Deck 3: Norms and Reliability

Question 1

The middlemost score when all the scores have been ranked is called the

A) mean
B) median
C) mode
D) second quartile

Answer 1

B

Question 2

The most frequently occurring score is the

A) mean
B) median
C) mode
D) all of the above

Answer 2

C

Question 3

In a frequency distribution, the sums of the frequencies for all intervals will ______the total number of scores in the sample.

A) exceed
B) equal
C) be less than
D) be less than or equal to

Answer 3

equal

Question 4

For which kind of distribution would the highest number of persons score in the superior range?

A) rectangular
B) normal
C) platonic
D) kurtic

Question 5

A small x is used to designate

A) the mean
B) correlation
C) individual scores
D) deviation scores

Question 6

If the number of scores is even, the median

A) is the average of the middlemost two scores
B) is the higher of the middlemost two scores
C) is the lower of the middlemost two scores
D) cannot be computed

Question 7

The initial outcome of testing is typically a

A) raw score
B) scaled score
C) subscale score
D) transformed score

Question 8

What statement best characterizes a "raw score"?

A) it is useless by itself
B) it is the end goal of test interpretation
C) it is a converted score based on a norm group
D) none of the above

Question 9

In a______ , the frequency of the class intervals is represented by single points rather than columns.

A) histogram
B) frequency table
C) frequency polygon
D) scattergram

Question 10

Consider the case where nine persons earn $10,000 and a tenth person earns $990,000. What is their average income?

A) $99,000
B) $111,000
C) $10,000
D) $100,000

Question 11

A sample of examinees who are representative of the population for whom the test is intended is called a ______group.

A) reference
B) criterion
C) norm
D) stratified

Question 12

In a smooth but skewed distribution, the most representative measure of central tendency is the

A) mean
B) median
C) mode
D) quartile

Question 13

What is the relationship between the variance and the standard deviation?

A) they are completely unrelated
B) they are usually correlated to a modest degree
C) one can be computed from the other
D) one is the inverse of the other

Question 14

The most commonly used statistical index of variability in a group of scores is the

A) range
B) variance
C) standard deviation
D) none of the above

Question 15

The essential objective of _______is to determine the distribution of raw scores in the norm group so that the test developer can publish derived scores known as norms.

A) statistical regression
B) test standardization
C) data correlation
D) obtaining a standard score

Question 16

For criterion-referenced tests, norms are

A) essential
B) common but not essential
C) uncommon and not essential
D) common and essential

Question 17

The Greek symbol "sigma" stands for

A) mean
B) sum of
C) correlation
D) difference

Question 18

Percentile ranks, age equivalents, grade equivalents, and standard scores are all examples of

A) norms
B) non-linear transformations
C) normal distributions
D) all of the above

Question 19

It is common for frequency distributions to include between ______class intervals.

A) three and five
B) three and eight
C) five and ten
D) five and fifteen

Question 20

Renorming of tests should

A) be the exception, not the rule
B) be the rule, not the exception
C) never be conducted
D) none of the above

Question 21

Suppose a group of scores has a mean of 40 and a standard deviation of 10. A score of 80 would equate to a standard score of

A) 1.0
B) 2.0
C) 4.0
D) 8.0

Question 22

The difference in underlying raw score points between percentiles of 90 and 99 is ________that between percentiles of 50 and 59.

A) greater than
B) less than
C) the same as
D) of unknown relation to

Question 23

Suppose that a college freshman earned 125 raw score points on a vocabulary test where the normative sample averaged 100 points with SD of 15 points). Suppose, in addition, he earned 110 raw score points on a spatial thinking test where the normative sample averaged 90 points with SD of 20 points). In which skill area does he show greater aptitude, vocabulary or spatial thinking?

A) vocabulary
B) spatial thinking
C) aptitude is equal in the two areas
D) insufficient data to answer the question

Question 24

A _______uses the standard deviation of the total distribution of raw scores as the fundamental unit of measurement.

A) percentile
B) percentile rank
C) standard score
D) T-score

Question 25

If test scores are piled up at the low end of the scale, the distribution is said to be

A) negatively skewed
B) positively skewed
C) leptokurtic
D) platykurtic

Question 26

In his initial investigations, Wechsler considered the belief that mental measures must distribute themselves according to the normal curve to be

A) axiomatic
B) probably correct
C) mistaken
D) merely an hypothesis

Question 27

Percentile scores have one major drawback:

A) they are difficult for most persons to comprehend
B) they often sum to more than 100 percent
C) an examinee's percentile score fluctuates significantly
D) they distort the underlying measurement scale

Question 28

For which types) of scores are negative values possible?

A) standard score
B) standardized scores
C) neither a nor b
D) both a and b

Question 29

A percentile of 25 is often denoted as

A) Q25
B) Q75
C) Q1
D) none of the above

Question 30

Why do psychologists prefer the normal distribution?

A) it has useful features for statistical inference
B) it can be defined with mathematical precision
C) it often arises spontaneously in nature
D) all of the above

Question 31

Suppose a subject scored at the 94th percentile on a psychological test? What does that mean?

A) the subject answered 94 percent of the questions correctly
B) the subject's score exceeded 94 percent of the standardization sample
C) the subject's score exceeded 6 percent of the standardization sample
D) none of the above

Question 32

A group of standard scores always possesses a mean of _____and a standard deviation of ______.

A) 1.0, 1.0
B) 1.0, 10.0
C) 0.0, 1.0
D) 0.0, 10.0

Question 33

The main advantage of_______ is retaining the relative magnitude of distances between successive values found in the original raw scores.

A) percentiles
B) standard scores
C) percentile ranks
D) composite scores

Question 34

When initial research indicates that an instrument produces skewed results in the standardization sample, test developers typically

A) use a statistical transformation
B) revamp the test at the item level
C) obtain data from a second sample
D) all of the above

Question 35

_______are the most common type of raw score transformation encountered in psychological testing.

A) Standard scores
B) Normalized scores
C) T-scores
D) Percentiles

Question 36

In a normal distribution, approximately ______percent of the scores will fall within one standard deviation of the mean in either direction.

A) 28
B) 48
C) 68
D) 88

Question 37

In a normal distribution, approximately______ percent of the scores will exceed the mean by two standard deviations.

A) 2
B) 5
C) 10
D) 25

Question 38

A z-score is the same thing as a

A) percentile
B) percentile rank
C) standard score
D) T-score

Question 39

In psychological testing, skewed distributions usually signify that the test developer has

A) a bias that favors one or more subgroups of persons
B) purposely set out to produce a test with skewness
C) underestimated the ability of some subjects
D) included too few easy items or too few hard items

Question 40

The distribution of mental test scores in standardization samples typically

A) approximates a normal curve
B) is negatively skewed
C) is positively skewed
D) is bimodal

Question 41

A C scale consists of______ units.

A) 9
B) 10
C) 11
D) 100

Question 42

Which of the following is NOT usually true of a norm group?

A) it is homogeneous
B) it is representative of the population
C) it is large, typically hundreds of subjects
D) it is tested according to standard procedures

Question 43

The mean of a standardized score is typically set at

A) 100
B) 500
C) 50
D) all of the above

Question 44

Which of the following necessarily has a mean of 50 and a standard deviation of 10?

A) standard score
B) normalized standard score
C) z-score
D) T-score

Question 45

When test scores are expressed as a percentage, with the passing level predetermined, the examiner is probably using

A) a norm-referenced test
B) a criterion-referenced test
C) a norm-referenced or criterion-referenced test
D) neither a norm-referenced nor criterion-referenced test

Question 46

What approach do most test developers use in choosing a norm group?

A) they strive to make a good faith effort to select a representative sample
B) they strive to obtain a pure random sample
C) they strive to obtain a sample on an opportunistic basis
D) they strive to use a strict strategy of diversified random sampling

Question 47

Which kind of norms are appropriate for aptitude tests?

A) age norms
B) grade norms
C) both age and grade norms
D) neither age nor grade norms

Question 48

The use of normalized standard scores is appropriate when

A) the normative sample is large and representative
B) the raw score distribution is only mildly non-normal
C) both a and b
D) neither a nor b

Question 49

T score scales are especially common for______ tests.

A) intelligence
B) personality
C) achievement
D) aptitude

Question 50

The restriction imposed by Hollerith cards was a main impetus for the development of _____scales.

A) stanine
B) sten
C) C
D) digit

Question 51

A tabular depiction of the established relationship between test scores and expected outcome on a relevant task is known as an)

A) correlation coefficient
B) histogram
C) frequency polygon
D) expectancy table

Question 52

In a ______scale, all raw scores are converted to a single-digit system of scores ranging from 1 to 9.

A) stanine
B) sten
C) C
D) digit

Question 53

The mean of ______scores is always 5, and the standard deviation is approximately 2.

A) stanine
B) sten
C) C
D) digit

Question 54

Which of the following is NOT true about the concept of variance?

A) variance is computed from the mean of the squared deviations
B) psychometricians prefer variance to standard deviation as an index of variability
C) variance is represented by sigma squared
D) variance is standard deviation squared

Question 55

To eliminate negative scores, the pre-selected mean of a standardized score should be at least ______times as large as the standard deviation.

A) two
B) three
C) five
D) ten

Question 56

A WAIS-III IQ of 115 could also be reported as a

A) standard score of +1.00
B) T score of 60
C) percentile rank of 84
D) all of the above

Question 57

The most popular approach for normalizing standard scores is conversion of

A) percentiles to normalized standard scores
B) raw scores to T scores
C) T scores to raw scores
D) percentiles to percentile ranks

Question 58

For which of the below is it LEAST likely that we would find proportionate representation of race, social class, age groups, etc.?

A) testing a large random sample
B) testing a small random sample
C) testing a stratified random sample
D) testing an entire population

Question 59

If 12 percent of the relevant population is African American, and a test developer chooses subjects randomly but with the constraint that 12 percent of the norm group is also African American, what approach is the developer using?

A) random sampling
B) classified sampling
C) stratified random sampling
D) selective random sampling

Question 60

For what purpose might ethnic norms for standardized intelligence tests be superior to nationally based norms?

A) for predicting competence within the child's non-school environment
B) for predicting performance within mainstream schools
C) for predicting college performance in mostly white schools
D) all of the above

Question 61

Typically, errors of measurement arise from

A) item selection
B) test administration
C) test scoring
D) all of the above

Question 62

The value of a reliability coefficient can vary between

A) -1.0 and +1.0
B) 0.0 and +1.0
C) 0.0 and 0.9
D) two numbers that depend upon the underlying units of measurement

Question 63

The sources of measurement error are

A) mysterious and unknown
B) potentially knowable for individual cases
C) of no significance in test development
D) all of the above

Question 64

Errors of measurement are

A) positive
B) negative
C) positive or negative
D) none of the above

Question 65

Measurement errors

A) are equally likely to be positive or negative
B) will average out to zero across a large group of subjects
C) act as random influences
D) all of the above

Question 66

Reliability and measurement error are

A) unrelated
B) moderately correlated
C) related in a precise statistical manner
D) none of the above

Question 67

Classical theory assumes that measurement errors

A) are not correlated with true scores
B) usually penalize the subject
C) are correlated from one test to the next
D) can be determined for the individual subject

Question 68

An important new use of criterion-referenced tests is:

A) establishing normative scores for IQ tests
B) identifying test-retest reliability on achievement tests
C) as a performance requirement for high-school graduation
D) determining entry into high-school programs

Question 69

The particular choice of items on a test is ______a source of measurement error in psychological testing.

A) always
B) usually
C) rarely
D) never

Question 70

Which of the following is an example of measurement error in test administration?

A) poor lighting
B) a fatigued examinee
C) an intimidating examiner
D) all the above

Question 71

Which of the following is most likely to be a source of measurement error?
a. projective tests
b. machine scored multiple-choice items
c. hand scored multiple-choice items
d. true-False items

Question 72

Unsystematic measurement errors behave like _______variables.

A) predictable
B) random
C) discrete
D) functional

Question 73

The variance of obtained scores is

A) the variance of true scores minus the variance of errors of measurement
B) the variance of true scores plus the variance of errors of measurement
C) The variance of errors of measurement minus the variance of true scores
D) none of the above

Question 74

As noted by FairTest, what determines whether or not criterion-referenced tests are fair?

A) results reflect an equal distribution of scores
B) how the cut-off scores are determined
C) results reflect a negative distribution of scores
D) results reflect a positive distribution of scores

Question 75

Regarding the true score, which statement is correct?

A) we can never know the true score with certainty
B) we can derive a probability that the true score resides within a certain interval
C) we can derive a best estimate of the true score
D) all of the above

Question 76

Correlations of +1.0 are extremely rare in psychological research and usually signify
a. an important breakthrough
b. a computational error
c. a trivial finding
d. a False relationship

Question 77

If scores on ability tests A and B are negatively correlated, this usually means that

A) scores on one of the tests are reported as errors
B) abilities measured by the two tests are unrelated
C) a high score on one test is associated with a low score on other
D) all of the above

Question 78

Which kind of test is interpreted in relation to specified standard of mastery?

A) criterion-referenced
B) norm-referenced
C) both of the above
D) neither of the above

Question 79

_________measurement errors arise when a test consistently measures something other than the trait for which it was intended.

A) Random
B) Systematic
C) Unsystematic
D) Probabilistic

Question 80

The classical theory of reliability is also known as the theory of

A) consistency
B) general and specific factors
C) true and error scores
D) generalizability