Deck 5: Designing Experiments

Question 1

The following examples are the independent / predictor variables in proposed studies. Based on these variables, which of the following most likely reflect(s) a type of "true" experimental design? (circle all that apply)

A) 2 (mood induction: happy, sad) x 3 (room temperature manipulation: hot, temperate, cold)
B) 2 (height: tall, short) x 4 (geographic region: West, South, Northeast, Midwest)
C) 3 (personality: introvert, mixed, extrovert) x 2 (adrenaline dosage: 0 cc, 10 cc)
D) 2 (pretest presented: yes, no) x 2 (manipulated seating: sat, stood)

Answer 1

D

Question 2

A posttest-only control group design would be considered a ____(a)______ design. A Solomon four-groups design would be considered a ____(b)______ design.

A) between-subjects; between-subjects
B) within-subjects; within-subjects
C) between-subjects; within-subjects
D) within-subjects; between-subjects

Answer 2

A

Question 3

Many experiments omit the pretest section of an experiment because it risks influencing the effects of the experimental treatments. Which of the following reflect possible outcomes of using a pretest? (circle all that apply)

A) The pretest will sensitize participants to the treatment, artificially enhancing the treatment effect.
B) The pretest will sensitize participants to the treatment, artificially weakening the treatment effect.
C) The pretest will change participants' responses such that their responses on the pretest no longer reflect their opinions or attitudes at the time of the experimental treatments.
D) The pretest will determine whether treatment groups' pretest scores are similar at the outset of a study (before exposure to any treatments).

Answer 3

B,C,D

Question 4

In a 2 (temperature: hot, cold) x 2 (experimenter: male, female) factorial design, researchers investigated whether ambient room temperature and the experimenter's gender influenced participants' aggression when participants were asked to complete a very boring and time-consuming task. The researchers found that generally, cold rooms incited more aggression than hot rooms. Furthermore, they found that room temperature only mattered when the experimenter was male, but not when the experimenter was female. This example reflects what effects of the independent variables?

A) one main effect
B) two main effects
C) an interaction
D) A and C
E) B and C
F) none of the above

Question 5

In a 2 (temperature: hot, cold) x 2 (experimenter: male, female) factorial design, researchers investigated whether ambient room temperature and the experimenter's gender influenced participants' aggression when participants were asked to complete a very boring and time-consuming task. The researchers also recorded participants' gender to see if gender correspondence between participant and experimenter influenced the overall treatment effects. Which of the following statements could be an accurate and appropriate interpretation of their findings? (circle all that apply)

A) The relationship between room temperature and aggression was moderated by the experimenter's gender.
B) The relationship between room temperature and experimenter's gender in predicting aggression was moderated by participants' gender.
C) Hot rooms with male experimenters caused participants' aggression to increase.
D) Hot rooms caused male participants' aggression to increase, but caused female participants' aggression to decrease.

Question 6

Which of the following experimental designs could also be considered a factorial design?

A) pretest-posttest control group design
B) posttest-only control group design
C) Solomon four-group design
D) A and B only
E) all of the above
F) none of the above

Question 7

In an experiment examining school identity induction (high, low), living arrangement (on-campus, off-campus), and wearing school colors, researchers found that when school identity was low, living arrangement did not influence whether students wore school colors. When school identity was high, people who lived off-campus wore school colors more often than those who lived off-campus. What type of interaction effect is described in this example?

A) divergent interaction
B) moderated interaction
C) crossover interaction
D) mixed interaction
E) not enough information

Question 8

An experiment examined the effects of classroom type (walled, open-air) and teaching approach (authoritarian, democratic) on students' learning. The researchers found that students taught by an authoritarian teacher performed the poorest of all experimental groups, and the type of room they were taught in did not matter. When students were taught by a democratic teacher, those in open-air classrooms performed significantly better than those in walled classrooms. What type of interaction effect is described in this example?

A) divergent interaction
B) moderated interaction
C) crossover interaction
D) mixed interaction
E) not enough information

Question 9

With within-subjects designs, particularly when the independent variable has more than two levels, the effects of exposure to one treatment may carry over to the next treatment (and thus influence response to that second treatment). To compensate for this possibility, researchers will often __________.

A) use a pretest to determine whether people starting in different treatments are initially similar
B) use a control group to compare responses to each treatment
C) arrange treatment presentation so that half of participants start with one treatment, while the other half starts on a different treatment.
D) counterbalance the order of treatments

Question 10

Researchers wanted to investigate the efficacy of four different studying strategies, but wanted to avoid possible carryover effects. Consequently, they organized the presentation of treatments like this, and randomly assigned participants across these ordered conditions:

A) complete counterbalancing
B) a factorial design
C) partial counterbalancing
D) a Latin square design

Question 11

It is possible to have an interaction effect with no main effects.

A) True
B) False
C) not enough information

Question 12

Provide an example for a 2 x 2 within-subjects factorial design. In your response, be sure describe what a within-subjects factorial design is, identify the independent and dependent variables involved, and explain how your example exhibits a within-subjects factorial design.

Question 13

Researchers were interested in the efficacy of a new parenting program designed to increase children's consumption of vegetables, wherein none of the strategies in the program have ever been seen or used before, and wanted to compare it to a group of normal parents who try to get their children to eat veggies in whatever way they can (i.e., a control group). However, the researchers also wanted to make sure that at the outset of the study children and parents' parenting styles were effectively similar, but worried that a pretest questionnaire about children eating their veggies might influence parents' emphasis on the family eating a balanced diet. Given this information, what experimental design would you suggest these researchers use and why? To aid in your explanation, provide a visual diagram for what this design might look like.

Question 14

Given two variables, Variable A and Variable M, draw two graphs illustrating the following effects: (a) two main effects with no interaction; (b) one main effect with a crossover interaction. Be sure to label key parts of the figure.

Question 15

Draw a graph representing a Solomon four-groups design with experimental treatment, Treatment B, wherein there were no main effects and a crossover interaction. Be sure to label the key parts of the figure.

Question 16

Though the Solomon four-group design may have many advantages, what might be some disadvantages or logistical challenges to using this design?

Question 17

Is it possible to have an interaction but no main effects? Two main effects but no interaction? Come up with an example to describe each scenario, if possible.

Question 18

Your advisor conducted a study on decision making using a 2 (mood prime: negative, positive) x 2 (physiological arousal: adrenaline injection, saline solution injection) factorial design, and had a final sample of 80 participants. However, upon further investigation, you realized that half of participants, randomly chosen, were administered a questionnaire about their attitudes toward authority figures before the primary dependent measure, and half completed the attitudinal questionnaire after the primary dependent measure. What sort of design would this most likely be? In this example, is it possible to obtain a significant three-way interaction, even if there are no significant main effects or two-way interactions? Do you have sufficient power to obtain a significant three-way interaction? How reliable would you expect the results of a three-way interaction test to be? Is it ever appropriate to divide participants into additional groups or categories after all data have been collected - why or why not?

Question 19

The theory of psychological reactance (Brehm, 1966; Brehm & Brehm, 1981) states that when people feel that their attitudinal or behavioral freedom is threatened or restricted, they will be motivated to restore that freedom. Freedom restoration can be achieved (and has been empirically observed) in several ways, including (a) increased positive evaluation of the object that is restricted, and (b) increased negative evaluation of the entity responsible for the freedom restriction. For example, if the government proposed to eliminate an ingredient (e.g., high-fructose corn syrup) from certain foods, people who perceive that to be a threat to their freedom to choose what they want to eat will be likely to rate high-fructose corn syrup more positively, and rate the government more negatively, relative to people's responses had the government not proposed such a ban. Furthermore, satisfying one of these methods of freedom restoration should serve to restore the freedom, thus making the second method of freedom restoration less likely to occur. In other words, if items assessing DV-1 (e.g., attitudes toward corn syrup) are presented first, responses on DV-2 (attitudes toward the government) are not necessarily going to demonstrate a reactance effect; the same outcome would occur if DV-2 is presented first. However, previous research has not investigated whether such an effect of presentation order on reactance actually occurs, and you want to test it. How might you be able to construct an experimental design that can test whether the presentation of the first dependent measure will dampen participants' responses to the second dependent measure? Is it even possible - why or why not?

Question 20

Using a modified pretest-posttest control group design, you investigated the effects of participants' gender and an experimental intervention on reducing prejudice toward people who are overweight or obese. Could this design also be conceptualized as a Solomon 4-group design - why or why not? Could it be conceptualized as a 2x2x2 factorial design - why or why not? Could it be conceptualized as a 2x2 factorial design - why or why not? If your answer to all of the previous questions is "no," what other label might you ascribe to this design (besides pretest-posttest control group), and why?