Question 1

Exhibit 13-6: A researcher collects the following measures on a sample of college students: scores on the SAT and the ACT, scores on an English composition test, high school class rank, scores on a noncognitive measure, whether or not students successfully completed the first year of college (pass-fail), and their GPA.Choose the statistical procedures that the researcher should use to answer the following questions.
-Refer to Exhibit 13-6.Is SAT or ACT the better single predictor of GPA?

Accepted Answer

The product-moment correlation is used to determine the strength and direction of the linear relationship between two continuous variables, such as SAT or ACT scores and GPA.

Question 2

A measure of political conservatism was administered to representative samples of persons of ages 18, 21, 30, 45, and 60, and their respective mean scores were 65, 90, 85, 75, 70.The correlation between age and political conservatism is

Accepted Answer

The pattern of mean scores across ages does not show a clear linear relationship, but rather a curvilinear one. There seems to be an increase in conservatism from age 18 to 30, a plateau from age 30 to 45, and a decline from age 45 to 60. Therefore, the correlation between age and political conservatism is best described as curvilinear.

Question 3

What is the best way for a researcher, in an exploratory study, to show the correlations among all the variables?

Accepted Answer

A correlation matrix is the best way to show the correlations among all the variables in an exploratory study. It displays a table of correlation coefficients between all possible pairs of variables, providing a comprehensive overview of the relationships among the variables. Scattergrams and linear regression equations are useful tools for visualizing and modeling individual relationships between two variables, while correlation ratios and expectancy tables are not commonly used to display correlations among multiple variables.

Question 4

If a test is going to predict a criterion well, the correlation between the test and the criterion should be

Accepted Answer

A high correlation between a test and a criterion, whether positive or negative, indicates that the test predicts the criterion well. A positive correlation means that as test scores increase, criterion measures increase, while a negative correlation means that as test scores increase, criterion measures decrease. Both scenarios indicate predictive validity.

Question 5

Exhibit 13-5.Choose the type of correlational statistic that would be used for each of the purposes listed.
-Refer to Exhibit 13-5.Used to describe the relationship between two variables measured on an interval scale

Accepted Answer

The product-moment correlation is used to describe the relationship between two variables measured on an interval scale.

Question 6

An advantage of correlational studies is that they

Accepted Answer

Correlational studies are useful because they provide a measure of the strength of the relationship between two variables, usually expressed as a correlation coefficient. However, they do not isolate cause-effect relationships or control extraneous variables as experimental studies do, and they do not focus on individuals or small groups in depth.

Question 7

In which college would you expect the correlation between aptitude test score and gradepoint average to be greatest? $\begin{array}{ll}
&&\underline {\text { College }}\
&\text { A } & \text { B } & \text { C }\
\text {Mean Aptitude } &650 & 775&800 \
\text { Standard Deviation } &71 & 80&68
\end{array}$

Accepted Answer

The answer of In which college would you expect the...

Question 8

The correlation between a predictor and a criterion is .84.This means that the criterion has about ____ percent of its variance in common with the predictor.

Accepted Answer

The correlation coefficient squared (R²) represents the proportion of variance in the criterion that can be predicted by the predictor. In this case, R² = (.84)² = .7056 or approximately 70.56%, so the criterion has about 70% of its variance in common with the predictor. Therefore, the best choice is C.

Question 9

Exhibit 13-6: A researcher collects the following measures on a sample of college students: scores on the SAT and the ACT, scores on an English composition test, high school class rank, scores on a noncognitive measure, whether or not students successfully completed the first year of college (pass-fail), and their GPA.Choose the statistical procedures that the researcher should use to answer the following questions.
-Refer to Exhibit 13-6.Does the combination of variables predict whether or not students successfully completed the first year of college?

Accepted Answer

Discriminant analysis is used to predict a categorical dependent variable, such as pass-fail, based on several predictor variables.

Question 10

Exhibit 13-6: A researcher collects the following measures on a sample of college students: scores on the SAT and the ACT, scores on an English composition test, high school class rank, scores on a noncognitive measure, whether or not students successfully completed the first year of college (pass-fail), and their GPA.Choose the statistical procedures that the researcher should use to answer the following questions.
-Refer to Exhibit 13-6.Do all the measures used in this study measure something in common?

Accepted Answer

Factor analysis will help identify any underlying dimensions or factors that the measures may be related to, indicating whether they have something in common or not.

Question 11

A study with college students found no correlation between hours of study for an examination and test performance.This probably means

Accepted Answer

The lack of correlation does not necessarily imply a lack of effect, a negative relationship, or the influence of a third variable. It simply means that in this study, no direct relationship was found between study hours and test performance. Other factors or methodological issues could be at play.

Question 12

Cohen's guidelines describe a correlation of +.50 as

Accepted Answer

Cohen's guidelines typically describe a correlation of +.50 as a large effect size.

Question 13

A researcher uses several different measures in an attempt to predict completion of a doctoral degree in the School of Education.What type of correlational analysis would be most appropriate to predict this criterion?

Accepted Answer

The answer of A researcher uses several different measures in...

Question 14

Statistical significance of a correlation tells us

Accepted Answer

Statistical significance of a correlation tells us how likely it is that the observed correlation is only a function of chance.

Question 15

Would a correlation of -.50 be described in the same way?

Accepted Answer

Yes, a correlation of -.50 would also be described as a moderate negative correlation.

Question 16

Exhibit 13-5.Choose the type of correlational statistic that would be used for each of the purposes listed.
-Refer to Exhibit 13-5.Used to determine the relationship between several independent variables and several dependent variables.

Accepted Answer

Canonical correlation is used to determine the relationship between several independent variables and several dependent variables.

Question 17

Exhibit 13-7: Choose the index of correlation appropriate for use in each of the following.
Exhibit
-Refer to Exhibit 13-7.Aptitude test scores and GPA

Accepted Answer

Since the variables being compared are both continuous (aptitude test scores and GPA), Pearson's r is the appropriate index of correlation. Spearman rho is used when at least one of the variables is ordinal, and Phi coefficient is used when both variables are dichotomous.

Question 18

A researcher wants to hold scholastic aptitude constant when investigating the relationship between need for achievement and actual achievement in the classroom.What statistical procedure should the researcher use?

Accepted Answer

The researcher should use partial correlation to hold scholastic aptitude constant and investigate the relationship between need for achievement and actual achievement in the classroom. Partial correlation is a statistical technique that removes the effect of a third variable (in this case, scholastic aptitude) on the relationship between two other variables (need for achievement and actual achievement).

Question 19

Exhibit 13-6: A researcher collects the following measures on a sample of college students: scores on the SAT and the ACT, scores on an English composition test, high school class rank, scores on a noncognitive measure, whether or not students successfully completed the first year of college (pass-fail), and their GPA.Choose the statistical procedures that the researcher should use to answer the following questions.
-Refer to Exhibit 13-6.What is the correlation between the combination of independent variables and GPA at the end of the freshman year as dependent variable?

Accepted Answer

Multiple regression is the appropriate statistical procedure to determine the correlation between the combination of independent variables (SAT, ACT, English composition test scores, high school class rank, and scores on a noncognitive measure) and the dependent variable (GPA at the end of the freshman year).

Question 20

Exhibit 13-6: A researcher collects the following measures on a sample of college students: scores on the SAT and the ACT, scores on an English composition test, high school class rank, scores on a noncognitive measure, whether or not students successfully completed the first year of college (pass-fail), and their GPA.Choose the statistical procedures that the researcher should use to answer the following questions.
-Refer to Exhibit 13-6.What is the correlation between the noncognitive measure and college GPA after the influence of SAT on both variables has been removed?

Accepted Answer

Partial correlation would allow the researcher to examine the relationship between the noncognitive measure and college GPA while controlling for the influence of SAT scores.

Quiz 13: Correlation Research