Unit 5 practice (notes)

Question 1

Which of the following r values is the strongest?
a. + 0.45
b. – 0. 72
c. + 0 .71
d. + 0.00

Answer 1

b

Question 2

Golf ability and golf scores are ________ correlated.
a. not
b. positively
c. negatively
d. inversely

Answer 2

c

Question 3

If you wanted to visually show somebody the correlation between his or her guesses of the
caloric value of certain foods and the actual caloric value of those foods you would produce a
____________.
a. line graph
b. bar graph
c. pie graph
d. scatterplot graph

Answer 3

d

Question 4

Strong correlations are those that have:
a. r values between 0.3 and 0.4.
b. r values between 0.7 and 0.8.
c. r values between 0.8 and 0.99.
d. r values between – 0.7 an – 0.8.

Answer 4

c

Question 5

You would probably agree that income and homelessness should be negatively correlated
(more income, less homelessness). However, in many large cities income is positively
correlated with homelessness (more incomes, more homelessness). How can this be?
a. A third variable, like housing cost, is positively correlated with both income and
homelessness and thus creates a spurious relationship.
b. The positive correlation exists because it should be looked at from the other perspective,
homelessness is positively correlated with income.
c. Both a and b are correct.
d. It can’t be. Make sure SPSS is working properly.

Answer 5

a

Question 6

Which of the following is most likely to yield data that is nominal, ordinal, and interval?
a. descriptive survey
b. analytical survey
c. test and inventory
d. both b and c

Answer 6

d

Question 7

A test format that requires participants to choose two of five words or phrases that seem
unrelated but equally acceptable is called:
a. multiple-choice.
b. forced choice.
c. Likert scale.
d. open-ended question.

Answer 7

b

Question 8

When participants choose responses on a survey or test that make them look good they are
creating bias or error. This bias or error is termed:
a. social acceptance.
b. social dissonance.
c. social desirability.
d. faking good.

Answer 8

c

Question 9

Which type of closed-end question best provides the researcher with data that can be used to
calculate means and standard deviations?
a. Only open-ended questions can give you this type of data.
b. forced choice questions
c. yes-no questions
d. Likert scale questions

Answer 9

d

Question 10

Eye, skin, hair, and car colour, are examples of:
a. spurious variables.
b. personal variables.
c. demographic variables.
d. dependent variables.

Answer 10

c

Question 11

Your test to measure humour in participants is complete. Before you begin to mass market it you
give it to a small group of people and ask for their opinions on the test as a whole and on
individual items. In addition, you take the test to experts in humour research. This process is
termed ___________.
a. piloting
b. generating data
c. revision
d. marketing

Answer 11

a

Question 12

Test-retest, alternate forms, internal consistency, and split-half all refer to:
a. test development.
b. piloting.
c. reliability.
d. validity.

Answer 12

c

Question 13

Jimmy wants to investigate how reliable his test is so he gives the test to the same group of
participants on two separate occasions. Jimmy is using:
a. test-retest reliability.
b. alternative forms reliability.
c. split-half reliability.
d. internal consistency.

Answer 13

a

Question 14

The greatest fault in using the test-retest method to provide a measure of reliability is that most
participants will score better on the second administration of the test because of:
a. better conditions of testing.
b. practice effects.
c. social desirability.
d. better understanding of test instructions and thus better mood

Answer 14

b

Question 15

KR-20 and coefficient alpha are both used to measure:
a. test-retest reliability.
b. alternative forms reliability.
c. error.
d. internal consistency of a test.

Answer 15

d

Question 16

Mary thinks her test is too short, so she adds 12 relevant questions. Mary has just:
a. increased her reliability.
b. decreased her reliability.
c. made her test more homogeneous.
d. made her test less homogeneous.

Answer 16

a

Question 17

Considering that all things are equal, which method should you choose to provide the best
possible estimate of reliability?
a. test-retest reliability
b. alternate forms reliability
c. split-half reliability adjusted with the Spearman-Brown formula
d. Cronbach’s alpha

Answer 17

d

Question 18

What is the difference between validity and reliability?

Answer 18

reliability: consistency of a test or measurement
validity: extent to which a test or measurement is in fact measuring what it is supposed to measure
These two concepts can be best distinguished via an example. Consider that I have a simple test to measure your knowledge in PSYC 2250. The test consists of two simple measurements: the circumference of your head and the distance between your eyes. Now you may say that this is a ridiculous test of PSYC 2250 knowledge, and you would be correct since there is no correlation between my two measures and PSYC 2250 knowledge. In other words, the test is not valid since head circumference and eye distance have very little to do with PSYC 2250 knowledge. But I say the test is reliable since I get the same result each time I measure your head circumference and eye distance. Thus, just because a test is reliable it is not necessarily valid. However, if a test is shown to have good validity it is most often reliable since reliable tests have little error and thus they actually measure something, and hopefully that something is what you intended to
measure. Unreliable tests measure mostly error and thus you cannot predict what they are in fact measuring.

Question 19

Tom constructs a test to measure aptitude. He gives the test to all applicants seeking a job with
IBM. A year later Tom discovers that scores on his aptitude test correlate strongly with job
performance. This type of validity is termed:
a. face validity.
b. content validity.
c. construct validity.
d. predictive criterion-related validity.
e. concurrent criterion-related validity.

Answer 19

d

Question 20

___________ refers to whether some test measures what it claims to measure.
a. Reliability
b. Validity
c. Spearman-Brown coefficient
d. Coefficient alpha

Answer 20

b

Question 21

Stella, 34, is interested in knowing her IQ. She visits a psychologist who gives her an IQ test.
The test is printed on bright orange paper and the first question instructs Stella to cut out circles
with a pair of left-handed scissors. Stella laughs but half heartedly begins to cut circles. From
Stella’s perspective this IQ test:
a. lacks face validity.
b. is unique but would be better suited for children.
c. has strong content validity.
d. has good construct validity

Answer 21

a

Question 22

Sharon has spent the last two years in the library learning every detail about anorexia and
bulimia. She constructs an intricate flow chart, which assigns different weights to likely
precipitating factors. She then uses this flow chart as a guide and constructs a test aimed at
predicting the likelihood of anorexia and bulimia. Sharon’s test likely has good ______________
validity, but the most important type of validity for her test will be ______________ and this will
take some time to gather.
a. concurrent validity; content validity
b. face validity; content validity
c. content validity; predictive validity
d. content validity; construct validity

Answer 22

c

Question 23

Which of the following is not a type of validity?
a. face validity
b. retest validity
c. construct validity
d. concurrent validity

Answer 23

b

Question 24

If a singing audition or test requires that you submit a representative sample of your singing
ability (i.e., a recording of your singing) this provides evidence for:
a. criterion-related validity.
b. content validity.
c. construct validity.
d. predictive validity.

Answer 24

b

Question 25

Which type of validity is important to all types of tests?
a. construct
b. content
c. criterion-related
d. predictive

Answer 25

b

Question 26

Carl studies how much people exercise and personality. He finds a correlation of + 0.7 between
time spent a week exercising and happiness. In other words, the more time people spent
exercising the happier they were. Which of the following statements is true?
a. There is no doubt that Carl’s finding will be stronger for 70% of the population.
b. If you want to be happy, exercise more.
c. If something happens in your life to make you sad, you will likely exercise less.
d. Persons who exercise more tend to be happier.

Answer 26

d

Question 27

Consider the following two Pearson r values: + 0.6 and +0.8. Which statement is true?
a. It is more reliable to draw cause-and-effect conclusions with +0.8 than +0.6.
b. It is more valid to draw cause-and-effect conclusions with +0.8 than +0.6.
c. Points on a scatterplot will be closer together for +0.6 than +0.8.
d. Points on a scatterplot will be closer together for +0.8 than +0.6.
e. None of the above are even remotely true.

Answer 27

d

Question 28

If a study is _____________ then it must be ______________.
a. reliable; valid
b. valid; reliable
c. both a and b are correct
d. none of the above are incorrect.

Answer 28

b