Question 1

Professional researchers often have misunderstandings about null hypothesis testing.

Accepted Answer

It is common for professional researchers to misunderstand the concept and interpretation of null hypothesis testing, leading to potential errors or misinterpretations in their research findings.

Question 2

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

Accepted Answer

This is the definition of a Type I error.

Question 3

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

Accepted Answer

This is a well-known phenomenon called publication bias, where journals are more likely to publish studies with statistically significant results than those without. This can lead to an overestimation of the true effects of a treatment or intervention in the published literature.

Question 4

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.

Accepted Answer

The null hypothesis for a paired-samples t test is that there is no significant difference between the mean of the difference scores and 0 in the population.

Question 5

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

Accepted Answer

Increasing the sample size can increase the power of studies by reducing the likelihood of chance findings and increasing the representativeness of the sample.

Question 6

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.

Accepted Answer

A Pearson's r of +.75 is a strong correlation and with a sample size of 240, it is likely statistically significant. However, without a formal statistical test, we cannot say for certain.

Question 7

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.

Accepted Answer

"Not statistically significant" means that the results were likely due to chance or sampling error, rather than a true difference in the population. Therefore, we cannot reject the null hypothesis and the result should be interpreted with caution.

Question 8

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

Accepted Answer

This is because the null hypothesis is assumed to be true until proven otherwise, so if you reject it based on a sample or experiment, there is always a chance that the result was due to chance or error rather than a true effect in the population. This probability is typically calculated as the p-value, which represents the probability of observing a result as extreme or more extreme than the one obtained, assuming that the null hypothesis is true. A commonly used significance level or alpha level is 0.05, meaning that if the p-value is less than or equal to 0.05, the result is considered statistically significant and the null hypothesis is rejected. However, this does not guarantee that the result is true or replicable, and there is always a chance of making a Type I error (false positive) or a Type II error (false negative) in hypothesis testing.

Question 9

Null hypothesis testing is the only approach to inferential statistics.

Accepted Answer

While null hypothesis testing is a commonly used approach to inferential statistics, it is not the only approach. Other approaches include confidence intervals, Bayesian inference, and permutation testing, among others.

Question 10

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.

Accepted Answer

A large difference of d = 1.00 between the means for two groups with a sample size of 100 is likely to result in a significant difference and rejection of the null hypothesis, especially if the appropriate statistical tests and significance level are used. Therefore, the student should reject the null hypothesis.

Question 11

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.

Accepted Answer

Just because the relationship is statistically significant does not mean that it is strong. It simply means that the relationship is unlikely to be due to chance. Additionally, the study does not establish causation.

Question 12

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

Accepted Answer

Researchers do null hypothesis testing to determine whether a sample relationship (such as a correlation or difference between groups) represents a real relationship in the population, or if it could have happened by chance. This is typically done by setting up a null hypothesis, which assumes that there is no true relationship in the population, and then testing whether the sample data supports or contradicts this null hypothesis. By doing so, researchers are able to draw conclusions about the population based on the sample data.

Question 13

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.

Accepted Answer

If you conclude that there is no relationship between two variables in the population based on a null hypothesis test, and there really is a relationship between them in the population, you have actually committed a Type II error. A Type I error occurs when you reject a null hypothesis that is actually true.

Question 14

Researchers in psychology usually set alpha to which of the following values?
A) .05
B) .50
C) .01
D) .10

Accepted Answer

Researchers in psychology usually set alpha to .05, meaning that they accept a 5% chance of making a Type I error (incorrectly rejecting a true null hypothesis). This is a common convention in the field and helps to establish a standard for interpreting statistical results.

Question 15

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

Accepted Answer

The file drawer problem refers to the bias introduced by the tendency for journals to publish studies with statistically significant results, while studies with non-significant results (null results) are less likely to be published. This leads to an overestimation of the strength of statistical relationships in the published literature, not an underestimation.

Question 16

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

Accepted Answer

Conducting a null hypothesis correctly only ensures that the statistical test was carried out properly. However, it does not guarantee that the conclusion drawn from the test is correct. The conclusion is based on the interpretation of the results and the context of the study, which can be influenced by various factors that were not considered in the statistical analysis. Therefore, it is important to critically evaluate the results and consider other factors before drawing conclusions.

Question 17

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

Accepted Answer

A large sample size increases the power of the test, making it more likely to detect a significant difference even with a strong relationship. A strong relationship between variables also increases the likelihood of rejecting the null hypothesis as it indicates a greater difference between groups. Therefore, option D - a strong relationship and a large sample - is the most likely scenario for rejecting the null hypothesis.

Question 18

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

Accepted Answer

This is the correct definition of statistical power.

Question 19

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

Accepted Answer

If Pearson's r in a sample is 0.00, it means there is no correlation between the two variables being measured. The null hypothesis in this case would be that there is no correlation. Therefore, there is no reason to reject the null hypothesis.

Question 20

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.

Accepted Answer

The null hypothesis is typically that there is no relationship or no difference between variables in the population. In this case, the null hypothesis is that there is no correlation between age and happiness in the population, meaning the observed correlation of .09 could be due to chance.

Quiz 13: Inferential Statistics

Professional researchers often have misunderstandings about null hypothesis testing.

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

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

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

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

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?

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

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

Null hypothesis testing is the only approach to inferential statistics.

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?

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?

In general, why do researchers do null hypothesis testing?

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.

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

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

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

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

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

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

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