Deck 16: Making Inferences From Sample Data I: The Null Hypothesis Significance Testing Approach

Question 1

A t value has a probability of .05. This means that:

A) There is a probability that 5 times out of 100 the alternative hypothesis is true
B) There is a probability that 5 times out of 100 the null hypothesis is true
C) There is a probability that 5 times out of 100 the null hypothesis will be rejected when it is false
D) There is a probability that 5 times out of 100 the null hypothesis will not be rejected when it is true

Answer 1

There is a probability that 5 times out of 100 the null hypothesis is true

Question 2

A null hypothesis states that there will be no significant differences in the population between the average weights of one group of women who have sports-related hobbies versus a group of women with non-sports-related hobbies. A Type I error would be to:

A) Reject the null hypothesis when it is true
B) Reject the null hypothesis when it is false
C) Not reject the null hypothesis when it is false
D) Not reject the null hypothesis when it is true

Answer 2

Reject the null hypothesis when it is true

Question 3

A statistical outcome that could occur less than 5 times in 100 (alpha = .05) should result in:

A) Rejection of the alternative hypothesis
B) Rejection of the null hypothesis
C) Not rejecting the null hypothesis
D) a and c

Answer 3

Rejection of the null hypothesis

Question 4

Type II error refers to:

A) The Beta
B) The probability of not rejecting the null hypothesis when it is false
C) The probability of rejecting the null hypothesis when it is true
D) a and b

Question 5

What significance level for a directional hypothesis is the same as .05 for a non-directional hypothesis?

A) .05
B) .025
C) .10
D) .01

Question 6

By divine revelation, a researcher discovers that his hypothesis was, in fact, true and he should never have rejected it. The researcher:

A) Did not reject the null hypothesis when it is true
B) Did reject the null hypothesis when it is false
C) Did reject the null hypothesis when in fact it is true
D) Did not reject the null hypothesis when it is false

Question 7

The situation above (problem 7) is an example of:

A) Type I error
B) Type II error
C) No significant differences between the population and sample means
D) a and c

Question 8

There are a number of ways to state the alternative hypothesis. Select which applies:

A) The population mean of the bird-watching intervention will be significantly different from the population mean of the non-bird-watching control group
B) The sample mean of the bird-watching intervention will be significantly different from the sample mean of the non-bird-watching control group
C) The population mean of the bird-watching intervention will be greater than the population mean of the non-bird-watching control group
D) a and c

Question 9

To estimate power, a researcher will need to know:

A) Size of the sample
B) The significance level (α)
C) The estimated effect size of the relationship
D) All of the above

Question 10

When calculating power, how does a researcher know what estimate to use for the effect size?

A) The researcher should always assume there will be a large effect size
B) The researcher should always assume there will be a medium effect size
C) The researcher should investigate past studies in the same content area to estimate the effect size
D) There is no single way to estimate effect size

Question 11

To increase power:

A) Alpha can be increased (i.e., set at .10 instead of .05)
B) A directional hypothesis can be tested
C) The error variance can be decreased
D) The number of participants can be increased
E) All of the above

Question 12

Which of the following is true about statistical power?

A) The greater the power, the more likely one can detect a true difference between groups
B) Power is the probability of rejecting a false null hypothesis
C) Power is the same as alpha, or the probability of a Type I error
D) The greater the power, the more generalizable the results
E) a and b

Question 13

Statistical power can be gained by:

A) Manipulating the alpha level
B) Decreasing participant heterogeneity
C) Using a measure for the DV that has documented reliability for a study such as yours
D) All of the above

Question 14

Rex All developed what he thinks is a miracle drug to cure baldness. He asks you for advice on how to maximize statistical power so that he can "prove" that Hair Today really works. You tell him that the best way to maximize power is to:

A) Increase the sample size
B) Set the alpha level at .10
C) Use a nondirectional hypothesis
D) Use a directional hypothesis

Question 15

A p value tells us:

A) Given these data, what is the probability of H0?
B) Given that H0 is false, what is the probability of these data?
C) Given that H0 is false, what is the probability that H0 may be true?
D) Given that H0 is true, what is the probability of these data?
E) None of the above

Question 16

Complications of NHST include:

A) The null hypothesis is rarely true
B) The null hypothesis is directional
C) A smaller p value indicates stronger relationship
D) A p value gives important information regarding substantive and clinical significance

Question 17

Null and alternative hypotheses refer to differences between sample means.

Question 18

Null and alternative hypotheses are stated in terms of population means.

Question 19

Nondirectional alternative hypotheses often are used when comparing two different treatment methods.

Question 20

We can never be sure if a decision based on sample data is actually true for the population.

Question 21

Directional alternative hypotheses often are used when comparing a treatment to a control condition.

Question 22

Directional alternative hypotheses should be used when there is previous research to support the intervention.

Question 23

The significance level or alpha should be determined at the end of the study.

Question 24

It is more difficult to reject the null hypothesis using a directional hypothesis.

Question 25

If the calculated value of an F statistic is relatively large, the probability is small.

Question 26

To interpret fully an inferential statistic, it is only necessary to know if p < .05.

Question 27

Power is the probability of rejecting a false null hypothesis.

Question 28

It is best to have power be less than .05.

Question 29

Power from a statistical point of view refers to the probability of not rejecting a false null hypothesis.

Question 30

Ideally, one should aim for a power of .80.

Question 31

Power can be determined by knowing the sample size, the significance level, and an estimate of the study's effect size.

Question 32

In order to gain power, the alpha level should be set at .01.

Question 33

Within-group variability decreases the statistical power of the study.

Question 34

An increased sample size will increase variability and decrease statistical power.

Question 35

A small sample size may not produce enough power to reject a false null hypothesis.

Question 36

The only proven method for increasing power is to increase the sample size.

Question 37

When estimating the effect size when calculating power, it is best to use .50, regardless of the content area being researched.

Question 38

The cut point of p < .05 is artificial.

Question 39

A p value of .06 should add supporting evidence to the results of a study.

Question 40

What is important about choosing a direction or non-directional hypothesis?

Question 41

Can the null hypothesis be accepted?