Deck 13: Advanced Statistics

Question 1

Matching subjects on a potentially confounding variable is one way to remove that variable's influence on the final results of a study. What is another technique to accomplish this?

A) ANCOVA
B) One-Way ANOVA
C) MANOVA
D) Factor analysis

Answer 1

A

Question 2

If the degrees of freedom for the MS error in a one-way ANOVA are 36, what would the number of degrees of freedom be if a one-way ANCOVA with two covariates were run on the same data?

A) 33
B) 34
C) 35
D) 36

Answer 2

B

Question 3

What problem might be reduced through the use of ANCOVA?

A) Unequal loss of subjects from the various treatment groups
B) Measurement regression
C) Initial difference between treatment groups
D) Lack of enough subjects

Answer 3

C

Question 4

What is the primary function of factor analysis?

A) To explain the variance in a data set in as concise a manner as possible
B) To determine the differences among means of several factors
C) To determine significant differences among several dependent variables
D) To correct final scores on the basis of assuming initial inequality

Question 5

Assume that you are conducting a MANOVA, but first you conduct a factor analysis. Why did you complete the factor analysis?

A) It is required to meet MANOVA assumptions
B) It reduces the possibility of a Type II error
C) It results in a smaller number of dependent variables
D) It allows a multivariate analysis to be conducted

Question 6

In general, how does the number of factors resulting from a factor analysis compare to the original number of factored variables?

A) The number of factors is somewhat smaller than the number of original variables.
B) The number of factors is less than the number of original variables.
C) It depends on the number of subjects-to-variables ratio.
D) None of the above is correct.

Question 7

What is the purpose of a factor rotation procedure in factor analysis?

A) To reduce the number of factors found
B) To increase the number of factors in the solution
C) To clarify the factors
D) To increase the total amount of variance that is explained

Question 8

With regard to independent and dependent variables and their levels of measurement, what is the opposite of the discriminant analysis procedure?

A) ANOVA
B) Factor analysis
C) ANCOVA
D) Nonparametric statistics

Question 9

In which statistical technique is the presence of outliers problematic?

A) ANCOVA
B) Multiple regression
C) Discriminant analysis
D) All of the above

Question 10

Which statistical procedure is most appropriate with a pretest/posttest design?

A) MANOVA, because there are multiple groups
B) An independent t-test
C) A four-way ANOVA
D) ANCOVA

Question 11

Which research problem is minimized when the researcher matches treatment and control subjects, assigns subjects to treatments randomly, or uses ANCOVA?

A) Having initial differences among treatment groups
B) Statistical regression
C) Contamination from a previous treatment
D) Significant interaction effects

Question 12

Of the following statistical procedures, which one "contains" all of the others?

A) Pearson correlation
B) Canonical correlation
C) Discriminant analysis
D) Multiple regression

Question 13

What is the number of degrees of freedom associated with the adjusted sum of squares within an ANCOVA with one covariate?

A) J - 1
B) N - J
C) N - 2
D) N - J - 1

Question 14

Which statistical test should be used if the dependent variable is in the form of categories reflecting discrete groups?

A) Factor analysis
B) Discriminate analysis
C) Multiple regression
D) ANCOVA

Question 15

Which statistical test should be used if the dependent variable is in the form of categories reflecting discrete groups?

A) Logistic regression
B) Partial regression
C) Multiple regression
D) Simple regression

Question 16

Which two pieces of information are most helpful in determining what statistical test to employ?

A) The number of levels of the independent variable(s) and the reliability of the measurement for the dependent variable
B) The amount of protection required against making a Type I error and the sample size
C) The mean of the population sampled and the validity of the criterion measure
D) The level of measurement for the independent and dependent variables and the number of independent and dependent variables

Question 17

Following the use of a principle components analysis, a researcher selected a four-factor solution that explained 75% of the variance in the original items. After a rotational procedure on the four-factor model, how much of the variance would be explained?

A) Less than 75%
B) 75%
C) more than 75%
D) Depends on the type of rotation used

Question 18

Of the following statistical techniques, which one does NOT allow the researcher to hypothesize a particular model to explain the data?

A) Path analysis
B) Structural equation modeling
C) Stepwise multiple regression

Question 19

Ultimately, inferential statistics can all be understood conceptually as which type of model?

A) A significance test of differences
B) A regression analysis
C) A t-test
D) An ANOVA

Question 20

Which variables are involved in a multiple regression?

A) Multiple independent variables
B) Multiple dependent variables
C) Both multiple independent and dependent variables
D) Could be either independent or dependent variables, depending on the model used

Question 21

What differentiates logistic regression from other types of regression models?

A) There are multiple independent variables
B) There are multiple dependent variables
C) There are multiple independent and dependent variables
D) The nature of the dependent variable-it is categorical

Question 22

Describe the dependent variable in logistic regression.

A) Continuous and normal
B) Categorical, with any number of levels
C) Simple, with two levels
D) Categorical, with only two levels

Question 23

What is the major difference between logistic regression (LR) and discriminant analysis (DA)?

A) The number of independent variables is greater in DA.
B) The dependent variable has two levels in LR and multiple levels in DA.
C) LR can have multiple outcome levels, whereas DA is more limited.
D) LR is based on the PPM correlation, and DA is based on the canonical model.

Question 24

What types of variables generally serve as covariates in ANCOVA?

A) Continuous only
B) Categorical only
C) Continuous and categorical
D) Adjusted

Question 25

A factorial MANCOVA can be considered a "marriage" between which two procedures?

A) MANOVA and ANOVA
B) ANOVA and multiple regression
C) ANCOVA and MANOVA
D) ANOVA and ANCOVA

Question 26

A researcher wants to determine whether various risk factors predict death. Which type of model should this researcher use?

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

Question 27

Choose the appropriate correlational technique to use
-To describe the strength of the relationship between several independent and several dependent variables

A) PPM
B) partial correlation
C) factor analysis
D) canonical correlation
E) logistic regression

Question 28

Choose the appropriate correlational technique to use
-To describe the strength of the relationship between two variables

A) PPM
B) partial correlation
C) factor analysis
D) canonical correlation
E) logistic regression

Question 29

Choose the appropriate correlational technique to use
-To determine whether a set of variables can be reduced to a smaller set of constructs

A) PPM
B) partial correlation
C) factor analysis
D) canonical correlation
E) logistic regression

Question 30

Choose the appropriate correlational technique to use
-To describe the strength of the relationship between two variables after the influence of a third variable has been removed

A) PPM
B) partial correlation
C) factor analysis
D) canonical correlation
E) logistic regression

Question 31

Choose the appropriate correlational technique to use
-To determine the increased odds of developing heart disease as a function of sedentary lifestyle behaviors and smoking

A) PPM
B) partial correlation
C) factor analysis
D) canonical correlation
E) logistic regression