Deck 6: Hypothesis Testing

Question 1

Typically, the null hypothesis is the statement that we are interested in _____.

A) Accepting
B) Disproving
C) Proving
D) Rejecting

Answer 1

D

Question 2

Consider a null hypothesis in which we state that the mean mass of male and female robins in a population is the same. Which of the following is an appropriate alternative hypothesis?

A) The mean mass of female robins is different from the mean mass of male robins.
B) The mean mass of female robins is equal to the mean mass of male robins.
C) The mean mass of female robins is less than the mean mass of male robins.
D) The mean mass of female robins is more than the mean mass of male robins.

Answer 2

A

Question 3

Which of the following is the best description of what a "two-sided" test means?

A) Either the null or alternative hypothesis may be false.
B) Either the null or alternative hypothesis may be true.
C) The alternative hypothesis includes parameter values of both sides of the null hypothesis parameter value.
D) The alternative hypothesis includes statistical values of both sides of the null hypothesis statistic value

Answer 3

C

Question 4

The sampling distribution for a test statistic if the null hypothesis is true is called which of the following?

A) The hypothetical distribution
B) The null distribution
C) The test distribution
D) The true distribution

Question 5

When presenting your results in a research paper or report, which of the following is not one of the pieces of information you should always include?

A) The P-value
B) The range of values
C) The sample size
D) The value of the test statistic

Question 6

Consider a hypothetical study in which ecologists looked at the sex ratios of clutches of frog eggs. In their sample, there were more clutches with a majority of males compared to those with a majority of females. The P-value of their test was 0.08. What is their best conclusion based on the evidence provided?

A) The frequencies of male and female majority clutches seem to differ in the population.
B) The frequencies of male and female majority clutches don't seem to differ in the population.
C) There are more female majority clutches than male majority clutches in the population.
D) There are more male majority clutches than female majority clutches in the population.

Question 7

Consider a hypothetical study in which ecologists looked at the sex ratios of clutches of frog eggs. In their sample, there were more clutches with a majority of males compared to those with a majority of females. The P-value of their test was 0.008. What is their best conclusion based on the evidence provided?

A) The frequencies of male and female majority clutches seem to differ in the population.
B) The frequencies of male and female majority clutches don't seem to differ in the population.
C) There are more female majority clutches than male majority clutches in the population.
D) There are more male majority clutches than female majority clutches in the population.

Question 8

Increasing the value of the significance level tends to do which of the following?

A) Increases Type I error and increases Type II error
B) Increases Type I error and reduces Type II error
C) Reduces Type I error and increases Type II error
D) Reduces Type I error and reduces Type II error

Question 9

The power of a statistical test is best described as which of the following?

A) The probability that the test will correctly accept a false alternative hypothesis.
B) The probability that the test will correctly accept a false null hypothesis.
C) The probability that the test will correctly reject a false alternative hypothesis.
D) The probability that the test will correctly reject a false null hypothesis.

Question 10

The ability of a statistical test to correctly reject a false null hypothesis is called which of the following?

A) Accuracy
B) Power
C) Precisions
D) Reliability

Question 11

Which of the following statements is correct?

A) Larger P-values will tend to give nonsignificant results.
B) Larger P-values will tend to give significant results.
C) Lower P-values will tend to give insignificant results.
D) Lower P-values will tend to give unsignificant results.

Question 12

Which of the following is our best attitude to a test result that returns a nonsignificant result?

A) The null hypothesis is conclusively demonstrated to be correct.
B) The null hypothesis is conclusively demonstrated to be incorrect.
C) The overall study is inconclusive and the null hypothesis is incorrect.
D) The overall study is inconclusive and we shouldn't make a definitive decision.

Question 13

Which of the following is true for a one-sided test?

A) The alternative hypothesis specifies a range of values.
B) The null hypothesis includes only a single value for the parameter of interest.
C) We calculate our P-value using the extremes of the null distribution.
D) We make a decision about using this test after looking at the data.

Question 14

Which of the following is correct?

A) We generally prefer doing one-sided tests followed by a two-sided test.
B) We generally prefer doing two-sided tests followed by a one-sided test.
C) We generally prefer doing one-sided tests unless we have a genuinely good reason to do a two-sided test.
D) We generally prefer doing two-sided tests unless we have a genuinely good reason to do a one-sided test.

Question 15

If a two-sided statistical test returns a P-value less than 0.05, then what can be said about the parameter value specified in the null hypothesis?

A) It is guaranteed to be inside the 95% confidence interval for the parameter.
B) It is guaranteed to be outside the 95% confidence interval for the parameter.
C) It is very likely inside the 95% confidence interval for the parameter.
D) It is very likely outside the 95% confidence interval for the parameter.

Question 16

Consider a situation in which we calculate a 95% confidence interval that ranges from 18 to 23. If we conducted a two-sided test with the null hypothesis of the population mean equaling 17, what would the likely result of our test be?

A) A P-value larger than 0.05 and we fail to reject the null hypothesis.
B) A P-value larger than 0.05 and we reject the null hypothesis.
C) A P-value smaller than 0.05 and we fail to reject the null hypothesis.
D) A P-value smaller than 0.05 and we reject the null hypothesis.

Question 17

Consider a situation in which we calculate a 95% confidence interval that ranges from 35 to 45. If we conducted a two-sided test with the null hypothesis of the population mean equaling 43, what would the likely result of our test be?

A) A P-value larger than 0.05 and we fail to reject the null hypothesis.
B) A P-value larger than 0.05 and we reject the null hypothesis.
C) A P-value smaller than 0.05 and we fail to reject the null hypothesis.
D) A P-value smaller than 0.05 and we reject the null hypothesis.

Question 18

Statistical hypotheses are statements about population values whereas scientific hypotheses are statements about natural phenomena that may account for population values.

Question 19

Hypothesis testing quantifies how unusual the data are, assuming that the alternative hypothesis is true.

Question 20

The null hypothesis is a specific claim about the value of a sample statistic.

Question 21

The alternative hypothesis is usually the hypothesis we would be most interested in determining is correct.

Question 22

The null distribution is always centered around zero.

Question 23

The y-axis for a null distribution is probability values.

Question 24

The most commonly used significance level in biology is a value of 0.05.

Question 25

The probabilities of Type I and Type II errors always add up to 1.0.

Question 26

As studies increase in sample size, they tend to have more statistical power.

Question 27

Nonsignificant tests are uninformative.

Question 28

Generally speaking, if we fail to reject the null hypothesis, we should therefore conclude that the null hypothesis is true.

Question 29

It is recommended practice to look at the data values prior to deciding whether to perform a one-sided or two-sided test.

Question 30

If a one-sided test rejects the null hypothesis, then a two-sided test would have rejected the null hypothesis as well.

Question 31

When calculating the 95% confidence intervals for the parameter value used in a nonsignificant statistical test, the region would usually (but not always) include the value specified in the null hypothesis.

Question 32

Describe the relationship between a statistical hypothesis and a scientific hypothesis.

Question 33

Is the null hypothesis always what we expect or hope to be true? Explain why or why not.

Question 34

In your own words, describe what a P-value is and how we use it to evaluate statistical hypotheses.

Question 35

When describing the results of a study, we should report three things. What are these three values and why should we include each of them?

Question 36

Describe the U.S. criminal justice system's assumptions about guilt and innocence using Type I and Type II terminology. Include reference to the concept of significance level as used in the system.

Question 37

Describe the major problem with using a nonsignificant result as evidence that the null hypothesis is true.

Question 38

Describe a data set and make two scientific hypotheses: one must be appropriate for studying with a two-sided test and the other must be appropriate for studying with a one-sided test.

Question 39

There is a risk relating to P-values when we look at our data before deciding on a one-sided or two-sided test. Describe this risk with specific reference to the P-value we calculate if we do the wrong test.

Question 40

Describe the relationship between the P-value of a two-sided statistical test and the confidence interval around the value listed in the null hypothesis.