Deck 13: Inferential Statistics

Question 1

Researchers usually hope to reject the null hypothesis.

Answer 1

True

Question 2

Researchers reject the null hypothesis when the p value is very low.

Answer 2

True

Question 3

The p value is the probability of a sample result as extreme as the one you found-if the null hypothesis were true.

Answer 3

True

Question 4

You reject the null hypothesis when the p value is less than alpha.

Question 5

When a researcher selects a nonrepresentative sample, they have committed a sampling error.

Question 6

The purpose of null hypothesis testing is to decide whether or not a sample result reflects a real relationship in the population.

Question 7

A statistically significant result is one that has practical importance.

Question 8

Sampling error is the sample-to-sample variability in a statistic.

Question 9

The mean extroversion score for a population of people is a parameter.

Question 10

The null hypothesis is usually that there is a strong relationship between two variables.

Question 11

When a researcher rejects the null hypothesis, he or she is concluding that there is a relationship between two variables.

Question 12

The p value is the probability that the null hypothesis is true.

Question 13

An ANOVA is used to compare the means of more than two groups or conditions.

Question 14

A strong relationship based on a large sample will be statistically significant.

Question 15

You reject the null hypothesis when the critical value of your test statistic is more extreme than the actual value of your test statistic.

Question 16

The null hypothesis for a paired-samples t test is that the mean difference score is zero.

Question 17

A one-sample t test compares the mean of a sample to a hypothetical population mean of interest.

Question 18

In psychological research, alpha is usually set to .05.

Question 19

To compare men and women in terms of their aggression levels, you would probably use a paired-samples t test.

Question 20

Researchers use parameters to estimate statistics.

Question 21

Professional researchers often have misunderstandings about null hypothesis testing.

Question 22

When you reject the null hypothesis but it is actually true, you have committed a Type I error.

Question 23

Statistically significant results are more likely to be published than statistically nonsignificant results.

Question 24

When you conduct a paired-samples t test, what is the null hypothesis?

A) The mean of the difference scores is 0 in the population.
B) One group scored significantly higher than the other.
C) The mean score is equal to the midpoint of the response scale.
D) People scored higher on the posttest than on the pretest.

Question 25

You can increase the power of your studies by increasing the sample size.

Question 26

A student researcher finds a Pearson's r of +.75 for a sample of 240 people. Even without doing a formal null hypothesis test, what can you tell her about her result?

A) The result was probably due to chance.
B) The result is statistically significant.
C) The null hypothesis is true.
D) She made a mistake in computing Pearson's r.

Question 27

What does it mean if the result of your study is "not statistically significant?"

A) You reject the null hypothesis.
B) The result represents a "real" result that is true in the population.
C) The result is was due to sampling error.
D) The result is unimportant.

Question 28

Any time you reject the null hypothesis, there is some probability that you could be mistaken.

Question 29

Null hypothesis testing is the only approach to inferential statistics.

Question 30

A research methods student finds a large difference between the mean scores for two groups (d = 1.00) for a sample of 100. Does she reject the null hypothesis?

A) yes
B) no
C) maybe
D) It is impossible to say.

Question 31

Imagine that you read a magazine about a "statistically significant" relationship between the amount of fish oil people consume and how long they live. The result is based on a sample of 10,000. Which of the following is true?

A) The relationship could be extremely weak.
B) The relationship is extremely strong.
C) The researchers failed to reject the null hypothesis.
D) The relationship must be a causal one.

Question 32

In general, why do researchers do null hypothesis testing?

A) to decide whether a sample relationship represents a real relationship in the population
B) to decide how strong a sample relationship is
C) to decide whether the independent variable is causally related to the dependent variable
D) none of the above

Question 33

When you conclude on the basis of a null hypothesis test that there is no relationship between two variables in the population-but there really is a relationship between them in the population-you have committed a Type I error.

Question 34

Researchers in psychology usually set alpha to which of the following values?

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

Question 35

Because of the file drawer problem, published research is likely to underestimate the strength of a statistical relationship in the population.

Question 36

If you conduct a null hypothesis correctly, then the conclusion you draw is guaranteed to be correct.

Question 37

You are most likely to reject the null hypothesis when you have which of the following?

A) a weak relationship and a small sample
B) a weak relationship and a large sample
C) a strong relationship and a small sample
D) a strong relationship and a large sample

Question 38

Statistical power is the probability of rejecting the null hypothesis for a given sample size and expected relationship strength.

Question 39

If Pearson's r in a sample is 0.00, you should reject the null hypothesis.

Question 40

A researcher finds that the correlation between age and happiness is .09. What is most likely the null hypothesis?

A) There is no correlation between age and happiness in the population.
B) There is a positive correlation between age and happiness in the population.
C) There is no correlation between age and happiness in her sample.
D) There is a positive correlation between age and happiness in her sample.

Question 41

When you conduct a two-sample t test, what is the null hypothesis?

A) The two sample means are the same.
B) The two sample means are different.
C) The two population means are the same.
D) The two population means are different.

Question 42

Which of the following describes the relationship between inferential statistics and null hypothesis testing?

A) They are the same thing.
B) They are direct opposites.
C) Null hypothesis testing has replaced inferential statistics in psychology.
D) Null hypothesis testing is one approach to inferential statistics.

Question 43

Imagine that you collect data on the relationship between the number of hours college students work and how stressed they feel. Which test should you use?

A) independent t test
B) null hypothesis test on Pearson's r
C) paired t test
D) z test

Question 44

Imagine that you want to compare people's blood pressure before and after meditating for an hour. Which test should you use?

A) one-way ANOVA
B) paired-samples t test
C) two-sample t test
D) chi-squared test

Question 45

Which of the following is true of a study with inadequate statistical power?

A) It is likely to result in a Type I error.
B) It has a good chance of producing a Type II error.
C) The relationship in the sample will almost certainly be extremely weak.
D) The null hypothesis will probably be rejected.

Question 46

A researcher conducts a two-sample t test comparing the mean sociability levels of 20 psychology majors and 15 chemistry majors. How many degrees of freedom are there for this test?

A) 1
B) 5
C) 33
D) 34

Question 47

What is a Type II error and how can it occur even though the researcher conducts the null hypothesis test correctly?

Question 48

If the value of Pearson's r in your sample is .45, what should you do?

A) reject the null hypothesis
B) fail to reject the null hypothesis
C) recompute Pearson's r because you must have made a mistake
D) There is not enough information to know what to do.

Question 49

Imagine that you want to compare the conscientiousness of men and women. Which test should you use?

A) one-sample t test
B) paired-samples t test
C) two-sample t test
D) null hypothesis test on Pearson's r

Question 50

Imagine you conduct a study and find that the correlation between the height of female college students and the heights of their boyfriends is +.36 and that the p value for this correlation is .02. This p value is the probability of something. But what is it the probability of?

Question 51

Imagine that a researcher has conducted a study and found that students exposed to a new method of learning fractions perform better on an end-of-year assessment than students exposed to the old method. Explain why this researcher conducts a null hypothesis test.

Question 52

When a study lacks statistical power, the researcher is in danger of which of the following?

A) committing a Type I error
B) committing a Type II error
C) committing a sampling error
D) all of the above

Question 53

Fred conducts a paired-samples t test. His critical t value is 2.36 and his actual t value is 4.17. Which of the following is true?

A) He should have conducted a two-sample t test.
B) His p value is greater than .05.
C) He should reject the null hypothesis.
D) He should fail to reject the null hypothesis.

Question 54

Which of the following is true of the statistical power?

A) It is equal to the probability of a Type I error.
B) It is two times the probability of a Type I error.
C) It is one minus the probability of a Type II error.
D) none of the above

Question 55

What null hypothesis test would you use to compare the mean empathy levels of three-year-olds and five-year-olds? Why?

Question 56

What is a Type I error and how can it occur even though the researcher conducts the null hypothesis test correctly?

Question 57

A research methods student carries out a null hypothesis test. She does everything correctly and rejects the null hypothesis. It is still possible, though, that she has done which of the following?

A) committed a Type I error
B) committed a Type II error
C) committed a sampling error
D) none of the above

Question 58

Imagine that you know the mean number of alcoholic drinks per week consumed by American college students. Now you want to compare the mean number for a sample at your university to the mean for American college students. Which test should you use?

A) single-sample t test
B) paired-samples t test
C) independent-samples t test
D) ANOVA

Question 59

The file drawer problem is caused partly by which of the following?

A) universities rewarding researchers for conducting lots of research
B) too much statistical power on the part of journal editors
C) scientists' skeptical attitude toward new results
D) journal editors' preference for publishing significant results

Question 60

What is the difference between statistical significance and practical significance? Give a concrete example to illustrate your answer.

Question 61

What is statistical power? What can researchers do to increase the power of their studies?