Methods Activity

Methods Activity
Social Psychology (PSYC 3311)

Part 1. Correlational vs. causation. Recallthat correlational research differs from true experimental designs in a number of important ways. There is 1) no manipulation of an independent variable, 2) random assignment to groups does not occur, 3) there is no control of extraneous (confounding) factors. Thus, we cannot make causal inferences from correlational studies.

Let’s practice the skill of distinguishing between correlational studies that provide descriptions or predictions about two or more variables (relational evidence), and experimental studies that provide explanations or causal evidence. To do so read the headline below and make a few decisions.

Keeping a diary doubles diet weight loss.

•    Without reading the associated article, does the headline seem to make a correlational or causal claim?         Correlational        Causal

•    If you think the headline makes a causal claim, what was the IV (independent variable) and what was the DV (dependent variable)?

•    If the headline makes a correlational claim, name possible third variables (a variable that might be related to both variables).

Part 1.1. Now that you have answered the questions above, read the article associated with the headline.

Keeping a food diary can double a person’s weight loss according to a study from Kaiser Permanente’s Center for Health Research. The findings, from one of the largest and longest running weight loss maintenance trials ever conducted, will be published in the August issue of the American Journal of Preventive Medicine.

Funded by the National Heart, Lung and Blood Institute at the National Institutes of Health, the study is one of the few studies to recruit a large percentage of African Americans as study participants (44 percent). African Americans have a higher risk of conditions that are aggravated by being overweight, including diabetes and heart disease. In this study, the majority of African American participants lost at least nine pounds of weight, which is higher than in previous studies.

“The more food records people kept, the more weight they lost,” said lead author Jack Hollis Ph.D., a researcher at Kaiser Permanente’s Center for Health Research in Portland, Ore. “Those who kept daily food records lost twice as much weight as those who kept no records. It seems that the simple act of writing down what you eat encourages people to consume fewer calories.”

In addition to keeping food diaries and turning them in at weekly support group meetings, participants were asked to follow a heart-healthy DASH (a Dietary Approaches to Stop Hypertension) diet rich in fruits and vegetables and low-fat or non-fat dairy, attend weekly group sessions and exercise at moderate intensity levels for at least 30 minutes a day. After six months, the average weight loss among the nearly 1,700 participants was approximately 13 pounds. More than two-thirds of the participants (69 percent) lost at least nine pounds, enough to reduce their health risks and qualify for the second phase of the study, which lasted 30 months and tested strategies for maintaining the weight loss.

Does the research support the claim? Why or why not? Defend your answer.

Part 2. Internal Validity. An experiment is described below. Before reading it recall the necessary ingredients of a true experiment and the meaning of internal validity. Refer to your notes and/or the textbook.

A true experiment includes_________________________________________


Internal validity_____________________________________________________________________

Part 2.1.
1. A researcher was interested in the effects of alcohol on perceptions of physical attractiveness of the opposite sex in a heterosexual sample. To study this, students from two of the professor’s classes were asked to participate, a senior seminar for psychology majors which met Thursday evening once a week from 6-9pm, and a freshman introductory psychology class, which met two mornings a week at 10 am. Because the seniors were all at least 21 and thus legally able to drink, they were assigned to the condition that received 2 oz. of alcohol mixed in with 6 oz. of orange juice. The freshman were assigned to the “placebo” alcohol condition, in which they received 2 oz. of tonic water (which tastes like alcohol) mixed in 6 oz. of orange juice. However, everyone is told they are really being served alcohol.

Students were invited to participate in the study if they had a free hour after their class with the professor. The professor conducted the study on a Thursday, on a day when the introductory class had had an exam. Students drank either the “alcohol” or the placebo drink, waited 30 minutes in a lounge for the alcohol to take effect, and then sat at a computer and performed a five-minute task in which they rated various faces of the opposite sex on physical attractiveness.

The group that had received alcohol rated the faces as more attractive than the group that did not receive alcohol and the professor concluded that alcohol makes people of the opposite sex appear more attractive.

A. What is the independent variable in the experiment? ______________________________________

B. What is the dependent variable in the experiment? _______________________________________

C. Is the professor’s conclusion a reasonable one? To answer this question, brainstorm any potential threats to internal validity.

Part 3.1. Operational definitions. Operational definitions are the specific criteria you use in a study to decide what “counts” as an example of a variable. For example, if you wanted to study obesity, you would need a specific criterion to decide who was obese and who wasn’t. Create an operational definition for each of the following concepts.



Risky Behavior

Part 3.2. Most experiments use more than one operational definition of a concept because the experiment may be more likely to find the desired results if variables are manipulated or measured in more than one way. Practice this skill below by writing at least 3 different operational definitions for the following terms.

Helping behavior

Self control

Part 4. Design an experiment. A hypothesis is listed below. Design an experiment to test the research question. Your experiment should follow established ethical guidelines. Imagine that you have unlimited funding.

2. Does alcohol consumption influence perceptions of attractiveness?

Describe your sample.

Operationally define the independent variable(s) and describe how you will manipulate it/them:

Operationally define the dependent variable(s) and describe how you will measure it/them including sample survey items if applicable:

Often, there are variables (called confounding variables) other than the independent variable that can systematically affect the dependent variable. Confounding variables interfere with an experiment because they can cause changes in the dependent variable, creating the illusion that the independent variable affects the dependent variable when it really doesn’t. Similarly, confounding variables can create the illusion that the independent variable doesn’t affect the dependent variable when it really does. Either way, confounding variables interfere with an experiment’s ability to draw correct conclusions about the relation between the independent and dependent variables. Which variables (if any) in this experiment might affect the dependent variable(s) other than the independent variable(s)?

Identify any confounding variables in your experiment.