Practice Quiz Chapter 12

. Of the following selections, which of the following can't be said about multiple regression?

. Which one of the following selections is not useful for understanding the idea of "controlling for" or "holding constant"?

. What is the formula for using multiple regression predict a value for the dependent variable?

. One of the following selections is not true of multiple regression–which one is it?

. What is model specification?

. One of the following selections is not a step necessary for calculating the predicted value on the dependent variable using multiple linear regression–which one is it?

. How is it possible to estimate the proportion of the relationship between an independent variable and a dependent variable that is unique to that independent variable alone?

. Calculate the constant coefficient for multiple linear regression using the following information: the mean of variable 1 is -437.55, the mean of variable 2 is 6.50, the partial slope coefficient for variable 1 is 27.12, the partial slope coefficient for variable 2 is 862.74, and ? is 45,476.81? What is a?

. Calculate the constant coefficient for multiple linear regression using the following information: the mean of variable 1 is 1437.55, the mean of variable 2 is 676.50, the partial slope coefficient for variable 1 is 47.12, the partial slope coefficient for variable 2 is 36.36, and ? is 155,876? What is a?

. Calculate the partial slope coefficient for multiple linear regression using the following information: the correlation between the first independent variable and the dependent variable is 0.5967, the correlation between the first independent variable and the second one is 0.2751, the correlation between the second independent variable and the dependent variable is 0.3614, and, finally, the standard deviation of the dependent variable is 34970.38 and the standard deviation for the first independent variable is 693.84. What is bx1?

. Calculate the partial slope coefficient for the second independent variable using the same information from the question above. What is bx2?

. What is a beta coefficient?

. When thinking about standardized slope coefficients, if one standard deviation of the variable "people's annual wages" is $34,970.38 then how many standard deviations is $27.13?

. When thinking about standardized slope coefficients again, if one standard deviation of the variable "number of hours worked on the job" is 693.84 then how many standard deviations is 1 hour?

. Which of the following statements can't be said about Ăź?

. Researchers consider several things when they decide which attribute or category to use as a reference group–which one of the following selections is not one of these things?

. What are dummy variables?

. What does rescaling a variable mean?

. Reporting the results of regressions requires some additional writing strategies–which of the following selections is not one of them?

. What is another name for dummy variables?

. In Canadian surveys, the measurement of disability relies on assessing how much difficulty a person has doing everyday tasks and how often these difficulties limit their daily activities; the definition does not consider self-identification.

. The best way to assess how multiple independent variables are related to a dependent variable is to perform a series of simple linear regressions that each predict the relationship between a different independent variable and the same dependent variable.

. The plane of best fit for multiple linear regression is the one that minimizes the sum of squared differences of each case from the plane.

. A multiple regression predicts the plane of best fit, but it is still a linear regression because it predicts straight-line relationships between variables.

. Subscript numbers are used to keep track of the different dependent variables and their respective slope coefficients.

. The multiple regression determines how much of the variation in the dependent variable is uniquely attributable to each independent variable alone, and accounts for correlation between them.

. The main advantage of using multiple regression is that it lets researchers sort out the unique relationship between each independent variable and the dependent variable, while accounting for the relationships between all the independent variables.

. Researchers seek to build regression models that capture all the predictors of an independent variable but exclude non-meaningful dependent variables.

. The starting point for selecting independent variables should always be the most current knowledge about what the dependent variable is related to.

. Typically, omitting an important independent variable from a regression makes the slope coefficients of other independent variables higher or lower than they would be if the variable were not omitted.

. Concerns about model specification and omitted variable bias motivate researchers to carefully consider which independent variables they will use in a regression.

. To account for correlation between two independent variables in multiple regression, researchers rely on two other relationships: the Pearson's correlation between two independent variables and the Pearson's correlation between the second independent variable and the dependent one.

. Standardized slope coefficients are necessary when slope coefficients are all measured in the same units.

. A standardized slope coefficient shows the "rise" in the dependent variable measured in standard deviations associated with a one standard deviation increase in the independent variable.

. Independent variables with two different standard deviations have different potential influences on the dependent variable.

. The independent variable with the largest absolute standardized slope coefficient has the weakest relationship with the dependent variable.

. When an independent variable is a dummy variable the regression slope coefficient shows the change in the dependent variable associated with having some characteristic.

. Categorical variables with more than two attributes can also be used as independent variables in a regression using more dummy variables.

. Researchers typically create dummy variables by recoding the original variable into new variables multiple times.

. Researchers tend to spend very little time developing and refining multiple regression models as they try to use these models to accurately capture the social characteristics and processes related to the dependent variable.

Back to top