Deck 5: Inferential Statistics: The Controlled Experiment, Hypothesis Testing, and the Z Distribution

Question 1

When findings in an experiment do not reach the conventional level of statistical significance, they are reported to be ___________.

A)nonsignificant
B)insignificant
C)significant
D)unworthy

Answer 1

nonsignificant

Question 2

If a magician claims that he can bend spoons with his mind, what did Carl Sagan propose to test the claim?

A)a lie detector test
B)the baloney detection kit
C)the salami detection kit
D)the Atkins detection kit

Answer 2

the baloney detection kit

Question 3

Which of the following is a Type I error?

A)rejecting H₀ when H₀ is false
B)rejecting H₀ when H₀ is true
C)not rejecting H₀ when H₀ is true
D)not rejecting H₀ when H₀ is false

Answer 3

rejecting H₀ when H₀ is true

Question 4

The probability of committing the Type I error is also called ___________.

A)alpha
B)beta
C)delta
D)omega

Question 5

In the section "Trends, and Does God Really Love the .05 Level of Significance More Than the .06 Level?" Rosnow and Rosenthal have argued that the strength of evidence for or against a null hypothesis should be a fairly continuous function of the size of

A)p (the significance level)
B)beta
C)the sample size
D)the size of the standard deviations for each group

Question 6

The minimum conventional level of statistical significance is ___________.

A).01
B).05
C).10
D).50

Question 7

The difference between two means in a controlled experiment that is just attributed to chance or random error is called ___________ in the signal-to-noise ratio.

A)signal
B)to
C)noise
D)ratio

Question 8

In the section "Can Statistics Solve Every Problem?" your course author argues

Question 9

A dependent variable in an experiment can be likened to the ___________ in the signal-to-noise ratio.

A)signal
B)to
C)noise
D)ratio

Question 10

Which of the following is true about correlational designs?

A)Finding a relationship between two variables does not mean that one causes the other but could give clues to set up experiments where causality may be determined.
B)Finding a relationship between two variables NEVER means that one causes the other.
C)Correlational designs are not very common.
D)Correlational designs are more powerful than a controlled experiment.

Question 11

Which of the following is more likely to end up being too sensitive to chance differences?

A)a nondirectional alternative hypothesis
B)a directional alternative hypothesis
C)a nondirectional research hypothesis
D)all of the above

Question 12

Because it is said in science "one experiment does not prove anything," statisticians rely on ___________ to test the usefulness of theories.

A)a controlled experiment
B)at least two experiments
C)parametric and nonparametric tests
D)replication

Question 13

The probability of committing the Type II error is also called ___________.

A)alpha
B)beta
C)delta
D)omega

Question 14

A trend in the data means that the experimenter

A)rejected the null hypothesis at p < .05
B)rejected the null hypothesis at p < .01
C)did not reject the null hypothesis but came close to doing so
D)none of the above

Question 15

In the classic two-group controlled experiment, what parameter is of central interest between the two groups?

A)the means
B)the standard deviations
C)the sample size
D)the median