Deck 3: Data Screening

Question 1

Which one of the following would suggest the assumption of univariate normality has been met?

A) boxplot evidenced by one or more outliers
B) nonstatistically significant Shapiro-Wilk test
C) Q-Q plot with data points not adhering to the diagonal line
D) skewness value greater than 10

Answer 1

B
Explanation: Nonstatistically significant results are desirable and indicate that the sample distribution is not statistically significantly different than what would be expected from a normal distribution.

Question 2

Which one of the following would be considered a relatively normal distribution?

A) Skewness = 2.0, kurtosis = 12.0
B) Skewness = 2.0, kurtosis = 1.0
C) Skewness = 20.0, kurtosis = 1.0
D) Skewness = 12.0, kurtosis = 20.0

Answer 2

B
Explanation: A relatively normal distribution will be evident when skewness is within an absolute value of 2.0 and kurtosis within an absolute value of 7.0.

Question 3

Which one of the following assumptions helps ensure that generalizations from the sample back to the population can be made?

A) homogeneity of variance
B) independence
C) linearity
D) normality

Answer 3

B
Explanation: If we have independence, then generalizations from the sample back to the population can be made (you may remember this as external validity, which was likely introduced in your research methods course).

Question 4

Which one of the following assumptions refers to dispersion of the dependent variable being approximately equal at all levels or categories of the independent variable?

A) homogeneity of variance
B) independence
C) linearity
D) normality

Question 5

Which one of the following statements is correct?

A) Meeting univariate normality implies tenable multivariate normality
B) Rejecting the assumption of univariate normality means that multivariate normality is not necessarily rejected
C) Univariate normality is a necessary condition for multivariate normality
D) Univariate normality is a sufficient condition for multivariate normality

Question 6

A researcher has computed Shapiro-Wilk. The level of significance was .05 and the p value was .03. Which one of the following is suggested?

A) There is evidence the assumption of linearity has been met
B) There is evidence the assumption of linearity has NOT been met
C) There is evidence the assumption of normality has been met
D) There is evidence the assumption of normality has NOT been met

Question 7

A researcher has computed Shapiro-Wilk. The p value was .10 and the significance level was .01. Which one of the following is suggested?

A) There is evidence the assumption of linearity has been met
B) There is evidence the assumption of linearity has NOT been met
C) There is evidence the assumption of normality has been met
D) There is evidence the assumption of normality has NOT been met

Question 8

Quantile-quantile plots can be used to determine evidence of which one of the following?

A) homogeneity of variance
B) independence
C) linearity
D) normality

Question 9

In the context of MANOVA, when all pairs of dependent variables are bivariate normally distributed (i.e., linear), which one of the following assumptions is met?

A) homogeneity of variance
B) independence
C) linearity
D) normality

Question 10

Which one of the following occurs when two or more variables perfectly predict and are therefore perfectly redundant?

A) homogeneity of variance
B) linearity
C) noncollinearity
D) singularity