Comparing Research Approaches (Coursework Sample)

Unit 1: Graded Assignment: Comparing Research Approaches
Comparing Research Approaches: The purposes of this assignment are to: (1) Learn about the differences between correlational, experimental, and non-experimental methods and (2) Understand what conclusions can appropriately be drawn from each type of research approach.
1. Read and comprehend the “Correlational, Experimental, and Non-Experimental Research Methods” handout.
Correlational, Experimental, and Non-Experimental Research Methods Handout
Correlational Methods
Goal: To determine how two, typically numerical and continuous, variables are related to each other. Correlational studies cannot identify causal relationships.
Procedure:
 Two variables are simply measured as they already exist in the world, with no manipulation by the experimenter.
 Because there is no manipulation of an independent variable by the experimenter, there likely are confounds, often called third variables in correlational studies. Third variables are unmeasured variables that relate to both measured variables, thus causing the relationship between the measured variables. Third variables are one reason why correlational studies do not allow drawing of causal conclusions.
 Another reason that correlational studies do not allow drawing of causal conclusions is because the direction of causality is not clear. That is, does measured variable A cause measured variable B or does measured variable B cause measured variable A?
 Relationships between variables in correlational studies are often expressed as correlation coefficients. Correlation coefficients range between –1 and +1. Negative correlations indicate inverse relationships, i.e., as the value of one variable increases the value of the other variable decreases. Positive correlations indicate direct relationships, i.e., as the value of one variable increases the value of the other variable increases.
Experimental Research Methods
Goal: To determine what causes a particular behavior.
Procedure:
 A potential cause (independent variable) is manipulated by the experimenter to determine its effect on a behavior (dependent variable). All other variables are controlled so they don’t become confounds (differences between the groups besides the independent variable that can provide alternative explanations for experimental results and thus threaten causal conclusions).
 Manipulation: Manipulation of the cause is often accomplished by using an experimental group and a control group. The experimental group gets the treatment and the control group does not. The control group allows us to obtain baseline data on the dependent variable (level of the behavior without the treatment) as well as to test for placebo effects (effects due to the belief that a treatment will work, rather than due to actual treatment effects)
 Control: Preventing variables from becoming confounds through control can be accomplished in three ways. (1) Usually, subjects are randomly assigned to their groups. Random assignment involves determining who goes in to what group purely on the basis of chance (e.g., a coin flip). Random assignment tends to even out participant variables. (2) Subjects can be matched according to important participant variables to even out those variables. (3) Important participant characteristics can be held constant, e.g., using only females.
Non-Experimental Research Methods
Goal: To examine relationships between variables by comparing groups of scores.
Procedure:
 In a nonequivalent groups design, people are assigned to groups on the basis of pre-existing participant characteristics (e.g., those with and without diabetes), rather than using random assignment. The groups may be non-equivalent because, besides differing on the variable used to create the groups (the quasi-independent variable), they may differ in other systematic ways, creating confounds. This often is considered to be a correlational method because there is no random assignment; it doesn’t necessarily look like the correlational method because there are groups.
 In a pretest-posttest design, one group of people is measured before and after a treatment. The researcher does not have control over what happens during the passage of time from before to after treatment except for the treatment. If the groups have systematic differential experiences, confounds are again created.
2. Outline the major differences between the three approaches by placing “yes” or “no” in each cell of the following table:
3. Test your knowledge of whether you can determine the research design used and the considerations in whether you can draw causal conclusions by completing the following worksheet:
Label the following according to what type of design it is. Your choices are below, a-d. Answer any other questions associated with each item.
a. Correlational Study
b. Experiment
c. Non-Experiment: Pretest-Posttest Design
d. Non-Experiment: Non-equivalent Groups design
a. Schacter and Singer randomly assigned each participant to one of two conditions: informed and uninformed. All participants received epinephrine, a drug that causes arousal, but those in the informed condition were told about the effects of the drug, whereas those in the uninformed condition were not. Those in the informed condition reacted with less anger than those in the uninformed condition to an insulting questionnaire.
What type of design is this? Why do you say this?________________________________
What is the independent variable?______________________________
What is the dependent variable?________________________________
b. Princeton and Dartmouth students were asked to view a match between their schools’ football teams and to determine which team broke more rules of the game. It was found that students at Princeton thought that the Dartmouth team broke more rules and
Dartmouth students thought that the Princeton team broke more rules.
What type of design is this? Why?_____________________
What is the “independent variable”?___________________________
What is the “dependent variable”?____________________________
c. Researchers examined whether there is a relationship between number of benefits obtained in a relationship (measured by the number of checkmarks on a checklist) and ratings of happiness in the relationship (1 = very unhappy, 7 = very happy).
What type of design is this? Why?_____________________
Consider interpretations with the ideas of direction of causality and third-variable explanations in mind._________________________________________________
d. In order to determine whether type of persuasion technique impacts behavior, researchers presented some participants with a scary lung cancer movie, some with pamphlets with advice on how to quit smoking, and some with both the movie and pamphlets. Participants were randomly assigned to the conditions. Participants were asked to keep a diary for the next three months of how many cigarettes they smoked.
What type of design is this? Why?_________________________________
What is the independent variable?__________________________
What is the dependent variable?_________________________
e. The scores on an anorexia scale of female members of a sorority were compared to anorexia scores of female members of the hockey team.
What type of design is this? Why?__________________________
What is the “independent variable”?________________________________
What is the “dependent variable”?____________________________________
Say that the average anorexia score for hockey team members was lower than for sorority members. Does the type of club cause differences in anorexia scores? What might be alternative explanations?_________________________________________________
f. Participants were asked to look at a set of comics and rate how funny they were. These same participants were then asked to hold a pencil in their teeth and look at another set of comics and rate of funny they were.
What type of design is this? Why?________________________________
Say that the set of comics rated while they held pencil in their teeth were rated as funnier. Was it because they were holding pencil in their teeth, do you think? What might be alternative explanations?___________________________________
g. Researchers wanted to examine the relationship between attractiveness and number of social interactions in a one-week period. Attractiveness of each person was rated by two undergraduate raters, one male and one female, based on a picture submitted by the research participants. Ratings were on a scale from 1= very unattractive to 7 = very attractive. Number of social interactions was evaluated by participants’ pushing a counter attached to their clothing at the beginning of each social interaction over the next week.
What type of design is this? Why?__________________________________
What are the variables in the study?_____________________________
Say that the two variables are found to be related. Explain the relationship. Consider direction of causality and third variables in your explanation.
______________________________
h. It has been found that people who go to church rate themselves as happier than those who do not go to church.
What type of design is this? Why?___________________________________
Explain the relationship. Consider direction of causality and third variables in your explanation.___________________________