Research Methods in Psychology, Spring, 2003     S. J. Gilbert, SUNY-Oneonta

STUDY QUESTIONS FOR TEST #2

The first box contains questions that are carried forward from Test-1. Some of them have been REWORDED or EXPANDED, so be careful. 

1.4 What are the 3 COMPONENTS OF AN EXPERIMENT? Why is each crucial to establishing CAUSE-AND-EFFECT RELATIONS between variables?

4.9 Distinguish between RELIABILITY and VALIDITY. How do you determine whether an inventory designed to measure a person’s level of SUSCEPTIBILITY TO MOTION SICKNESS is RELIABLE? How do you determine whether the inventory is VALID? [Note: It is NOT sufficient to say "choose questions that are about motion sickness."

5.2 Give an OPERATIONAL DEFINITION of the variable "admiration for Steve Gilbert."

5.5 The following experiment was run over the course of a one week period, during which some subjects went home for the weekend, whereas others stayed on campus. Each subject sat in a classroom and watched a videotaped presentation by a prospective teacher, in which the teacher either spilled a cup of coffee on himself, or didn’t. The "spill" group watched the video in a classroom, over a standard 21" video monitor. The "no spill" group watched the video across the hall, in a similar classroom, over a recently installed 52" projection screen monitor. After watching the video, all of the students filled out a questionnaire that evaluated the teacher. The mean scores of the students in the spill vs no-spill conditions were compared to determine whether people evaluate a person who has committed some kind of blunder more negatively than a person who has not committed a blunder.

(A) What is the DEPENDENT VARIABLE? (B) What is the INDEPENDENT VARIABLE? How many levels of this variable are there? (C) Identify an EXTRANEOUS VARIABLE that is CONFOUNDED with the INDEPENDENT VARIABLE. (D) Identify a NUISANCE VARIABLE. (E) State a DIRECTIONAL HYPOTHESIS. (F) Is this an INDEPENDENT (Between) GROUPS design, or a CORRELATED GROUPS (Within-Subjects) design? (G) What is the appropriate inferential statistical test for testing the hypothesis (give its FULL NAME)? (H) Make up a result of this statistical test that would SUPPORT your hypothesis, and then report it as you would in the results section of a research paper. (I) Make up a result of this statistical test that would NOT SUPPORT your hypothesis, and then report it as you would in the results section of a research paper.

Here are the new questions.

Chapter 13.

13.1 What are the major components of an APA format paper, and in what order do they appear?

13.2 What information is contained in an ABSTRACT?

13.3 What information is contained in an INTRODUCTION?

13.4 What are the principal subsections of a METHOD section?

13.5 What information is contained in a RESULTS section, and how does it differ from a DISCUSSION section?

13.6 What is the difference between a TABLE and a FIGURE? Under what circumstances would one be better than the other? Under what circumstances would you include NEITHER in your research report?

13.7 What is the difference between a REFERENCE section and a BIBLIOGRAPHY? Which is included in an APA format paper?

13.8 Each of the following sentences violates a grammatical guideline of the APA publication manual. Rewrite each sentence to eliminate the problem.

> After filling out the questionnaire, the participants were told to raise their hands.

> The questions which were not answered correctly were removed from the inventory.

> Since the child could not follow the instructions, he could not perform the task.

> While the majority of participants completed the task, many of them could not.

 

Chapter 7a.

7.1 What is the preferred measure of CENTRAL TENDENCY? Under what circumstances would an alternative measure of central tendency be more accurate? What is that alternative measure?

7.2 Hollie did an experiment in which subjects heard music at either a low, moderate, or high volume while they studied for a french test. Half of the subjects in each group were males and half were females. Here are the mean scores the subjects received on the test: high-volume males: 82.5; moderate-volume males: 75; low volume males: 64.5; high-volume females: 80; moderate-volume females: 81; low volume females: 79.

Draw a properly proportioned and labeled LINE GRAPH depicting the results of this experiment.

7.3 Make up a MEAN and a STANDARD DEVIATION for the variable "weights of 100 randomly selected female SUNY-Oneonta sophomores. Explain what information we can derive from the STANDARD DEVIATION statistic that you made up.

7.4 Leo gave a VOCABULARY TEST and a SELF-CONFIDENCE inventory to 137 PSYC 100 students. Then he calculated a PEARSON CORRELATION coefficient. Draw a SCATTERGRAM (with properly labeled x and y axes) depicting each of the following results: (a) r = -85; (b) r = +.50; (c) r = +.03. Which of these correlations shows the STRONGEST relationship between the two variables?

 

Chapter 7b.

7.5 Harriet did the BLUE vs WHITE paper experiment (using an independent groups design, with 25 subjects in each group) and got the following results: BLUE: Mean=81.3, SD = 5.6; WHITE: Mean = 75.2, SD = 4.8. The means for the two groups look pretty different (DIFFERENCE = 6.1). (A) There are TWO KINDS OF  EXPLANATIONS for why the mean for the subjects in one group of an experiment would not be the same as the mean for the subjects in the other group (in Harriet's case, why the BLUE and WHITE paper means are not the same). What are they? (B) What should Harriet do now to help her decide which of the two kinds of explanations to accept?

7.6 (A) Give an example of a TYPE-I error, and explain why it is. (B) Give an example of a TYPE-II error, and explain why it is.

7.7 What information is expressed in the NUMERATOR and DENOMINATOR of the equation for inferential tests such as the t-test and the F-test (ANALYSIS OF VARIANCE)? Suppose the independent variable SIZE OF CLASSROOM (30 vs 60 desks) has NO ACTUAL EFFECT on the dependent variable SATISFACTION WITH CLASS (measured on a 1-10 scale). I do an experiment in which students are RANDOMLY ASSIGNED to one of the classrooms, and then rate their satisfaction with the class. (A) Suppose the mean satisfaction rating of the students assigned to the smaller room is 7.0. What mean satisfaction rating would you expect for the students assigned to the larger room? (B) Suppose you did this experiment over and over, and did a t-test each time. Then you calculated the MEAN of all these t-tests. What mean would you expect to get? Why?

7.8 What is the meaning of p < .05. (A) What decision do we make about our hypothesis and our null hypothesis if a t-test produces a significance level of p < .13? (B) What decision do we make about our hypothesis and our null hypothesis if a t-test produces a significance level of p < .03?

 

 

Chapter 8a.

8.1 Madonna decided to do a simple experiment using a Two Group design, to test the hypothesis that a person who talks FASTER than average, will be viewed as more intelligent than a person who talks SLOWER than average. She hasn’t made a final decision concerning whether you will use a TWO-INDEPENDENT-GROUPS (Between-Groups) design or a TWO -CORRELATED-GROUPS (Within-subject or Repeated-Measures) design. (A) Draw a figure that clearly shows how participants would be assigned to the experimental conditions in a TWO -INDEPENDENT-GROUPS design. (B) Draw a figure that clearly shows how participants would be assigned to the experimental conditions in a TWO-CORRELATED-GROUPS design. (C) Suppose Madonna decided to use an INDEPENDENT-GROUPS design. Is there a CONTROL GROUP in her design? If so, what is it? If not, why not?

8.2 Astonishingly, after spending a semester in Steve Gilbert’s research methods class, Alfredo becomes interested in the effect the COLOR OF PAPER (blue vs white) has on PERFORMANCE on a TEST. Alfredo decides to do an experiment, using a simple TWO -INDEPENDENT-GROUPS (Between-Groups) design. He has 120 participants available to participate in his experiment. He wants to assure that the participants who end up in each group are, on average, equal with respect to variables such as intelligence, health, experience, and any other variables that might affect how they perform on the DEPENDENT VARIABLE of the experiment. (A) In light of the number of participants available to him, what method for assigning participants to groups should Alfredo use? (B) What is the rationale for using this method?

8.3 It turns out that only 24 participants were available for Alfredo’s experiment (see 8.2 above), and Alfredo realized that he would have to use one of THREE special techniques for assigning participants to groups, if he was to assure that the participants who end up in each group are, on average, equal with respect to the variable INTELLIGENCE (which would have an important effect on how they perform on the DEPENDENT VARIABLE of the experiment). The text describes three special techniques for assigning participants to groups that can be used in such a case. Name and describe each one, and give the rationale for how it achieves equality between groups of participants, on a potential confounding variable (such as intelligence).

 

Chapter 8b.

ADVANTAGES

DISADVANTAGES

INDEPENDENT GROUPS DESIGN

CORRELATED GROUPS DESIGN

8.4 Any researcher who wishes to do a simple TWO-GROUP experiment, must decide whether to use an INDEPENDENT-GROUPS or a CORRELATED-GROUPS design. Each has advantages and disadvantages. Draw a table like the one shown above, and fill it in, showing the advantages and disadvantages of each design.

8.5 (A) What inferential statistical test is appropriate for analyzing the results of a TWO-GROUP experiment that used an INDEPENDENT GROUPS design?

Imagine that Carrie did such an experiment, using COKE vs PEPSI as the levels of the INDEPENDENT VARIABLE, and a 1 (HATE IT) to 10 (LOVE IT) scale for the DEPENDENT VARIABLE. Her DIRECTIONAL hypothesis was that subjects would prefer COKE over PEPSI. She had 20 participants in each group. Make up MEANS and STANDARD DEVIATIONS for each group, reflecting a set of results Carrie reasonably could expect to get in her experiment. Then, make up the results of a t-test performed on these results.

(B) Write a short, succinct RESULTS SECTION of a research paper for Carrie’s experiment, using the information you made up. Make it as succinct as possible – no unnecessary words!

8.6 (A) What inferential statistical test is appropriate for analyzing the results of a TWO-GROUP experiment that used a CORRELATED GROUPS design?

Imagine that Carl did such an experiment, using KENTUCKY FRIED CHICKEN vs BROOKS CHICKEN as the levels of the INDEPENDENT VARIABLE, and a 1 (HATE IT) to 10 (LOVE IT) scale for the DEPENDENT VARIABLE. His DIRECTIONAL hypothesis was that subjects would prefer BROOKS CHICKEN over KENTUCKY FRIED CHICKEN. He had 10 participants available for his experiment. Make up MEANS and STANDARD DEVIATIONS for each LEVEL of Carl’s INDEPENDENT VARIABLE, reflecting a set of results Carl reasonably could expect to get in his experiment. Then, make up the results of a t-test performed on these results.

(B) Write a short, succinct RESULTS SECTION of a research paper for Carl’s experiment, using the information you made up. Make it as succinct as possible – no unnecessary words!

 

Chapter 9.

9.1 Make up and describe an experiment that has: (A) One independent variable with three levels; (B) One independent variable with four levels; (C) Two independent variables, each with two levels.

9.2 Lenny wishes to do an experiment that looks at how the difficulty of a problem affects people’s feelings of depression. Her Independent Variables is DIFFICULTY OF ANAGRAM PROBLEM, with the three levels being LOW, MODERATE, and HIGH Difficulty. Lenny’s DEPENDENT VARIABLE measure is TEMPORARY FEELINGS OF DEPRESSION as measured by a standard Depression Inventory.

Lenny wishes to use an INDEPENDENT GROUPS design, but only has 27 participants available, and isn’t confident that simple RANDOM ASSIGNMENT of nine participants to her three groups would be sufficient to assure that each group started the experiment in the same MOOD. She decides, instead, to use PRIOR MOOD (as measured by a mood inventory) as a MATCHING VARIABLE. (A) Describe the exact procedure that Lenny should use in assigning participants to the three groups, using the MATCHED SETS procedure. (B) Why is this procedure superior to simple RANDOM ASSIGNMENT?

9.3 (A) What statistical test would I use (give its exact and complete name) to analyze the results of an experiment with more than two levels of the INDEPENDENT VARIABLE, in which:(A) each participant is assigned to RECEIVE ONLY ONE LEVEL OF THE INDEPENDENT VARIABLE, via simple RANDOM ASSIGNMENT; (B) each participant receives EVERY LEVEL of the INDEPENDENT VARIABLE; (C) each participant within a MATCHED SET (on a matching variable) is assigned to RECEIVE ONLY ONE LEVEL OF THE INDEPENDENT VARIABLE, via simple random assignment.

9.4 Anne has 80 participants available for an experiment she wishes to do on the effect of ambient noise (ZERO vs LOW vs MODERATE vs HIGH) on PHYSIOLOGICAL AROUSAL. (A) What is the DEPENDENT VARIABLE? (B) What is the INDEPENDENT VARIABLE? (C) How many levels of the INDEPENDENT VARIABLE are there? (D) How many levels of the INDEPENDENT VARIABLE would each subject participate in, if Anne used a REPEATED MEASURES design? (E) How many levels of the INDEPENDENT VARIABLE would each participant experience if Anne used an INDEPENDENT GROUPS design?

9.5 Both the t-test and the ANOVA (Analysis of Variance) share the same, basic logic. They are based on an equation in which a term in the NUMERATOR is divided by a term in the DENOMINATOR to produce a statistic (either t or F). (A) Describe the meaning of the term in the DENOMINATOR; (B) Describe the meaning of the term in the NUMERATOR; (C) Suppose that the NULL HYPOTHESIS really is true (i.e., that the INDEPENDENT VARIABLE in fact, has no effect on the DEPENDENT VARIABLE). What t would you expect when you do a t-test? What F would you expect when you do an ANALYSIS of VARIANCE? Why? (D) Suppose that the HYPOTHESIS really is true (i.e., that the INDEPENDENT VARIABLE, in fact, has a powerful effect on the DEPENDENT VARIABLE). What t might you expect when you do a t-test? What F might you expect when you do an ANALYSIS of VARIANCE? Why?

9.6 According to the text, there is a two stage process involved when an (ANOVA) ANALYSIS of VARIANCE is used to analyze the results of an experiment in which there is a single INDEPENDENT VARIABLE with MORE THAN two levels.

The FIRST STAGE is to perform an ANALYSIS of VARIANCE, to determine whether the MEANS for the levels of the INDEPENDENT VARIABLE differ from each other to a greater extent than would be expected by chance – i.e., if the INDEPENDENT VARIABLE makes a difference. Suppose that the ANOVA produces a result that is statistically significant, at p < .05.

(A) Describe the SECOND STAGE of this two stage process – exactly what you do, and how you do it. (B) Why is this second stage necessary? What information does it provide that cannot be derived from the ANOVA itself? (C) Give two hypothetical outcomes of this second stage process, from an experiment in which participants’ liking for COORS, MILLER, and SAM ADAMS beer is compared.