Test Score Interpretation and Appendix A

Question 1

The ____ ____ ____ obtained by an examinee has only limited meaning. It is only when the score is tired of one of two “____” – either to the ____ of ____ ____ or to an ____ ____ of ____ - that it becomes meaningful. The former method is referred to as ____-____ ____; the latter is called ____-____ ____.

Answer 1

Raw Test Scores; Anchors; Performance of Other Examinees; Established Standard of Performance; Norm-Referenced Interpretation; Criterion-Referenced Interpretation

Question 2

____-____ ____ involves comparing an examinee’s test score to scores obtained by people included in a normative (standardization) sample and is useful for identifying ____ ____. To interpret scores in terms of norms, an examinee’s raw test score is ____ to ____ ____ that indicates their ____ ____ in the ____ ____.

Answer 2

Norm-Referenced Interpretation; Individual Differences; Converted to Another Score; Relative Standing; Norm Group

Question 3

The adequacy of norm-referenced interpretation relies on the extent to which the ____ ____ ____ those of the ____ in the ____ ____. If an examinee’s characteristics do not match, the interpretation of his or her score may be ____. As an example, an inexperienced sales applicant’s assertiveness test score might be misinterpreted if it is compared to the distribution of scores obtained by a sample consisting of only experienced salespeople. A major difficulty with norm-referenced interpretation is finding ____ that were ____ from people whose ____ are ____ to those of the ____; and this problem is compounded by the fact that, for many tests, ____ ____ can become ____ ____ ____.

Answer 3

Examinee’s Characteristics Match; People in the Norm Sample; Misleading; Norms; Derived; Characteristics are Similar; Examinee; Normative Data; Obsolete Rather Quickly

Question 4

A ____ ____ (PR) expresses an examinee’s raw score in terms of the percentage of examinees in the norm sample who achieved lower scores. The primary advantage of percentile ranks is that they are ____ to ____: If a sales applicant’s raw score on the assertiveness test is equivalent to a percentile rank of 88, this means that 88% of the people in the norm sample obtained scores ____ than the applicant’s score.

Answer 4

Percentile Rank; Easy to Interpret; Lower

Question 5

A distinguishing characteristic of percentile ranks is that their distribution is always ____ (rectangular) in ____ regardless of the shape of the raw score distribution. This is because percentile ranks are ____ ____ throughout the ____ of ____. In other words, an ____ ____ of ____ falls between the 10th and 20th percentile, between the 20th and 30th percentile, and so on. Whenever a distribution of transformed scores differs in shape from the distribution of raw scores, as it does with percentile ranks, the score transformation is referred to AS A ____ ____.

Answer 5

Flat; Shape; Evenly Distributed; Range of Scores; Equal Number of Scores; Nonlinear Transformation

Question 6

The main disadvantage of percentile ranks is that they represent an ____ ____ of ____. As a result, while percentile ranks indicate an examinee’s ____ ____ in a ____, they do not provide information about ____ ____ between ____. If Applicants A, B, And C obtain percentile ranks of 30, 40, and 50, respectively, on the Interpersonal assertiveness test, it would be possible to say that Applicant B is more assertive than Applicant A and less assertive than Applicant C.

Answer 6

Ordinal Scale of Measurement; Relative Position in a Distribution; Absolute Differences; Examinees

Question 7

It would not be possible, to conclude that the difference in assertiveness between Applicants B and A is the same as the difference between Applicants B and C. Another problem is that the transformation of scores to percentile ranks has the effect of ____ ____ between examinees whose raw in the ____ of the ____ while ____ (minimizing) ____ between examinees whose raw scores are at the ____.

Answer 7

Maximizing Differences; Middle of the Distribution; Compressing Differences; Extremes

Question 8

____ ____: when an examinee’s raw test score is converted to a ____ ____, the transformed score indicates the examinee’s position in the normative sample in terms of ____ ____ from the ____. A primary advantage of standard scores is that they permit ____ of ____ obtained from ____ ____.

Answer 8

Standard Scores; Standard Score; Standard Deviations from the Mean; Comparisons of Scores; Different Tests

Question 9

The most commonly used standard score is the _-____. The _-____ equivalent for an examinees raw score is calculated by subtracting the mean of the distribution from the raw score to obtain a deviation score and then dividing the deviation score by the distribution’s standard deviation. Z = (X – M) / SD

Answer 9

Z-Score; Z-Score

Question 10

For example, assume that the assertiveness test has a mean of 50 and a standard deviation of 10. and that a job applicant receives a score of 60. The applicant’s z-score is + 1.0: (60 50j/IO = + 1.0. This score indicates that the applicant received a score that is ____ ____ ____ ____ the ____ achieved by people in the ____ ____.

Answer 10

One Standard Deviation Above the Mean; Normative Sample

Question 11

The distribution of z-scores has the following properties: (l) The mean of the z-score distribution is equal to _. (2) The standard deviation is equal to _. (3) All raw scores below the mean of the distribution are ____ _-____, scores above the mean are ____ _-____, and scores equal to the mean are equal to a z-score of _. (4) Unless it is “normalized,” the z-score distribution has the same shape as the ____ ____ ____. In other words, the transformation of raw scores to z-scores is a ____ ____.

Answer 11

0; 1; Negative Z-Scores; Positive Z-Scores; 0; Raw Score Distribution; Linear Transformation

Question 12

Other standard scores include _-____, deviation IQ scores, and stanines. These scores are similar to z-scores but have different ____ and ____ ____.

Answer 12

T-Scores; Means and Standard Deviation

Question 13

The T-score distribution has a mean of ¬_ and standard deviation of ¬_. Consequently, a person whose raw score is one standard deviation below the mean would obtain a T-score of _.

Answer 13

50; 10; 40

Question 14

Many intelligence tests provide deviation __ ____. For example, the Stanford-Binet Intelligence Scales, 5th Edition (SB5) and the Wechsler Adult Intelligence Scales, 4th Edition (WAIS-IV) provide deviation IQs that have a mean of _ and standard deviation of _. A person who obtains a Full Scale score of 130 on the WAIS-IV has obtained a score that Is ____ ____ ____ ____ the ____.

Answer 14

IQ Scores; 100; 15; Two Standard Deviations Above the Mean

Question 15

____ ____ divide a distribution of scores into nine parts and have a mean of _ and standard deviation of approximately _. Stanines are expressed as ____ ____ that range from _ to _, with each unit representing ____-____ of a ____ ____ — e.g., a stanine of 5 ranges from -.25 standard deviations to + .24 standard deviations, and a stanine of 6 ranges from + .25 standard deviations to + .74 standard deviations.

Answer 15

Stanine Scores; 5; 2; Whole Numbers; 1 to 9; One-Half of a Standard Deviation

Question 16

You want to be familiar with the equivalence of different norm-referenced scores in a normal distribution. Note that it is possible to calculate a percentile rank equivalent of a z-score using the areas under the ____ ____. For instance, in a normal distribution, a percentile rank of 84 is equivalent to a z-score of + 1.0: __ of cases fall below the mean of the distribution and __ (half of 68%) fall between the mean and one standard deviation above the mean: 50 + 34 = 84.

Answer 16

Normal Curve; 50%; 34%

Question 17

____-____ ____ involves interpreting scores in terms of a prespecified standard. A ____ ____ (percent correct) is one type of criterion-referenced score: It indicates the ____ of the ____ ____ that an ____ ____ ____. If the maximum total score on a test is 200, for instance, an examinee who obtains a raw score of 150 would have a percentage score of 75. when percentage scores are used as the method of score interpretation, a ____ ____ is ____ ____ (e.g., 70 percent) so that examinees who score at or ____ the ____ have “____” the test, while examinees who score below the cutoff have “____.”

Answer 17

Criterion-Referenced Interpretation; Percentage Score; Percentage; Test Content; Examinee Answered Correctly; Cutoff Score is Usually Set; Above the Cutoff; Passed; Failed

Question 18

This method of score interpretation is often employed in ____ (____-____) ____, which involves specifying the ____ ____ of ____ required for ____ ____ (e.g., 90% correct) and periodically administering a test to ____ to assess their degree of ____. If a learner’s test performance indicates deficiencies, he or she is given appropriate ____ ____. (Note that percentage scores are also referred to as content- and domain-referenced scores.)

Answer 18

Mastery (Criterion-Referenced) Testing; Terminal Level of Performance; All Learners; Learners; Mastery; Deficiencies; Remedial Instruction

Question 19

Another type of criterion-referenced interpretation involves ____ ____ ____ ____ in terms of their ____ ____ on an ____ ____. This can be accomplished through the use of a ____ ____ or an ____ ____.

Answer 19

Interpreting Examinees’ Test Scores; Likely Status; External Criterion; Regression Equation; Expectancy Table

Question 20

Example: Using the data collected in her concurrent validity study, the psychologist constructs an ____ ____ that allows her to predict a job applicant’s average monthly sales based on the applicant’s assertiveness test score.

Answer 20

Expectancy Table

Question 21

Have percentile ranks and standard scores Linked with ____-____ ____, and percentage scores, regression equation, and expectancy table associated with ____-____ ____.

Answer 21

Norm-Referenced Interpretation; Criterion-Referenced Interpretation

Question 22

Scores on objective tests are sometimes ____ for ____ to ensure that an examinee does not benefit from ____ ____. When test scores will be corrected for guessing, an examinee’s best strategy is to ____ ____ to ____ that are ____ rather than to ____ ____. Several correction for guessing ____ are available. Most involve calculating a ____ ____ by taking into account the ____ of ____ ____, the number of ____ ____, and the number of ____ for ____ ____. When the correction involves ____ ____ from ____ ____, the resulting score distribution will have a ____ ____ and a ____ ____ ____ than the ____ ____.

Answer 22

Corrected for Guessing; Random Guessing; Omit Answers to Questions; Unknown; Guess Wildly; Formulas; Corrected Score; Number of Correct Responses; Incorrect Responses; Alternatives for Each Item; Subtracting Points; Examinee’s Scores; Lower Mean; Larger Standard Deviation; Original Distribution

Question 23

An examinee’s raw test score is usually easier to interpret when it can be compared to the scores of other examinees or to a prespecified standard of performance. The former technique is referred to as (1) ____-referenced interpretation and may entail converting an examinee’s raw score to a percentile rank or standard score. A percentile rank indicates the percent of examinees in the normative sample who obtained (2) ____ scores.

Answer 23

(1) norm; (2) lower

Question 24

The transformation of raw scores to percentile ranks is (3) ____ because the shape of the distribution of percentile ranks is always (4) ____ regardless of the shape of the raw score distribution. Standard scores anchor an examinee’s test score to those of the norm group by reporting the examinee’s score in terms of (5) ____ from the mean. One of the most commonly used standard scores is the z-score. The z-score distribution has a mean of (6) ____ and a standard deviation of 1.0. A z-score of (7) ____ indicates that an examinee’s raw score is one standard deviation below the mean.

Answer 24

(3) nonlinear; (4) flat (rectangular); (5) standard deviations; (6) 0; (7) —1.0

Question 25

Scores can also be interpreted in terms of a prespecified standard of performance. (8) ____ scores are one means of criterion-referenced interpretation. They indicate the proportion of the test content (e.g., percent of test items) that examinees answered correctly. An (9) ____ table makes it possible to use an examinee’s predictor (test) score to estimate the probability that he or she will attain different scores on a (10) ____. Scores on objective tests are sometimes (11) ____ to ensure that examinees don’t benefit from random guessing.

Answer 25

(8) Percentage; (9) expectancy; (10) criterion; (11) corrected for guessing

Question 26

Factor analysis is used either to identify the ____ ____ of ____ “____” (dimensions) needed to explain the ____ among a ____ of ____, ____, or ____ ____ (exploratory factor analysis) or to ____ ____ about the ____ and ____ of ____ ____ (confirmatory factor analysis).

Answer 26

Minimum Number of Underlying “factors;” Intercorrelations; Set of Tests, Subtests, or Test Items; Test Hypotheses; Number and Nature; Underlying Factors

Question 27

Factor analysis serves a number of functions including a test’s ____ ____, creating ____, and ____ the ____ of intelligence, personality, attitudes, interests, and other attributes. Factor analysis is a very ____ ____ both ____ and ____ but can be described in four steps.

Answer 27

Construct Validity; Subtests; Identifying the Components; Complex Procedure; Conceptually and Mathematically

Question 28

Step _: ____ ____ ____, ____, or ____ ____ to a group of examinees and ____ ____ on ____ ____ of tests, subtests, or test items to obtain a ____ ____. Use the information in the correlation matrix to determine how many factors to “____” in the factor analysis.

Answer 28

1; Administer Several Tests, Subtests, or Test Items; Correlate Scores on All Pairs; Correlation Matrix; Extract

Question 29

An inspection of the coefficients in the correlation matrix reveals that there are two “clusters” of tests: Tests A. B, and C have relatively high correlations with each other (.56, .58, and .52) and Tests X, Y, and Z have relatively high correlations with each other (.52, .55, and .60), but Tests A, B. and C have low correlations with Tests X, Y, and Z. The high intercorrelations between tests within each of the two clusters suggest that the tests in each cluster are measuring a ____ ____ (factor). Since there are two clusters of tests, ____ ____ will be ____ in the ____ ____. (If there had been three clusters, three factors would have been extracted, and so on.)

Answer 29

Common Trait; Two Factors; Extracted; Factor Analysis

Question 30

Step _: Use one of several ____ ____ ____ (e.g., maximum likelihood, alpha, principal components) to ____ the ____ ____ in the ____ ____ to a ____ ____.

Answer 30

2; Factor Analytic Techniques; Convert the Data Contained; Correlation Matrix; Factor Matrix

Question 31

The initial factor matrix is difficult to interpret because it typically has a pattern of ____ ____ ____ (correlations) for ____ ____ on the ____ ____, ____ ____ and ____ ____ on the ____ ____, and ____ ____ on any ____ ____. In most situations, the ideal pattern is for ____ ____ to ____ ____ with ____ fac____ tor and other tests to ____ ____ with ____ ____. Consequently, the initial factor matrix is ordinarily not ____ but is “____” to obtain a ____ ____ ____.

Answer 31

High Factor Loadings; Each Test; First Factor; Moderate Positive and Negative Loadings; Second Factor; Declining Loadings; Succeeding Factors; Some Tests to Correlate Highly with One Factor; Correlate Highly with Another Factor; Interpreted; Rotated; Rotated Factor Matrix

Question 32

Step _: ____ the ____ and ____ of the ____ ____ by ____ the ____ using one of several available techniques (e.g., varimax, quartimax). Rotation actually involves ____ ____ ____ that represent the ____ ____. As shown below, when there are ____ ____, there are two axes, one representing ____ _ and the other representing ____ _. The tests included in the factor analysis are ____ on the ____ ____ of ____ on the basis of their ____ ____. When the axes are rotated, the process is ____: The new factor loadings are derived from the ____ of the ____ ____ to the ____ ____.

Answer 32

3; Simplify the Interpretation and Naming; Identified Factors; Rotating the Factors; Physically Rotating Axes; Extracted Factors; Two Axes; Factor I; Factor II; Plotted; Initial Set of Axes; Factor Loadings; Reversed; Position of the Tests Relative to the Rotated Axes

Question 33

There are two types of ____: In an ____ ____, the axes remain at a 90 degree angle; in an ____ ____, the angle between the axes is something other than 90 degrees. An ____ ____ produces ____ ____, which are ____, while an oblique rotation produces ____ ____, which are ____. The investigator’s theory about the nature of the constructs measured by the tests included in the analysis determines whether the rotation should be ____ or ____.

Answer 33

Rotation; Orthogonal Rotation; Oblique Rotation; Orthogonal Factors; Uncorrelated; Oblique Factors; Correlated; Orthogonal or Oblique

Question 34

Example: The psychologist believes that interpersonal assertiveness and aggressiveness are unrelated characteristics, so she conducts an ____ ____ and obtains the following rotated factor matrix.

Answer 34

Orthogonal Rotation

Question 35

The orthogonal rotation is depicted below. Because it was orthogonal, the axes were ____ at a _ ____ ____. Note that a rotation is always done so that each axis passes as closely as possible to a ____ of ____, which represent the ____ ____ in the ____.

Answer 35

Maintained; 90 Degree Angle; Cluster of Points; Tests Included in the Analysis

Question 36

Step _: ____ and ____ the ____ ____. The rotated factors are named by considering the ____ of the ____ that ____ and ____ ____ ____ with ____ ____. (See Section ill.B.2 of this chapter for a description of the interpretation of the rotated factors.)

Answer 36

4; Interpret and Name the Rotated Factors; Nature; Tests; Do and Do Not Correlate; Each Factor

Question 37

The primary purpose of ____ ____ ____ is to identify a set of variables that explains all (or nearly all) of the total variance in a set of test scores. The variables are called ____ ____ (or ____), and a ____ ____ is defined as the linear combination of the set of tests that describes as much of the intercorrelations between the tests as possible. Principal components analysis is used as a technique in itself to identify the ____ or ____ that ____ a ____ of ____ and as an initial step in ____ ____-____ ____ ____.

Answer 37

Principal Components Analysis; Principal Components; Eigenvectors; Principal Component; Variables or Dimensions; Underline a Set of Tests; Principal Components-Based Factor Analysis

Question 37

The primary purpose of ____ ____ ____ is to identify a set of variables that explains all (or nearly all) of the total variance in a set of test scores. The variables are called ____ ____ (or ____), and a ____ ____ is defined as the linear combination of the set of tests that describes as much of the intercorrelations between the tests as possible. Principal components analysis is used as a technique in itself to identify the ____ or ____ that ____ a ____ of ____ and as an initial step in ____ ____-____ ____ ____.

Answer 37

Principal Components Analysis; Principal Components; Eigenvectors; Principal Component; Variables or Dimensions; Underline a Set of Tests; Principal Components-Based Factor Analysis

Question 38

The output of a principal components analysis looks very similar to a ____ ____ except that the number of components is always the ____ as the ____ of ____ included in the ____.

Answer 38

Factor Matrix; Same as the Number of Tests; Analysis

Question 39

Because principal components are ____, each component accounts for a ____ ____ of the ____ in ____ ____. The eigenvalues listed at the bottom of the table indicate the ____ of ____ accounted for by ____ ____. A component’s eigenvalue is calculated by ____ the ____ ____ between that ____ and ____ ____. For example, the eigenvalue for Component I is equal to .80^2 + .77^2 + .75^2 + .76^2.

Answer 39

Orthogonal; Unique Amount; Variability in Test Scores; Amount of Variability; Each Component; Summing the Squared Correlations; Component and Each Test

Question 40

As can be seen from the eigenvalues, the principal components are ____ so that the ____ ____ accounts for the ____ ____ of ____ in ____ ____, the second component accounts for the second largest amount of variability, etc. When the eigenvalues are added, the sum equals (or is about equal to) the ____ ____ of ____ included in the ____; i.e., 2.37 + .79 + .72 + .08 = 3.96. To facilitate interpretation of the eigenvalues, they can be ____ by the ____ of ____ to determine the ____ of ____ explained by each ____. For the first component in the above example, 2.37/4 = .59. This means that Component I explains __ of the ____ ____ in the ____ ____.

Answer 40

Extracted; First Component; Largest Amount of Variability in Test Scores; Total Number of Tests; Analysis; Divided; Number of Tests; Precent of Variability; Component; 59%; Total Variability; Four Tests

Question 41

As in factor analysis, the goal in principal components analysis is to ____ the ____ ____ that explain the ____ in ____ ____. There are several different techniques for determining which components to ____. One method involves retaining only those components that have an eigenvalue of _ or ____. The rest of the analysis is similar to ____ ____. The components can be ____ and ____ as they ____ or, to facilitate interpretation, the components can be ____ and then ____ and ____.

Answer 41

Identify the Fewest Components; Variability in Test Scores; Retain; 1.0 or Higher; Factor Analysis; Named and Interpreted; Are; Rotated; Named and Interpreted