary for a useable test, but that while a test can be reliable without being valid it is impossible for a test to be valid if it is not reliable. Reliability concerns a tests consistency across time, items and raters and concerns temporal consistency reliability, test-retest reliability, internal consistency reliability, etc. Mention of Cronbachs alpha as the most widely used indicator of reliability and that a range of .70 to .80 is generally acceptable would be good.

Validity concerns the basic issue of whether or not a test measures what it is designed to test and involves such issues as construct validity, face validity, criterion validity, convergent validity, and divergent validity. Providing an example of how one of these types of validity is tested would be good. For example, comparing the results of the test in question to some external standard or behavior tests criterion validity.

(Answer to Q. 6)  Answer: b: Agreeableness

Thank you for answering correctly!