Module 4 Lab

The experimental research strategy

For this assignment use the topic you intend to write your paper on.

Design a between-groups experiment with two groups and one dependent variable.

The questions are listed below, and here   Module 4 lab.docx Download Module 4 lab.docx

There is also a sample below for parts A & B and a few comments on part C

_______________

A. Your task is to begin to design a between-subjects experiment related to the topic you are interested in writing your paper about.

1. In a few sentences, describe your population of interest, research question, and design.

2. What is the hypothesis you are testing?

3. Who will your participants be, that is who will be in each of the two groups?

4. What is your independent variable? Be specific!

5. What is your dependent variable? Again, be specific.

6. What is one variable, either an individual difference or other variable, that might be a confounding variable? Why is it a potential confound?

7. How could you deal with this potential confound? Should you hold the variable constant, or restrict the range or match participants, or use a different strategy (and name that strategy) to minimize the potential confound?

8. Briefly describe your sampling procedure including what type of sampling procedure it is.

9. What outcome do you expect given your hypothesis?

10. What statistical test should you use to analyze your data?

B. Now redesign your study so it is similar, except now uses a with-in subjects design.

1. In a few sentences, describe your research question and design.

2. What is the hypothesis you are testing?

3. Who will your participants be?

4. What is your independent variable? Be specific!

5. What is your dependent variable? Again, be specific.

6. What outcome do you expect given your hypothesis?

7. What statistical test should you use to analyze your data?

Sample Lab:

My population of interest is people with delusional disorders. I want to see if acceptance and commitment therapy (ACT) can help people with delusional disorders cope better with their unusual beliefs. I will use a between-groups design where one group receives Acceptance and Commitment Therapy and a second group receives a widely accepted form of cognitive therapy for delusional beliefs.

The hypothesis is that people who receive ACT will have better coping than people who receive Cognitive Therapy as evidenced by lower scores on the Peter’s Delusional Inventory (PDI), a questionnaire that asks about degree of belief, frequency, and distress associated with several common delusional beliefs.

The participants in each group will be people who have been diagnosed with a delusional disorder.

The independent variable is:

Exposure to eight sessions of acceptance and commitment therapy Vs. eight sessions of cognitive therapy for delusions

The dependent variable is score on The Peter’s Delusional Inventory

Type of delusional belief could be a confound. People with paranoid delusions have a harder time learning to trust a therapist. Age could also be a confound, as older people may have longer-standing delusions that are harder to treat. I would use matching participants with respect to type of delusions so the groups have the same number of participants with each type of delusion. I think randomization would take care of concern about age.

My accessible population is people diagnosed with a delusion living within 20 miles of UCF. I would recruit them from two large community mental health centers in Orlando. It will be a convenience sampling in that people have to want treatment, so not every person with a delusional disorder will be a potential participant.  People who agree to participate will be randomly assigned to each treatment group with restricted randomization so each group is matched with respect to type of delusional belief.

The expected outcome is that those who receive ACT will have lower scores on the PDI following treatment than will those receiving cognitive therapy.

An independent measures t-test will be used to determine if the means on the PDI differ following treatment.

To make this a within-subject design I would have one group of participants who would receive one treatment and then the other. I would randomize them with respect to which treatment is delivered first. They would complete the PDI after each course of treatment.

My hypothesis is that there will be a larger change in PDI score after ACT than after cognitive therapy.

The independent variable is type of treatment and the dependent variable is change in PDI score following treatment.

The participants are individuals with a diagnosis of delusional disorder.

I would expect that scores on the PDI will drop more following ACT than following cognitive therapy.

I would use a repeated measures t-test to analyze the results.

In this sample, I am not going to tell you which design I like better.

I will tell you that in order to decide I would weigh the pros and cons – such as limiting individual differences (advantage within), the possibility of carryover effects from having both levels of the independent variable (advantage between), the difficulty of recruiting people with delusional disorders (advantage within), and so on…and look at how big an advantage/disadvantage each creates.