AP Psych Free Response Question for Problems in Research

These questions are modeled after Free Response Questions (FRQs) of the AP Psychology Exam. Below would most likely represent one of two questions that could appear on the FRQ.

The instructions on the most recent FRQs have included the following: It is not enough to answer a question by merely listing facts. You should present a cogent argument based on your critical analysis of the questions posed, using appropriate psychological terminology.

Use the items below to assess your mastery of knowledge of the topics below, but also how you might apply your knowledge to present your “cogent argument.”

Question

Psychologists who conduct research studies encounter difficulties that can threaten the validity or ethical aspects of their work. Describe how each of the following potential problems in research can best be prevented.

1. Poor External Validity

2. Coercion

3. Framing Effect

4. Correlation-Causation Fallacy

5. Placebo Effect

Sample Answers and Feedback

FRQ Practice Submission 1

(a) Poor external validity is having a study that can generalized outside the context of the study. To avoid low external validity, researchers must use random selection so everyone has an equal chance of being in the study and it’s a representative sample.

(b) Coercion means getting a person to do a study through force. This can be avoided as ethics in psychological studies require no coercion to be used in studies and that it must be voluntary, and studies with coercion cannot be approved by the IRB.

(c) Framing effect happens when options or questions have a negative or positive connotation to them and create bias. This can be avoided by researchers making sure questions or options present no positive or negative effect to them. For instance instead of saying “Since smoking kills people everyday, how do you feel about smoking” change it to “How do you feel about smoking?”

(d) Correlation-Causation Fallacy is the tendency to think that correlation causes causation, but actually correlation doesn’t imply causation. This can be avoided by not using any spurious correlations in the study such as connecting ice cream process to forest fires increasing.

(e) The Placebo effect happens when participants believe they have had or show benefits due to the placebo they took. To minimize this effect, researchers can use double-blind studies, this will make sure researchers don’t give any hints to the participants on whether they have the placebo or the actual pill.

Teacher FRQ Feedback

Part (a) is correct. Random selection can help maintain external validity. For part in (b), stressing that a study is voluntary reduces the threat of perceived coercion. For part (c), the example of rephrasing the question is an effective way of ensuring that the point will be scored. Part (d) is a tough question. Spurious correlations are likely to be found if you search for them in data. How can we avoid making that mistake? Part (e) would not score. Consider what the placebo effect is. Is it avoidable?

FRQ Practice Submission 2

(a) Poor External Validity means the research cannot be replicated in other situations. In order to prevent this, you can have fellow researchers skim through your experiment plan, before you start, and determine whether or not they would be able to replicate it.

(b) Coercion means forcing people to participate in your research study. This can be prevented if you get word out for your research early and have your volunteers ready. Also, it is unethical to force people to participate, so the study would not be approved.

(c) The framing effect is a principle that influences people’s decisions based on how the options were presented (tone, diction, etc). To prevent this, make sure your voice is neutral while presenting the various options, so that bias is not formed.

(d) The Correlation-Causation Fallacy emphasizes that the IV may not be the sole reason of the change in the DV. To prevent this, make sure you account for the other variables like confounding, external, etc before you start, so that you can figure out a way to prevent them from influencing your study,

(e) The placebo effect states that the expectations and biases of the participants could influence their behavior in the study. To prevent this, researchers should use a double blind study, so that neither the psychologist nor the participants are influenced by anything. Also, researchers can tell the participants as minimal as possible, so that they don’t know what is supposed to happen to them if they do not receive the placebo.

Teacher FRQ Feedback

Part (a) would not score. External validity and procedures are not too closely related. For part (b), how would having them ready early reduce coercion? For part (c), tone of voice is possible, but I would also stress wording as well. In part (d), the definition is incorrect but the application will likely score. For part (e), double blind by itself would not address the placebo effect. To what are you comparing placebo?

FRQ Practice Submission 3

(a) Poor External Validity is the inability of a research study to generally apply to the targeted population. This potential problem is best prevented by using random selection and random assignment in the conducted research study.

(b) Coercion is when a participant in a study is forced into participating using threats or other forms of pressure. This can be avoided by having a random selection of participants, and if participants choose not to volunteer, others who are randomly selected who do choose to participate will help keep the study ethical and maintain its validity.

(c) Framing Effect is the usage of diction to manipulate a positive or negative correlation to something–resulting in a bias in the research study. This is best prevented by ensuring that there is no preset bias in the experiment by looking at the way questions and phrases are worded.

(d) The Correlation-Causation Fallacy is the false belief that two variables that are correlated were caused by one or the other. This can be prevented by ensuring the claim in a research study frames only the information that was found, and does not assume any information that was not found or backed up by the study.

(e) The Placebo Effect is a false belief that a participant taking a placebo has felt a difference. This results in bias from researchers as they are already aware that they are taking the placebo, and is best prevented by having a double-blind study take place.

Teacher FRQ Response

For part (a), random selection scores the point, but not random assignment. For part (b), random selection would not help avoid the threat of coercion. Part time (c) would score. Part time (d) is lacking some detail that would score the point. Part (e) needs more detail to score the point; placebo compared to what?

FRQ Practice Submission 4

(a) Poor external validity is when outside factors are heavily influencing the experiment. This can best be prevented by random selection, which will allow participants to be representatives of their population without the worry of other factors. Coercion is when participants are forced into the study.

(b) Coercion can best be prevented by giving individuals the option to participate in an experiment/survey (without the use of a reward if they decide to go ahead with participating).

(c) The Framing Effect is when a participant reacts to something the experimenter says based on how it’s presented in front of them. This can be prevented by giving just the amount of information a participant needs without going too far in depth in order to prevent unnecessary skewing of results.

(d) The Correlation-Causation Fallacy is when two or more things are believed to be correlated to each other, but are not. This can be prevented by making some sort of scatter plot after conducting an experiment showing how one variable may influence the likes of another.

(e) The placebo effect is when a drug doesn’t have any effect on the participant/test subject. This can be prevented by conducting an experiment and administering the drug without telling the participants there is a drug at all (single-blind studies).

Teacher FRQ Feedback

In part (a), the definition is not correct but the application scores the point. Part (b) would score the point. Part (c) likely scores the point because the definition gives context. Part (d) likely would not score: a scatterplot tells you whether there is a correlation. I don’t think part (e) would score. What you describe would still result in a placebo effect.