Deck 17: Bivariate Statistics

Question 1

If the F value indicates that the between-groups variance estimate is significantly greater than the within-groups variance estimate in an ANOVA, we can conclude that

A) there is random error.
B) the group means are from the same population.
C) the null hypothesis that the differences between the groups are due only to chance can be rejected.
D) there was an error in the calculation.

Answer 1

C

Question 2

Within-group variance in an ANOVA is considered to indicate

A) systematic error.
B) differences among the group means.
C) that the group means are from the same population.
D) random error.

Answer 2

D

Question 3

Which of the following is NOT a bivariate statistic?

A) Pearson correlation coefficient
B) Chi-square
C) One-way ANOVA
D) Multiple regression

Answer 3

D

Question 4

Researcher Jones is interested in assessing the relationship between sex of participant and helping behavior (whether the person helped or did not help). Which of the following statistics should be used?

A) Chi-square statistic
B) Kappa
C) Bivariate regression
D) Pearson's r

Question 5

A psychologist wanted to determine whether attitudes toward capital punishment were affected by religious commitment. He chose a sample of sixty people, thirty of whom were "very religious" and thirty of whom were "not religious." He then assessed the person's attitude toward capital punishment. Thirteen of the "very religious" people were opposed to capital punishment, whereas only eight of the "not religious" people were opposed. What would be the most appropriate statistical analysis for this experimental data?

A) Bivariate regression
B) One-way analysis of variance
C) Correlation coefficient
D) Chi-square test for independence

Question 6

A negative correlation coefficient indicates that

A) there is no relationship between the variables.
B) high scores on one variable are associated with low scores on the other variable.
C) high scores on one variable are associated with high scores on the other variable.
D) it was not calculated accurately.

Question 7

As the value of r increases, the proportion of variability statistic

A) increases.
B) decreases.
C) does not change.
D) may increase or decrease.

Question 8

A researcher is interested in whether students' grade point averages differed in between those enrolled in lab science, social science, humanities, and mathematics courses. She compared students from a chemistry, an economics, an art, and a calculus class. What would be the most appropriate statistical analysis for these data?

A) Bivariate regression
B) One-way analysis of variance
C) Correlation coefficient
D) Chi-square test for independence

Question 9

Statistical methods used to measure the relationships between two variables are referred to as

A) inferential statistics.
B) correlational statistics.
C) bivariate statistics.
D) multivariate statistics.

Question 10

For a 2 x 2 contingency table, the appropriate effect size statistic is

A) r.
B) Cohen's kappa.
C) Phi.
D) Cramer's statistic.

Question 11

A researcher calculates a regression equation in order to

A) predict a score on a second variable from a known score on a first variable.
B) determine the strength of the relationship between two nominal variables.
C) predict the differences between scores on one variable from the scores on a second variable
D) determine the causal relationship between two variables.

Question 12

Which of the following Pearson correlation coefficients indicates the strongest relationship between two variables?

A) .56
B) -.81
C) -.05
D) .34

Question 13

If the F value indicates that the between-groups variance is not significantly greater than the within-groups variance in an ANOVA, we can conclude that

A) there is random error.
B) the group means are from the same population.
C) the null hypothesis that the differences between the groups are due only to chance can be rejected.
D) there was an error in the calculation.

Question 14

A researcher wanted to know if aggressive behavior was related to watching violent programs on TV. He selected some students and measured how many violent programs they watched per week (from 0 to 25) and how aggressive their behavior was (as rated on a 1 to 7 scale). What would be the most appropriate statistical analysis for these data?

A) Bivariate regression
B) One-way analysis of variance
C) Multiple regression
D) Chi-square test for independence

Question 15

A correlation coefficient specifies the __________ and __________ of the linear association between two variables.

A) direction; magnitude
B) strength; shape
C) direction; distribution
D) shape; distribution

Question 16

The proportion of variance measure for the Pearson correlation coefficient is

A) r.
B) the effect size.
C) the coefficient of determination.
D) the contingency table.

Question 17

__________ is used to compare the means on a dependent variable between two or more groups of participants who differ on a single independent variable.

A) One-way analysis of variance
B) Regression
C) A contingency table
D) Kappa

Question 18

If a correlation is positive, this indicates that __________ scores on one variable are associated with __________ scores on another variable.

A) low; high
B) high; low
C) high; high
D) high; medium

Question 19

What statistic is used to assess the degree of association between two nominal variables?

A) The chi-square statistic
B) Bivariate regression
C) ANOVA
D) Pearson's r

Question 20

The coefficient of determination is to the Pearson correlation coefficient as which of the following is to chi-square?

A) R
B) F
C) The p-value
D) Phi

Question 21

The coefficient of determination indicates

A) the proportion of variance accounted for by the correlation coefficient.
B) whether the null hypothesis can be rejected.
C) whether the correlation is positive or negative.
D) whether the population has been randomly sampled.

Question 22

In a contingency table of opinion about a new state delegate, out of the 300 total respondents 132 favor the new delegate. Which of the following equations would be used to find the expected count of the fifty African Americans within this study who favored the new delegate?

A) (50 x 132) / 300
B) (50 x 132) + 300
C) (132 / 300) - 52
D) (132-300) x 52

Question 23

To predict a dependent variable from knowledge of an independent variable, which of the following would be most useful?

A) A regression equation
B) A post-hoc test
C) Means comparisons
D) A random sample

Question 24

The effect size for a one-way ANOVA is

A) the F value.
B) the degrees of freedom.
C) eta.
D) the Pearson r.

Question 25

The regression line measures

A) the overall effectiveness of the dependent variable.
B) the ability to predict a correlation coefficient.
C) the average effect size.
D) the best possible prediction of the dependent variable

Question 26

If r =1.00, then

A) all of the points on the scatterplot will fall on the line of best fit.
B) there is a weak positive correlation.
C) variable x causes variable y.
D) the r will not be statistically significant.

Question 27

The ratio of the between-groups variance estimate to the within-groups variance estimate is indicated by

A) the F value.
B) the degrees of freedom.
C) the sum of squares.
D) eta.

Question 28

Kramer is conducting a chi-square analysis on a 2 x 2 contingency table. Which statistic should he use to calculate the effect size?

A) Kappa
B) Alpha
C) Phi
D) r

Question 29

In a one-way ANOVA the grand mean is defined as

A) the between groups variance.
B) the means across all of the participants in the study.
C) only the means from the control group.
D) n times the degrees of freedom.

Question 30

Which of the following statistical procedures could be used to measure the relationship between two quantitative variables?

A) Chi-square analysis
B) One way ANOVA
C) Contingency table analysis
D) Bivariate regression

Question 31

In a one-way ANOVA, eta is the measure of

A) effect size.
B) mean square variance.
C) statistical significance.
D) the interaction.

Question 32

In a one-way ANOVA, if the F-obtained is less than F-critical, then

A) the null hypothesis can be rejected.
B) a Type 1 error has been made.
C) a lower alpha needs to be used.
D) the null hypothesis cannot be rejected.

Question 33

Which of the following is the measure of effect size in a one-way ANOVA?

A) eta
B) F
C) b
D) p