Deck 5: Measurement

Question 1

What is the generic term for any type of measurement device?

A) gauge
B) measure
C) tool
D) dial

Answer 1

B

Question 2

A measure is said to be valid to the extent that it does what?

A) accurately measures what it purports to measure
B) measures within the margin of error
C) produces the same results every time
D) produces results that confirm the hypothesis

Answer 2

A

Question 3

According to this topic, no test is perfectly valid.

Answer 3

True

Question 4

How is content validity established?

A) by comparing results to the hypothesis
B) by testing a measure and analyzing the answers
C) with statistical tests such as p-value
D) by asking participants for feedback

Question 5

How is face validity determined?

A) by observing the reactions of participants
B) by testing a measure and analyzing the answers
C) with statistical tests such as p-value
D) by considering if what is asked is relevant to the study

Question 6

When might researchers prefer a survey with low face validity?

A) when using statistical tools of analysis
B) to boost reliability
C) to mask what is being investigated
D) to increase the content validity

Question 7

What two other terms are used to refer to "empirical validity"?

A) face validity or content validity
B) predictive validity or criterion-related validity
C) content validity or criterion-related validity
D) predictive validity or face validity

Question 8

If a test is perfectly valid, what value will its validity coefficient have?

A) 0.0
B) <0.1
C) 1.0
D) >1.0

Question 9

If a test has no validity at all, what value will its validity coefficient have?

A) 0.0
B) <0.1
C) 1.0
D) >1.0

Question 10

To what does predictive validity refer?

A) the degree to which the results correlate to something in the future
B) the validity predicted by the researcher before testing
C) the likelihood that a measure confirms a hypothesis
D) the reliability of the content in a measure

Question 11

Construct validity relies on which two elements?

A) the construct and tests of validity
B) reliability and subjective judgments
C) empirical data and statistical tests
D) subjective judgments and empirical data

Question 12

Construct validity relies on which two elements?

A) the construct and tests of validity
B) reliability and subjective judgments
C) empirical data and statistical tests
D) subjective judgments and empirical data

Question 13

To determine construct validity, researchers begin by hypothesizing what?

A) the relationship between components of the construct and other variables
B) the effects of assumptions on dependent variables
C) the relationship of the independent variables to the population
D) the mortality rate of participants

Question 14

A test with high reliability will necessarily have low validity

Question 15

Why is high validity more important than high reliability?

A) better to measure accurately than measure the wrong thing consistently
B) better to measure consistently than measure the wrong thing accurately
C) without high validity, reliability cannot be achieved
D) they are equally important

Question 16

According to this topic, most published tests have reliability coefficients of what?

A) about 0.60
B) about 0.70
C) about 0.80
D) about 0.90

Question 17

Which of the following describes a basic reliability test?

A) measure twice and compare the results
B) compute the mean of results and compare to hypothesis
C) have a second observer record the same data
D) divide the mean of results by the standard deviation

Question 18

What is a correlation coefficient?

A) the relationship between results and hypothesis
B) the efficiency with which results are correlated
C) the degree of relationship between participants
D) the degree of relationship between results

Question 19

What do interobserver reliability and parallel-form reliability have in common?

A) both are best used with large quantitative datasets
B) both correlate results between different data collection instruments
C) both are weak measures of reliability
D) both are used exclusively for qualitative studies

Question 20

Which two methods for estimating reliability require two testing sessions?

A) interobserver and parallel-form
B) correlation coefficient and interobserver
C) test-retest and parallel-form
D) test-retest and correlation coefficient

Question 21

How many test administrations are needed in the split-half method?

A) none
B) one
C) two
D) three or more

Question 22

Which reliability method for determining internal consistency is currently more common?

A) parallel-form
B) Cronbach's alpha
C) split-half
D) test-retest

Question 23

Internal consistency measures the reliability of which of these elements?

A) the entire test
B) the test takers' performance
C) the individual test items
D) the assessment given by different graders

Question 24

What is the highest possible value for a split-half reliability coefficient?

A) 0.0
B) 0.1
C) 1.0
D) 1.1

Question 25

To obtain Cronbach's alpha, calculations are used to obtain the equivalent of what?

A) median of all possible interobserver reliability coefficients
B) average of all possible split-half reliability coefficients
C) standard deviation of all possible reliability coefficients
D) average of all other alpha coefficients

Question 26

A norm-referenced test is designed to reveal what?

A) an individual's performance relative to his or her prior performances
B) a group's performance relative to results obtained from other tests
C) an individual's performance relative to that of a group's
D) an individual's performance relative to his or her results on other tests

Question 27

A criterion-referenced test is designed to reveal what?

A) an individual's performance relative to his or her prior performances
B) an individual's performance relative to that of a group's
C) an individual's content knowledge relative to that of a group's
D) an individual's ability to meet performance standards

Question 28

When designing a norm-referenced test, researchers will favor items that are answered correctly by roughly what percentage of examinees in tryouts of the test?

A) <25%
B) 35%
C) 50%
D) >70%

Question 29

Which of the following is an advantage of multiple-choice achievement tests over holistic performance-based assessments?

A) more valid because they can measure more skills
B) more reliable than subjective holistic assessments
C) multiple-choice achievement tests are more criterion-referenced
D) holistic assessments cannot be norm-referenced

Question 30

Which of the following best describes what a holistic performance-based assessment such as an IQ test measures?

A) innate (inborn) intelligence
B) one's particular likelihood to be successful in college
C) skills that have been acquired within a specific cultural setting
D) abilities independent of cultural influences and formal instruction

Question 31

Which type of test would be most likely to predict an individual's ability to achieve success as a physician?

A) aptitude test
B) achievement test
C) intelligence test
D) multiple-choice test

Question 32

Which type of test would be most likely to measure an individual's knowledge of botany?

A) aptitude test
B) achievement test
C) intelligence test
D) multiple-choice test

Question 33

Which type of test would be most likely to measure an individual's ability to solve problems?

A) aptitude test
B) achievement test
C) intelligence test
D) multiple-choice test

Question 34

What do researchers reduce by observing behavior unobtrusively?

A) social-desirability effects
B) Likert-scale testing bias
C) need to administer tests anonymously
D) margin of error

Question 35

Loosely structured stimuli are used in which type of personality measure?

A) Likert-scale testing
B) norm-referenced achievement
C) projective
D) social desirability

Question 36

Likert-scale tests present what range of choices?

A) multiple choice
B) strongly agree to strongly disagree
C) true or false
D) open ended

Question 37

Data triangulation describes what research technique?

A) comparing data from multiple studies
B) comparing past data to present results
C) obtaining data from multiple participants
D) obtaining related data by multiple methods

Question 38

Mixed-method describes what research technique?

A) comparing data from multiple studies of different types
B) comparing past data to present results
C) obtaining both qualitative and quantitative data
D) obtaining data from multiple sources

Question 39

When a qualitative research team compares the coding of data done independently by team members, when are the results considered dependable?

A) the results are consistent with quantitative data
B) there is interobserver agreement
C) the results are consistent with prior research
D) when each team member uses a different method