R. Ch 8 - Appraisal - TM
601. Appraisal can be defined as a. the process of assessing or estimating attributes. b. testing which is always performed in a group setting. c. testing which is always performed on a single individual. d. a pencil and paper measurement of assessing attributes.
A
634. Which method of reliability testing would be useful with an essay test but not with a test of algebra problems? a. test-retest. b. alternate forms. c. split-half. d. interrater/interobserver.
D
641. _______ did research and concluded that intelligence was normally distributed like height or weight and that it was primarily genetic. a. Spearman. b. Guilford. c. Williamson. d. Francis Galton.
D
645. The first intelligence test was created by a. David Wechsler. b. J. P. Guilford. c. Francis Galton. d. Alfred Binet and Theodore Simon.
D
647. IQ stands for intelligence quotient, which is expressed by a. CA/MA x 100. b. CA/MA x 100. c. MA/CA x 50. d. MA/CA x 100.
D
660. The MMPI-2 is a. an IQ test. b. a neurological test. c. a projective personality test. d. a standardized personality test.
D
674. The NCE is a. an intelligence test. b. an aptitude test. c. a personality test. d. an achievement test.
D
678. Your supervisor wants you to find a new personality test for your counseling agency. You should read a. professional journals. b. the Buros Mental Measurements Yearbook. c. classic textbooks in the field as well as test materials produced by the testing company. d. all of the above.
D
692. _______ would be an informal method of appraisal. a. IQ testing. b. Standardized personality testing. c. GRE scores. d. A checklist.
D
602. A test can be defined as a systematic method of measuring a sample of behavior. Test format refers to the manner in which test items are presented. The format of an essay test is considered a(n) _______ format. a. subjective. b. objective. c. very precise. d. concise.
A
606. The _______ index indicates the percentage of individuals who answered each item correctly. a. difficulty. b. critical. c. intelligence. d. personal.
A
614. An achievement test measures maximum performance while a personality test or interest inventory measures a. typical performance. b. minimum performance. c. unconscious traits. d. self-esteem by always relying on a Q-Sort design.
A
617. A test battery is considered a. a horizontal test. b. a vertical test. c. a valid test. d. a reliable test.
A
621. In the field of testing, validity refers to a. whether the test really measures what it purports to measure. b. whether the same test gives consistent measurement. c. the degree of cultural bias in a test. d. the fact that numerous tests measure the same traits.
A
622. A counselor peruses a testing catalog in search of a test which will repeatedly give consistent results. The counselor a. is interested in reliability. b. is interested in validity. c. is looking for information which is not available. d. is magnifying an unimportant issue.
A
625. Face validity refers to the extent that a test a. looks or appears to measure the intended attribute. b. measures a theoretical construct. c. appears to be constructed in an artistic fashion. d. can be compared to job performance.
A
631. One method of testing reliability is to give the same test to the same group of people two times and then correlate the scores. This is called a. test-retest reliability. b. equivalent forms reliability. c. alternate forms reliability. d. the split-half method.
A
643. J. P. Guilford isolated 120 factors which added up to intelligence. He also is remembered for his a. thoughts on convergent and divergent thinking. b. work on cognitive therapy. c. work on behavior therapy. d. work to create the first standardized IQ test.
A
654. The best intelligence test for a kindergartner would be the a. WPPSI-III. b. WAIS-III. c. WISC-IV. d. Myers-Briggs Type Indicator.
A
656. Group IQ tests like the Otis Lennon, the Lorge-Thorndike, and the California Test of Mental Abilities are popular in school settings. The advantage is that a. group tests are quicker to administer. b. group tests are superior in terms of predicting school performance. c. group tests always have a higher degree of reliability. d. individual IQ tests are not appropriate for school children.
A
658. In a culture-fair test a. items are known to the subject regardless of his or her culture. b. the test is not standardized. c. culture-free items cannot be utilized. d. African Americans generally score higher than Whites.
A
668. Test bias primarily results from a. a test being normed solely on White middle-class clients. b. the use of projective measures. c. using Whites to score the test. d. using IQ rather than personality tests.
A
669. A counselor who fears the client has an organic, neurological, or motoric difficulty would most likely use the a. Bender Gestalt. b. Rorschach. c. Minnesota Multiphasic Personality Inventory. d. Thematic Apperception Test.
A
672. Interest inventories are positive in the sense that a. they are reliable and not threatening to the test taker. b. they are always graded by the test taker. c. they require little or no reading skills. d. they have high validity in nearly all age brackets.
A
675. The _______ are examples of aptitude tests. a. GATB, the O*NET Ability Profiler, and the MCAT. b. GZTS and the MMPI. c. CPI and the MMPI. d. Strong and the LSAT.
A
689. One major testing trend is a. computer-assisted testing and computer interpretations. b. more paper and pencil measures. c. to give school children at least three IQ tests per year. d. to train pastoral counselors to do projective testing.
A
693. The WAIS-III is given to 100,000 individuals in the United States who are picked at random. A counselor would expect that a. approximately 68% would score between 85 and 115. b. approximately 68% would score between 70 and 130. c. the mean IQ would be 112. d. 50% of those tested would score 112 or above.
A I know, I see it too, a question with numbers—lots of numbers, but don't panic. Let's walk through this one together. First, the Wechsler IQ test has been administered to a very large group of people so chances are the distribution of scores will be normal. This tells you that the mean score will be 100 (i.e., the average IQ) and the standard deviation will be 15 (if the question were asked about the Binet you'd use 16 as the standard deviation). In a normal distribution approximately 68% of the population will fall between +/-1 standard deviation of the mean. With a standard deviation of 15 you simply subtract 15 from 100 to get the low score (i.e., 85) and add 15 to 100 to get 115. Choice "b" would be correct if the 68% was changed to 95%, since about 95% of the people in a normal distribution fall between +/-2 standard deviations of the mean. (You simply subtract 30 from 100 to get 70 and add 30 to 100 to yield the upper IQ score of 130.) Keep in mind that choice "c" should read 100 while choice "d" ought indicate that 50% would score above 100
690. One future trend which seems contradictory is that some experts are pushing for a. a greater reliance on tests while others want to rely on them less. b. social workers to do most of the testing. c. psychiatrists to do most of the testing. d. counselors to ban all computer-assisted tests.
A It seems we counselors just can't agree on anything. Many counselors would like to see a greater emphasis in the future on tests which assess creative and motivational factors.
650. Today the Stanford-Binet is used from ages 2 to adulthood. The IQ formula has been replaced by the a. SAS. b. SUDS. c. entropy. d. ACPA.
A SAS =. Standard Age Score
684. Counselors often shy away from self-reports since a. clients often give inaccurate answers. b. ACA ethics do not allow them. c. clients need a very high IQ to understand them. d. they are generally very lengthy.
A Say a client is monitoring her behavior and does not wish to disappoint her therapist. The report could be biased. This is a "reactive effect" of the self-monitoring.
637. A researcher working with a personality test discovers that the test has a reliability coefficient of .70 which is somewhat typical. This indicates that a. 70% of the score is accurate while 30% is inaccurate. b. 30% of the people who are tested will receive accurate scores. c. 70% of the people who are tested will receive accurate scores. d. 30% of the score is accurate while 70% is inaccurate.
A Seventy percent of the obtained score on the test represented the true score on the personality attribute, while 30% of the obtained score could be accounted for by error. Seventy percent is true variance while 30% constitutes error variance.
642. Francis Galton felt intelligence was a. a unitary faculty. b. best explained via a two factor theory. c. best explained via the person's environment. d. fl uid and crystallized in nature.
A Sir Francis Galton of England has been recognized as one of the major pioneers in the study of individual differences. A half- cousin of Charles "Origin of Species" Darwin, he believed that exceptional mental abilities were genetic and ran in families, and said just that in his 1869 work Hereditary Genius.
663. The 16 PF refl ects the work of a. Raymond B. Cattell. b. Carl Jung. c. James McKeen Cattell. d. Oscar K. Buros.
A Tests and inventories like the 16 PF, which analyze data outside of a given theory, are called factor-analytic tests or inventories rather than theory-based tests.
626. A job test which predicted future performance on a job very well would a. have high criterion/predictive validity. b. have excellent face validity. c. have excellent construct validity. d. not have incremental validity or synthetic validity.
A Choice "d" introduces you to the terms incremental validity and synthetic validity. Although incremental validity and synthetic validity are not considered two of the five or six major types of validity, don't be too surprised if they pop up on an advanced exam question. The term incremental validity has been used to describe a number of testing phenomena. First and foremost, incremental validity has been used to describe the process by which a test is refined and becomes more valid as contradictory items are dropped. Incremental validity also refers to a test's ability to improve predictions when compared to existing measures that purport to facilitate selection in business or educational settings. When a test has incremental validity, it provides you with additional valid information that was not attainable via other procedures. Synthetic validity is derived from the word synthesized. Synthetic validity was popularized by industrial organizational (I/O) psychologists who felt the procedure had merit, especially when utilized for smaller firms who did not hire a large number of workers. In synthetic validity, the helper or researcher looks for tests that have been shown to predict each job element or component (e.g., typing, filing, etc.). Tests that predict each component (criterion) can then be combined to improve the selection process.
679. The standard error of measurement tells you a. how accurate or inaccurate a test score is. b. what population responds best to the test. c. the accuracy for personality but not IQ tests. d. the number of people used in norming the test.
A If a client decided to take the same test over and over and over again you could plot a distribution of scores. This would be the standard error of measurement for the instrument in question. Suffice it to say, the lower the better. A low standard error means high reliability. Social loafing describes a phenomenon in which a person in a group puts forth less effort than if he or she were attempting to accomplish the same goal individually.
627. A new IQ test which yielded results nearly identical to other standardized measures would be said to have a. good concurrent validity. b. good face validity. c. superb internal consistency. d. all of the above.
A - You also should be familiar with the terms convergent and discriminant validity. These terms relate to both criterion validity and construct validity. The relationship or correlation of a test to an independent measure or trait is known as convergent validity. Convergent validity is actually a method used to assess a test's construct/ criterion validity by correlating test scores with an outside source. Say, for example, that a measure purports to measure phobic responses. A client, who has a snake phobia, is then exposed to a snake and experiences extreme panic. If the client scores higher on the test than he would in a relaxed state, then this would display convergent validity. The test also should show discriminant validity. This means the test will not reflect unrelated variables. Hence, if phobias are unrelated to IQ, then when one correlates clients' IQ scores to their scores on the test for phobias, this should produce a near zero correlation. Similarly, if discriminant validity is evident, a counselor who is genuinely qualified to sit for a state licensing exam should score higher on the exam than a student who flunked an introductory counseling course.
603. The National Counselor Exam (NCE) is a(n) _______ test because the scoring procedure is specific. a. subjective. b. objective. c. projective. d. subtest.
B
605. The NCE is a(n) _______ test. a. free choice. b. forced choice. c. projective. d. intelligence.
B
610. A client who takes a normative test a. cannot legitimately be compared to others who have taken the test. b. can legitimately be compared to others who have taken the test. c. could not have taken an IQ test. d. could not have taken a personality test.
B
611. In an ipsative measure the person taking the test must compare items to one another. The result is that a. an ipsative measure cannot be utilized for career guidance. b. you cannot legitimately compare two or more people who have taken an ipsative test. c. an ipsative measure is never valid. d. an ipsative measure is never reliable.
B
630. A valid test is _______ reliable. a. not always. b. always. c. never. d. 80%.
B
636. An excellent psychological or counseling test would have a reliability coefficient of a. 50 b. 0.90 c. 1.00 d. -0.90
B
646. Today, the Stanford-Binet IQ test is a. a nonstandardized measure. b. a standardized measure. c. a projective measure. d. b and c.
B
655. The mean on the Wechsler and the Binet is _______ and the standard deviation is _______. a. 100; 100. b. 100; 15 Wechsler, 16 Stanford-Binet. c. 100; 20. d. 100; 1.
B
664. The Myers-Briggs Type Indicator refl ects the work of a. Raymond B. Cattell. b. Carl Jung. c. William Glasser. d. Oscar K. Buros.
B
666. An aptitude test is to _______ as an achievement test is to ____ ___. a. what has been learned; potential. b. potential; what has been learned. c. profit from learning; potential. d. a measurement of current skills; potential.
B
676. One problem with interest inventories is that the person often tries to answer the questions in a socially acceptable manner. Psychometricians call this response style phenomenon a. standard error. b. social desirability (the right way to feel in society). c. cultural bias. d. acquiescence.
B
696. A good practice for counselors is to a. always test the client yourself rather than referring the client for testing. b. never generalize on the basis of a single test score. c. stay away from culture-free tests. d. stay away from scoring the test yourself.
B Also, although choice "c" represents an ideal measure, most experts believe that as of this date no such animal exists
632. One method of testing reliability is to give the same population alternate forms of the identical test. Each form will have the same psychometric/statistical properties as the original instrument. This is known as a. test-retest reliability. b. equivalent or alternate forms reliability. c. the split-half method. d. internal consistency.
B Counterbalancing is necessary when testing reliability in this fashion. That is to say, half of the individuals get parallel form A first and half get form B initially. This controls for variables such as fatigue, practice, and motivation.
682. A counselor can utilize psychological tests to help secure a ____ ___ diagnosis if third party payments are necessary. a. AACD. b. DSM or ICD. c. percentile. d. standard error.
B Diagnosis is a medical term which asserts that you classify a disease based on symptomatology. CPT (Current Procedural Terminology Codes) are used to let insurance companies, managed care firms, etc. know which service you provided, such as individual therapy or family therapy.
685. In most instances, who would be the best qualified to give the Rorschach Inkblot Test? a. A counselor with NCC, NCCC after his or her name. b. A clinical psychologist. c. A D.O. psychiatrist. d. A social worker with ACSW after his or her name.
B Generally, a clinical psychologist would have the most training in projective measures while the social worker would have the least education regarding tests and measurements.
661. The word psychometric means a. a form of measurement used by a neurologist. b. any form of mental testing. c. a mental trait which cannot be measured. d. the test relies on a summated or linear rating scale.
B Psychometrics literally refers to the branch of counseling or psychology which focuses on testing. Choice "d" is used to describe answer scales in which various values are given to different responses. For example, on a Likert Scale a "strongly agree" might be given a 5, yet an "agree response" might be rated a 4. The clients score is the "sum" of all the items.
652. The best IQ test for a 22-year-old single male would be the a. WPPSI-III. b. WAIS-III. c. WISC-IV. d. Computer-based testing.
B Quick hints for dealing with WAIS-IV questions: • The test is based on neurocognitive research and the Cattell-Horn-Carroll leading theory of human intelligence. • It can be administered and scored online. • The exam takes 60 to 90 minutes to complete. • When compared to the previous version of the exam, object assembly and picture arrangement have been dropped. • Ten subject areas, also called subtests on some exams (with a mean of 10 and a standard deviation of 3) make up four index scores: verbal comprehensive index (VCI), perceptual reasoning index (PRI), working memory index (WMI), and processing speed index (SPI). • FSIQ merely stands for full scale IQ. FSIQ and indexes sport a mean of 100 with a standard deviation of 15. • Less emphasis than the previous version on crystallized intelligence. • Can measure IQ from 40 to 160. Since the Stanford- Binet 5 has a wider range (e.g., it can measure an IQ up to 180) it would be a better instrument than the Wechsler for measuring extremely low IQs or giftedness.
673. A counselor who had an interest primarily in testing would most likely be a member of a. HS-BCP. b. AARC. c. NASW. d. ACES. a. ASGW. - Association for Specialists in Group Work B. Association for Assessment and Research in Counseling b. AMECD. - Association for Measurement and Evaluation in Counseling and Development c. NASW. - National Association of Social Workers
B The AARC (Association for Assessment and Research in Counseling) is one of 20 ACA divisions. Can you name the other choices?
695. Infant IQ tests are a. more reliable than those given later in life. b. more unreliable than those given later in life. c. not related to learning experiences. d. never used.
B These "toddler tests" are sometimes capable of picking up gross abnormalities such as severe intellectual disabilities.
683. A colleague of yours invents a new projective test. Seventeen counselors rated the same client using the measure and came up with nearly identical assessments. This would indicate a. high validity. b. high reliability. c. excellent norming studies. d. culture fairness.
B This is known as "inter-rater" reliability.
633. A counselor doing research decided to split a standardized test in half by using the even items as one test and the odd items as a second test and then correlating them. The counselor a. used an invalid procedure to test reliability. b. was testing reliability via the split-half method. c. was testing reliability via the equivalent forms method. d. was testing reliability via the inter-rater method.
B When a researcher does not use even versus odd questions to split the test, he or she may do so using random numbers (merely dividing a test according to first half versus second half could confound the data due to practice and fatigue effects).
607. Short answer tests and projective measures utilize free response items. The NCE and the CPCE uses forced choice or so-called _______ items. a. vague. b. subjective. c. recognition. d. numerical.
C
608. A true/false test has _______ recognition items. a. similar. b. free choice. c. dichotomous. d. no.
C
612. Tests are often classified as speed tests versus power tests. A timed typing test used to hire secretaries would be a. a power test. b. neither a speed test nor a power test. c. a speed test. d. a fine example of an ipsative measure.
C
613. A counseling test consists of 300 forced response items. The person taking the test can take as long as he or she wants to answer the questions. a. This is most likely a projective measure. b. This is most likely a speed test. c. This is most likely a power test. d. This is most likely an invalid measure.
C
616. In a cyclical test a. the items get progressively easier. b. the difficulty of the items remains constant. c. you have several sections which are spiral in nature. d. the client must answer each question in a specified period of time.
C
619. The most critical factors in test selection are a. the length of the test and the number of people who took the test in the norming process. b. horizontal versus vertical. c. validity and reliability. d. spiral versus cyclical format.
C
620. Which is more important, validity or reliability? a. Reliability. b. They are equally important. c. Validity. d. It depends on the test in question.
C
624. Construct validity refers to the extent that a test measures an abstract trait or psychological notion. An example would be a. height. b. weight. c. ego strength. d. the ability to name all men who have served as U.S. presidents.
C
629. A reliable test is _______ valid. a. always. b. 90%. c. not always. d. 80%.
C
635. A reliability coefficient of 1.00 indicates a. a lot of variance in the test. b. a score with a high level of error. c. a perfect score which has no error. d. a typical correlation on most psychological and counseling tests.
C
649. Simon and Binet pioneered the first IQ test around 1905. The test was created to a. assess high school seniors in America. b. assess U.S. military recruits. c. discriminate normal from retarded Parisian children. d. measure genius in the college population.
C
651. Most experts would agree that the Wechsler IQ tests gained popularity, as the Binet a. must be administered in a group. b. favored the geriatric population. c. didn't seem to be the best test for adults. d. was biased toward women.
C
653. The best intelligence test for a sixth-grade girl would be the a. WPPSI-III. b. WAIS-III. c. WISC-IV. d. Merrill-Palmer.
C
657. The group IQ test movement began a. in 1905. b. with the work of Binet. c. with the Army Alpha and Army Beta in World War I. d. with the AGCT in World War II.
C
648. The Binet stressed age-related tasks. Utilizing this method, a 9- year-old task would be one which a. only a 10-year-old child could answer. b. only an 8-year-old child could answer. c. 50% of the 9-year-olds could answer correctly. d. 75% of the 9-year-olds could answer correctly.
C A 9-year-old task was defined as one in which one half of the 9-year-olds tested could answer successfully.
640. IQ means a. a query of intelligence. b. indication of intelligence. c. intelligence quotient. d. intelligence questions for test construction.
C A quotient is the result when you perform division. The early ratio formula for the Binet IQ score was MA/CA (i.e., mental age divided by your chronological age) × 100. The score indicated how you compared to those in your age group. Memory device: An MA is a high degree so put it on top of the equation as the numerator.
691. Most counselors would agree that a. more preschool IQ testing is necessary. b. teachers need to give more personality tests. c. more public education is needed in the area of testing. d. the testing mystique has been beneficial to the general public.
C Again, the public needs to know the limitations of testing (i.e., that they are fallible). If you've been doing counseling for any length of time then you've surely come in contact with clients who have been harmed by hearing a score (e.g., their IQ) and then reacting to it such that it becomes a negative, self-fulfilling prophecy.
671. One major criticism of interest inventories is that a. they have far too many questions. b. they are most appropriate for very young children. c. they emphasize professional positions and minimize bluecollar jobs. d. they favor female pursuits.
C Also take note of the fact that contrary to popular opinion interests and abilities are not—that's right, not—highly correlated. A client, for example, could have tremendous musical ability in music yet could thoroughly dislike being a musician
680. A new IQ test has a standard error of measurement of 3. Tom scores 106 on the test. If he takes the test a lot, we can predict that about 68% of the time a. Tom will score between 100 and 103. b. Tom will score between 100 and 106. c. Tom will score between 103 and 109. d. Tom will score higher than Betty who scored 139.
C Calculated simply by taking: 106 - 3 = 103 and 106 + 3 = 109. Hint: Your exam could refer to this as the "68% confidence interval" (i.e., 103 to 109). Classical test theory suggests the formula, X = T + E, where X is the obtained score, T is the true score, and E is the error. Hence, psychometricians know that if a client takes the same test over and over, random error (i.e., E in the formula) will cause the score to fluctuate.
659. The Black versus White IQ controversy was sparked mainly by a 1969 article written by _______. a. John Ertl. b. Raymond B. Cattell. c. Arthur Jensen. d. Robert Williams.
C Here are four names with which to be familiar. Choice "a," John Ertl, claimed he invented an electronic machine to analyze neural efficiency and take the place of the paper and pencil IQ test. The device relies on a computer, an EEG, a strobe light, and an electrode helmet. The theory is that the faster one processes the perception, the more intelligence he or she has. I might add that thus far, counselors don't seem to be buying the idea! Choice "b," Raymond B. Cattell, is responsible for the fluid (inherited neurological that decreases with age and is not very dependent on culture) and crystallized intelligence (intelligence from experiential, cultural, and educational interaction). Crystallized intelligence is measured by tests that focus on content. Fluid intelligence is tested by what has been called "content-free reasoning" such as a block design or a pictorial analogy problem. Jensen, choice "c" mentioned earlier, sparked tremendous controversy—actually that's putting it mildly— when he suggested in a 1969 Harvard Educational Review article ("How Much Can We Boost IQ and Scholastic Performance?") that the closer people are genetically, the more alike their IQ scores. Adopted children, for example, will sport IQs closer to their biological parents than to their adopted ones. Jensen then leveled the charge that whites score 11 to 15 IQ points higher than African Americans (regardless of social class). His theory stated that due to slavery it was possible that African Americans were bred for strength rather than intelligence. He estimated that heredity contributed 80%, while environment influenced 20% of the IQ. Urie Bronfenbrenner, who could be included here if we had a choice "e," claimed that Jensen relied on twin studies with poor internal validity. Other researchers (e.g., Newman, Freeman, and Holzinger; Fehr) felt that genetic influences contributed less than 50% to IQ. In the final choice, Choice "d" , the African American psychologist Robert Williams created the Black Intelligence Test of Cultural Homogeneity (BITCH) to demonstrate that African Americans often excelled when given a test laden with questions whose answers would be familiar to members of the African American community. Williams charged that tests like the Binet and the Wechsler were part of "scientific racism." Williams—a victim of the system himself—scored an 82 on an IQ test at age 15 and his counselor suggested bricklaying since he was good with his hands! Williams rejected the advice and went on to put PhD after his name! IQ tests, though controversial to say the least, are, however, excellent predictors of school success in most cases since schools emphasize values that have been heavily influenced by European cultures.
670. An interest inventory would be least valid when used with a. a first-year college student majoring in philosophy. b. a third-year college student majoring in physics. c. an eighth-grade male with an IQ of 136. d. a 46-year-old White male construction worker.
C Interest inventories work best with individuals who are of high school age or above inasmuch as interests are not extremely stable prior to that time. Interests become quite stable around age 25.
644. A counselor is told by his supervisor to measure the internal consistency reliability (i.e., homogeneity) of a test but not to divide the test in halves. The counselor would need to utilize a. the split-half method. b. the test-retest method. c. the Kuder-Richardson coefficients of equivalence. d. cross-validation.
C Internal consistency or homogeneity of items also is known as "inter-item consistency." In plain everyday verbiage, the supervisor wants the counselor to find out if each item on the test is measuring the same thing as every other item. Is performance on one item truly related to performance on another? This can be done by using the Kuder-Richardson reliability/item consistency estimates, which are often denoted on exams as the KR-20 or KR-21 formulas. Lee J. Cronbach's alpha coefficient, also has been used in this respect. Cronbach's alpha and the KR-20 or KR-21 are alternatives to the split-half method. Choice "d," "cross- validation," is another popular term used in this area of study. Cross-validation takes place when a researcher further examines the criterion validity (and in rare instances, the construct validity) of a test by administering the test to a new sample. This procedure is necessary to ensure that the original validity coefficient is applicable to others who will take the exam. This method helps guard against error factors, which are likely to be present if the original sample size is small. In most cases a cross-validation coefficient is indeed smaller than the initial validity coefficient. This phenomenon is called "shrinkage."
686. Your client, who is in an outpatient hospital program, is keeping a journal of irrational thoughts. This would be a. an unethical practice based on NBCC ethical guidelines. b. considered a standardized test. c. an informal assessment technique. d. an aptitude measure.
C Self-reports, case notes, checklists, sociograms of groups, interviews, and professional staffings would also fall into the informal assessment category.
677. An aptitude test predicts future behavior while an achievement test measures what you have mastered or learned. In the case of a test like _______ the distinction is unclear. a. Binet b. Wechsler c. GRE d. Bender
C Sure, the GRE attempts to predict graduate school performance, but it also tests your level of knowledge. Some exams will refer to tests like the GRE, MAT, MCAT, SAT, etc., as "aptitude-achievement tests." Now here's where a counselor's life gets really complicated. Say your exam presents you with one of the aforementioned tests and gives you "aptitude" as one choice, and "achievement" as another, but does not give you "aptitude achievement" as an alternative (yipes!). Well, I certainly won't condone the practice, but based on my investigation of the textbook taxonomy of tests I'd opt for the "aptitude" option and latch onto the first good four-leaf clover I could get my hands on.
667. Both the Rorschach and the Thematic Apperception Test (TAT) are projective tests. The Rorschach uses 10 inkblot cards while the TAT uses a. a dozen inkblot cards. b. verbal and performance IQ scales. c. pictures. d. incomplete sentences.
C The TAT consists of 31 cards. The test, which is intended for ages 4 and beyond, uses up to 20 cards when administered to any given individual (i.e., 19 selected to fit the age and sex of the client, plus one blank card). The pictures on each card are intentionally ambiguous, and the client is asked to make up a story for each of them. Choice "d" would describe a projective test such as the Rotter Incomplete Sentence Blank (RISB) in which the subject completes an incomplete sentence with a real feeling.
700. In constructing a test you notice that all 75 people correctly answered item number 12. This gives you an item difficulty of a. 1.2. b. .75. c. 1.0. d. 0.0.
C The item difficulty index is calculated by taking the number of persons tested who answered the item correctly/total number of persons tested. Hence, in this case 75/75 = 1.0. This maximum score for item 12 tells you it is probably much too easy for your examinees.
687. You are uncertain whether a test is intended for the population served by your not-for-profit agency. The best method of researching this dilemma would be to a. contact a local APA clinical psychology graduate program. b. make a long distance call to the person who created the test. c. read the test manual included with the test. d. give the test to six or more clients at random.
C The manual should specify the target population for the test in question.
638. A career counselor is using a test for job selection purposes. An acceptable reliability coefficient would be _______ or higher. a. .20. b. .55. c. .80. d. .70.
C This is a tricky question. Although .70 is generally acceptable for most psychological attributes, for admissions for jobs, schools, and so on, it should be at least .80 and some experts will not settle for less than .90.
604. A short answer test is a(n) _______ test. a. objective. b. culture free. c. forced choice. d. free choice.
D
609. A test format could be normative or ipsative. In the normative format a. each item depends on the item before it. b. each item depends on the item after it. c. the client must possess an IQ within the normal range. d. each item is independent of all other items.
D
615. In a spiral test a. the items get progressively easier. b. the difficulty of the items remains constant. c. the client must answer each question in a specified period of time. d. the items get progressively more difficult.
D
618. In a counseling research study two groups of subjects took a test with the same name. However, when they talked with each other they discovered that the questions were different. The researcher assured both groups that they were given the same test. How is this possible? a. The researcher is not telling the truth. The groups could not possibly have taken the same test. b. The test was horizontal. c. The test was not a power test. d. The researcher gave parallel forms of the same test.
D
623. Which measure would yield the highest level of reliability? a. A TAT, projective test popular with psychodynamic helpers. b. The WAIS-III, a popular IQ test. c. The MMPI-2, a popular personality test. d. A very accurate scale.
D
628. When a counselor tells a client that the Graduate Record Examination (GRE) will predict her ability to handle graduate work, the counselor is referring to a. good concurrent validity. b. construct validity. c. face validity. d. predictive validity.
D
694. A word association test would be an example of a. a neuropsychological test. b. a motoric test. c. an achievement test. d. a projective test.
D Although it is rare, some texts and exams take issue with the archaic word projective and refer to such tests as "self- expressive."
688. Clients should know that a. validity is more important than reliability. b. projective tests favor psychodynamic theory. c. face validity is not that important. d. a test is merely a single source of data and not infallible.
D Although the first three choices are important to the counselor, the final statement should be explained to the client. An extremely high score—say on a mechanical aptitude test—does not automatically imply that the client will prosper as a mechanic.
665. The counselor who favors projective measures would most likely be a a. Rogerian. b. strict behaviorist. c. TA therapist. d. psychodynamic clinician.
D Choices "a," "b," and "c" all reflect positions that do not rely heavily on the unconscious mind (especially the behaviorists, who believe that if you can't directly measure the behavior, it is not meaningful).
639. The same test is given to the same group of people using the test-retest reliability method. The correlation between the first and second administration is .70. The true variance (i.e., the percentage of shared variance or the level of the same thing measured in both) is a. 70%. b. 100%. c. 50%. d. 49%.
D Here's the key to simplifying a question such as this. To demonstrate the variance of one factor accounted for by another you merely square the correlation (i.e., reliability coefficient). So .70 × .70 = .49 and .49 × 100 = 49%. Your exam could refer to this principle as the coefficient of determination.
697. You want to admit only 25% of all counselors to an advanced training program in psychodynamic group therapy. The item difficulty on the entrance exam for applicants would be best set at a. 0.0. b. .5 regardless of the admission requirement. c. 1.0. d. .25.
D In most tests the level is set at .5 (i.e., 50% of the examinees will answer correctly while 50% will not). However, in this case the .25 level would allow you to ferret out the lower 75% you do not wish to admit. Item difficulty ranges from 0.0 (choice "a") to 1.0 (choice "c"). The higher the index number, the greater the number of examinees who will answer the question correctly. Or simply: The higher the number, the easier the question is to answer.
681 A counselor created an achievement test with a reliability coefficient of .82. The test is shortened since many clients felt it was too long. The counselor shortened the test but logically assumed that the reliability coefficient would now a. be approximately .88. b. remain at .82. c. be at least 10 points higher or lower. d. be lower than .82.
D Increasing a test's length raises reliability. Shorten it and the antithesis occurs. Note: The Spearman Brown formula is used to estimate the impact that lengthening or shortening a test will have on a test's reliability coefficient.
698. According to Public Law 93-380, also known as the Buckley Amendment, a 19-year-old college student attending college a. could view her record, which included test data. b. could view her daughter's infant IQ test given at preschool. c. could demand a correction she discovered while reading a file. d. all of the above.
D Persons over age 18 can inspect their own records and those of their children. The Family Educational Rights and Privacy Act (FERPA) also stipulates that information cannot be released without adult consent.
699. Lewis Terman a. constructed the Wechsler tests. b. constructed the initial Binet prior to 1910. c. constructed the Rorschach. d. Americanized the Binet.
D Since Terman was associated with Stanford University the test became the Stanford-Binet.
662. In a projective test the client is shown a. something which is highly reinforcing. b. something which is highly charged from an emotional standpoint. c. a and b. d. neutral stimuli.
D there are several acceptable formats for projective tests: First, Association—such as "What comes to mind when you look at this inkblot?" Second, Completion—"Complete these sentences with real feelings." Third, Construction—such as drawing a person. The theory is that self-report inventories like the MMPI do not reveal hidden unconscious impulses. In order to accomplish this the client is shown vague, ambiguous stimuli such as a picture or an inkblot. Some counselors believe that by using projective measures a client will have more difficulty faking his or her responses and that he or she will be able to expand on answers. It should be noted that examiner bias is common when using projectives and a therapist using projective measures needs more training than one who merely works with self-report tests.
