Deck 22: Analysis and Interpretation of Complex Research Questions

Question 1

A researcher is doing a study on the effects of TV watching on adolescents' level of aggressive behavior in the classroom. The following could be source(s) of error variance:

A) TV watching
B) Gender
C) Aggressive behavior
D) a and b

Answer 1

Gender

Question 2

In the above example, to control for error variance, the researcher should:

A) Introduce a second independent variable, such as TV watching
B) Introduce a second independent variable, such as aggressive behavior
C) Introduce a second independent variable, such as gender
D) None of the above

Answer 2

Introduce a second independent variable, such as gender

Question 3

A researcher is testing the effects of a new medication on two groups of asthmatics. Participants are randomly assigned to either a treatment group or a control group. The researcher is concerned about participants' ethnicity and adds ethnicity as a second IV. The overall research approach is called:

A) Randomized experimental
B) Quasi-experimental
C) Comparative
D) Associational

Answer 3

Randomized experimental

Question 4

A researcher is testing the effects of a new medication on men's baldness. Two baldness clinics are recruited for the study. One clinic group receives the medication and the other group receives a placebo. The researcher is concerned about the effects of age on outcomes and includes age as a second IV. The overall research approach is:

A) Randomized experimental
B) Quasi-experimental
C) Comparative
D) Associational

Question 5

The appropriate statistic for the above study is:

A) Independent samples t test
B) Factorial ANOVA
C) Repeated-measures ANOVA
D) Factorial ANOVA with repeated measures

Question 6

A researcher wants to know how anxiety affects serum pH levels. He has three groups of participants: low, medium, and high anxiety. He obtains blood samples and measures the serum pH of each participant. He is also interested in the effects of gender on serum pH levels and adds gender as a second IV to his research design. The research approach is:

A) Randomized experimental
B) Quasi-experimental
C) Comparative
D) Associational

Question 7

In a 4 × 2 factorial design, how many interaction effects are there?

A) 1
B) 2
C) 4
D) 6
E) 8

Question 8

In a 4 × 2 factorial design, how many main effects are there?

A) 1
B) 2
C) 4
D) 6
E) 8

Question 9

A 4 × 2 factorial design yields a significant interaction effect. How many individual cell means must be examined?

A) 1
B) 2
C) 6
D) 8
E) 16

Question 10

Which of the following are examples of a mixed factorial design?

A) One between-groups IV and two within-subjects IVs
B) One between-groups IV and one within-subjects IV
C) Two between-groups IVs and one within-subjects IV
D) All of the above

Question 11

A researcher randomly assigned a group of children to either a treatment condition or a control group. All of the children filled out an assessment tool before and after the intervention. This is an example of what research approach:

A) Randomized experimental
B) Quasi-experimental
C) Comparative
D) Associational

Question 12

A researcher had two intact groups of students. The two groups were randomly assigned to either a treatment condition or a control group. Both groups of students complete measures at the pretest and again at the posttest. This is an example of what research approach:

A) Randomized experimental
B) Quasi-experimental
C) Comparative
D) Associational

Question 13

The following pertain(s) to a true mixed design:

A) The between-groups IV is always active
B) The within-subjects IV is always active
C) All the IVs must be active
D) The between-groups IV is always an attribute

Question 14

The following pertain(s) to a pretest-posttest design:

A) The between-groups IV must be active
B) The between-groups IV must be an attribute
C) The within-subjects IV must be active
D) The within-subjects IV is change over time

Question 15

The questions answered by a mixed ANOVA include:

A) Is there a significant main effect for the between-groups IV?
B) Is there a significant main effect for the within-subjects IV?
C) Is there an interaction between the two IVs?
D) All of the above

Question 16

The major difference between the two-factor ANOVA and the two-factor ANOVA with repeated measures on the second factor is that F values are calculated differently. The following is/are true:

A) There are two error terms, one for the between-groups IV and one for the within-subjects IV
B) The between groups independent variable F uses the between groups error term
C) The interaction effect F uses the within-subjects error term
D) All of the above

Question 17

A mixed ANOVA approach is used to analyze a pretest-posttest design. The following is/are true:

A) A significant effect for the between-groups IV yields the most important information
B) A significant effect for the within-subjects IV yields the most important information
C) A significant interaction effect yields the most important information
D) None of the above

Question 18

The following is/are true for the mixed ANOVA approach to analysis of a pretest-posttest design:

A) Meaningful analysis is dependent on a significant interaction between the treatment conditions and time
B) Tests of simple main effects must be done when there is a significant interaction
C) The treatment only can interact with the posttest
D) All of the above

Question 19

The following is/are true for the gain score approach to analysis of a pretest-posttest design:

A) The mixed design is converted to a factorial design with three independent variables
B) The dependent variable is the difference in scores between pretest to posttest
C) A two-factor design becomes a single-factor design with four levels
D) b and c

Question 20

A gain score analysis of a pretest-posttest design uses:

A) A single-factor ANOVA
B) Simple main effects analysis
C) Information about differences prior to the intervention (pretest scores)
D) All of the above

Question 21

When used in the analysis of a pretest-posttest design, the ANCOVA:

A) Changes a mixed design to a single-factor design
B) Uses differences in pretest scores to adjust the posttest scores
C) Analyzes only the posttest scores after adjustments have been made with the pretest scores
D) All of the above

Question 22

A researcher uses a single-factor ANCOVA to analyze a pretest-posttest design. The between-groups IV has three levels. There is a significant F value. The researcher should:

A) Use multiple t tests to interpret the results
B) Use a Tukey HSD post hoc test to interpret the results
C) Use simple main effects analysis to interpret the interaction
D) Graph the pretest and posttest means for comparison

Question 23

A researcher's study is a mixed design with two levels of the within-subjects IV (pretest- posttest). The researcher is using a randomized experimental research approach. The analysis recommended in the text is:

A) The gain score approach
B) The mixed ANOVA approach
C) A single-factor ANCOVA
D) A factorial ANCOVA

Question 24

Different approaches to analyzing pretest-posttest data have their own disadvantages. Which of the following is a disadvantage of using single-factor ANOVA with gain scores?

A) The treatment may interact with the posttest
B) Assumptions of linearity and parallel slopes must be satisfied
C) It does not provide information about pretest differences
D) A significant interaction between the between- and within-subjects IVs needs to be decomposed into main effects before one can interpret the results

Question 25

Examples of complex associational analyses include:

A) MANOVA
B) Multiple regression
C) Multiple, univariate ANOVAs
D) All of the above

Question 26

Which of the following is true of hierarchical multiple regression?

A) The order to enter each variable into the equation is decided ahead of time
B) The researcher can control the effects of a particular variable by entering it first
C) Type I error is reduced by using fewer steps to enter variables in the equation
D) All of the above

Question 27

Problems with the stepwise regression approach include:

A) The computer decides which variables should be included in the equation
B) Only a small number of predictor variables are included in the equation
C) The probability of a Type II error is increased
D) A correlation matrix is not used

Question 28

A researcher is interested in studying the effects of smoking and alcohol use on post-operative complications in the hospital. The criterion variable is categorical with only two levels (complications: Yes or No). Smoking is operationalized as number of packs smoked per week, and alcohol use is measured as number of alcoholic beverages consumed per week. What would be the appropriate complex statistic to use?

A) Multiple regression
B) Discriminant analysis
C) Factor analysis
D) MANOVA

Question 29

A researcher is interested in studying the effects of amphetamine use and exercise on weight loss. The criterion variable is normally distributed. Amphetamine use and amount of exercise are both normally distributed. The research should use which statistic?

A) Multiple regression
B) Discriminant analysis
C) Logistic regression
D) MANOVA

Question 30

Assume that the above researcher (in question 29) added gender, ethnicity (Caucasian vs other), and tobacco use (yes vs. no) as predictors. What type of analysis should be used?

A) Hierarchical multiple regression
B) Discriminant analysis
C) Logistic regression
D) MANOVA

Question 31

A significant "main effect" signifies a causal relationship: That the IV caused the DV.

Question 32

Error variance is variability attributed to differences among individual participants.

Question 33

Error variance can be significantly reduced by adding a second independent variable that might be affecting the DV.

Question 34

The second IV added to a research design to control for error variance is usually an active IV.

Question 35

The addition of a second IV affects the study's overall quantitative research approach.

Question 36

In an ANOVA Source Table, the first thing to look for is significant main effects.

Question 37

In an ANOVA Source Table, the first thing to look for is significant interaction effects.

Question 38

Where there is more than one IV involved, the data should be graphed and examined.

Question 39

Simple main effects analyses should be conducted when there is a significant interaction in an ANOVA Source Table.

Question 40

If a main effect is significant and there are two levels to the IV, a Tukey HSD should be performed to see if the means are significantly different.

Question 41

The simplest mixed design is a two-factor design with repeated measures on the second factor.

Question 42

Mixed-factorial designs only apply to quasi-experimental and randomized experimental approaches.

Question 43

With mixed-factorial designs, the research approach is named after the highest approach.

Question 44

The three research questions answered by a two-factor ANOVA are similar to the questions answered by a mixed ANOVA.

Question 45

A two-factor ANOVA and a mixed ANOVA use the same statistical approaches.

Question 46

Significant interaction effects from a mixed ANOVA when the between-groups IV has more than two levels should be followed up with simple main effects analysis.

Question 47

The mixed ANOVA approach to analyze a pretest-posttest design yields three separate questions.

Question 48

When using a mixed ANOVA approach to analyze a pretest-posttest design, the interaction effect is the most meaningful information.

Question 49

When using a mixed ANOVA approach to analyze a pretest-posttest design, a significant main effect for the within-subjects IV could be due to the internal validity threat of maturation.

Question 50

When using a gain score analysis for a pretest-posttest design, post hoc comparisons are never required.

Question 51

A single-factor ANOVA has less power than a single-factor ANCOVA.

Question 52

ANCOVA is commonly used in situations when a third variable may be responsible for the relationship between the IV and the DV.

Question 53

A major limiting factor to using complex associational analyses is the number of participants required.

Question 54

Regression analyses are used to predict a criterion variable from several predictor variables.

Question 55

With multiple regression, it is possible to have a significant R², even if none of the individual predictor variables are statistically significant.

Question 56

Regression coefficients need to be standardized or converted to z scores for comparisons among them.

Question 57

A correlation matrix is used for almost all multiple regression and multivariate procedures.

Question 58

With discriminant analysis, the dependent variable is continuous with multiple levels.

Question 59

Why is it best practice to interpret the interaction effect first?

Question 60

Why should one be cautious when using gain scores?

Question 61

What does a multiple R tell you?

Question 62

What are the problems with using a stepwise linear regression?