Chapter 24: Using Statistics to Predict

Chapter 24: Using Statistics to Predict

Complete Chapter Questions With Answers

Sample Questions Are Posted Below

MULTIPLE CHOICE

1. A researcher is studying daily carbohydrate intake and the next day’s first morning Accu-Chek value. After drawing a scatter plot, the researcher develops a multiple regression equation. What does the value R represent?

ANS:  D

R is defined as the correlation between the actual y values and the predicted y values using the new regression equation. The R² represents the percentage of variance in y explained by the predictor and is called the coefficient of determination.

DIF:    Cognitive Level: Application           REF:   Page 572

2. A researcher is studying stress and various factors that seem to be related to it, in women with terminally ill spouses.Among other things, the researcher finds out that depression contributes to total stress; depression contributes to lack of exercise; a decreased amount of exercise  contributes to depression; exercise relieves stress; guilt contributes to depression; exercise has no effect on guilt, but it affects total stress; stress worsens depression; and stress decreases motivation to perform exercise. This is an example of what problem commonly encountered in regression analyses?

ANS:  A

Multicollinearity occurs when the independent variables in a multiple regression equation are strongly correlated with one another. The presence of multicollinearity does not affect predictive power; rather it causes problems related to generalizability and the stability of the findings. The first step in identifying multicollinearity is to examine the bivariate correlations among the independent variables.

DIF:    Cognitive Level: Application           REF:   Page 573

3. Why does a researcher decide to calculate odds ratio instead of calculating linear regression when comparing whether a person voted in the last election and the person’s gender?

ANS:  B

When both the predictor and the dependent variable are dichotomous (having only two values; also called binary), the odds ratio is a commonly used statistic to obtain an indication of association. Linear regression is used to determine relationships with interval or ratio data. The odds ratio (OR) is defined as the ratio of the odds of an event occurring in one group to the odds of it occurring in another group. Put simply, the OR is a way of comparing whether the odds of a certain event is the same for two groups.

DIF:    Cognitive Level: Analysis                REF:   Page 575