Deck 26: Ethics in Statistical Studies

Question 1

How can the 'experimenter effect' be minimized?

A)Third-party randomization.
B)Double-blind procedures.
C)A standard protocol for the treatment of all participants that must be strictly followed.
D)All of the above.

Answer 1

All of the above.

Question 2

If a researcher has a desired outcome for a study and analyzes the data using different methods until the desired outcome is reached, this is called 'data __________.'

Answer 2

snooping

Question 3

Explain why it would not be ethical to report the sample size of a survey without also reporting the percentage or number of people who responded to it.

Answer 3

the results are only based on the respondents, and if they differ from nonrespondents, this could bias the results.

Question 4

Name a situation where the quality of data becomes an ethical issue.

Question 5

Which of the following would not fall under the requirements of informed consent?

A)Telling participants about any foreseeable discomfort or pain that might be involved.
B)Telling participants which of the drug treatments they will receive before the experiment begins.
C)Telling participants they have a right to withdraw from a study once it has started.
D)All of the above statements fall under the requirements of informed consent.

Question 6

Give two benefits of computer-assisted information collection (versus other methods).

Question 7

The 'experimenter effect' can __________the results of a study.

Question 8

Which of the following pertains to the issue of humane care and use of animals in research, as outlined by the American Psychological Association?

A)You can use procedures subjecting animals to pain, stress, or privation, but only when an alternative procedure is unavailable and the goal is justified by its prospective scientific, educational, or applied value.
B)In certain cases it is necessary that an animal's life must be terminated; in those cases, psychologists must proceed rapidly, with an effort to minimize pain and in accordance with accepted procedures.
C)Animals do not have the same rights as human participants in experiments.
D)All of the above.

Question 9

How is it possible that two different statistical analyses of the same data set can produce different results and more importantly, different conclusions?

Question 10

Which of the following should be decided before the data have been collected?

A)What level of confidence or level of significance should be used?
B)How will outliers be handled?
C)Which statistical method for analyzing the data is the most appropriate?
D)All of the above.

Question 11

If a researcher has a desired outcome for a study and conditions are not very carefully controlled, it is quite likely that the researcher will influence the outcome.This is called the __________ effect.

Question 12

When should a researcher decide whether an alternative hypothesis should be one-sided or two-sided?

A)Before collecting the data.
B)While the data are being collected, and a pattern begins to emerge.
C)After the data have been collected and the researcher has had a chance to summarize them to see what direction the results may take.
D)No single rule will apply to all situations; each one is different.

Question 13

What is 'data snooping' and why is it a practice to be avoided?

Question 14

Name three things that psychologists are supposed to inform participants about when obtaining informed consent, according to the American Psychological Association's code of ethics.

Question 15

Which of the following is an example of the 'experimenter effect'?

A)The results of a study confirmed what the researcher was hoping they would confirm.
B)Researchers who interacted with participants knew which treatments were given to which participants before the data collection process began, and had certain expectations for the outcomes.
C)The researchers were not properly trained on how to record the data and this affected the accuracy of the results.
D)All of the above.

Question 16

Explain what it means for a federal statistical agency to practice a 'strong position of independence,' according to the National Research Council's Committee on National Statistics.

Question 17

A board that all research institutions are required to maintain for oversight of research involving human and animal participants is called a(n) _______________ Board.

Question 18

Virtually all experiments with human participants __________that the researchers obtain the informed consent of the participants.

Question 19

When should a researcher decide which statistical method to use to analyze their data (assuming there is more than one statistical method available)?

A)Before collecting the data.
B)Once the data have been collected and summarized and the researcher can see what direction the results may take, then he/she needs to choose an analysis.
C)Once the data have been collected, the researcher should try different analyses and select the one that gives them the most desirable results.
D)None of the above.

Question 20

A researcher is required to get your 'informed consent' before you participate in a study.List two major requirements that he/she has to meet in order to do this correctly.

Question 21

Running multiple tests on the same data set at the same stage of an analysis __________ (choose from: increases, decreases, doesn't affect) the chance of obtaining at least one invalid result.

Question 22

Whether or not a study achieves statistical significance depends on which of the following?

A)The magnitude of whatever effect or relationship may actually exist.
B)The size of the study.
C)The power of the study to detect an effect that would result from the given sample size.
D)All of the above.

Question 23

Suppose a researcher conducts 100 hypothesis tests on her large data set and reports on the five results that were found to be significant, but doesn't inform the reader about the other 95 tests that were done.Is there a problem with this? Explain why or why not.

Question 24

Explain the impact resulting from the selective reporting of statistical results by the media.

Question 25

Which of the following is an example where the conclusions made are stronger than are justified?

A)"The correlation between the two variables in this observational study was found to be statistically significant (p<.05), which implies a cause and effect relationship."
B)"The relationship between these two variables was found to be statistically significant (p = .049; n = 10,000), thus indicating a strong relationship." Note: p reflects the p-value, and n the sample size.
C)"Venting anger can be helpful to your heart health! A study based on 1,000 middle-aged males followed over a period of 10 years found that men who were able to vent their anger were less likely to have heart attacks."
D)All of the above.

Question 26

Explain the ethical dilemma faced by a researcher when he/she is trying to determine what sample size to use in his/her study.

Question 27

__________ is one of the pressures that can mislead a researcher to weaken their conclusions more than is justified.(Note this is unethical, yet it happens sometimes.)

Question 28

Before examining the results of a statistical analysis, make sure that the appropriate __________ have been investigated and verified.If the investigation shows that there is a problem, you have to disregard the results of the analysis.

Question 29

Name two ways in which statistical results can be unfairly reported by researchers or the media.