We've detected that Javascript is not enabled. It is required for an optimal survey taking experience. Please check your browser's settings and make sure Javascript is turned on. Learn how to enable Javascript.

Please answer each of the following questions to help you self-assess your understanding of "Chapter 9: Making Sense of Multivariate Statistics" (Remler & Van Ryzin, 2010)

1. (OPTIONAL) Your email address This question requires a valid email address.

2. Please Match the Term to Its Definition *This question is required.

Space Cell

An adjusted version of R-squared that takes into consideration the number of independent variables. Technically, an unbiased estimator of the population R-squared - the proportion of the dependent variable variance explained by all the independent variables in the population.

A statistical analysis that compares the means across groups, normally used in analysis of experimental data.

The best linear predictor of a dependent variable using more than one independent variable.

Making a prediction using a fitted model (particularly regression) far from the data used to fit the model.

Statistics examining the relationships between multiple (more than two) variables at the same time.

Phenomenon in which an independent variable is a linear combination of two more of the other independent variables.

Adjusted R-Squared

An adjusted version of R-squared that takes into consideration the number of independent variables. Technically, an unbiased estimator of the population R-squared - the proportion of the dependent variable variance explained by all the independent variables in the population.

A statistical analysis that compares the means across groups, normally used in analysis of experimental data.

The best linear predictor of a dependent variable using more than one independent variable.

Making a prediction using a fitted model (particularly regression) far from the data used to fit the model.

Statistics examining the relationships between multiple (more than two) variables at the same time.

Phenomenon in which an independent variable is a linear combination of two more of the other independent variables.

ANOVA

An adjusted version of R-squared that takes into consideration the number of independent variables. Technically, an unbiased estimator of the population R-squared - the proportion of the dependent variable variance explained by all the independent variables in the population.

A statistical analysis that compares the means across groups, normally used in analysis of experimental data.

The best linear predictor of a dependent variable using more than one independent variable.

Making a prediction using a fitted model (particularly regression) far from the data used to fit the model.

Statistics examining the relationships between multiple (more than two) variables at the same time.

Phenomenon in which an independent variable is a linear combination of two more of the other independent variables.

Multiple Regression

An adjusted version of R-squared that takes into consideration the number of independent variables. Technically, an unbiased estimator of the population R-squared - the proportion of the dependent variable variance explained by all the independent variables in the population.

A statistical analysis that compares the means across groups, normally used in analysis of experimental data.

The best linear predictor of a dependent variable using more than one independent variable.

Making a prediction using a fitted model (particularly regression) far from the data used to fit the model.

Statistics examining the relationships between multiple (more than two) variables at the same time.

Phenomenon in which an independent variable is a linear combination of two more of the other independent variables.

Multicollinearity

An adjusted version of R-squared that takes into consideration the number of independent variables. Technically, an unbiased estimator of the population R-squared - the proportion of the dependent variable variance explained by all the independent variables in the population.

A statistical analysis that compares the means across groups, normally used in analysis of experimental data.

The best linear predictor of a dependent variable using more than one independent variable.

Making a prediction using a fitted model (particularly regression) far from the data used to fit the model.

Statistics examining the relationships between multiple (more than two) variables at the same time.

Phenomenon in which an independent variable is a linear combination of two more of the other independent variables.

Out-of-Sample Extrapolation

An adjusted version of R-squared that takes into consideration the number of independent variables. Technically, an unbiased estimator of the population R-squared - the proportion of the dependent variable variance explained by all the independent variables in the population.

A statistical analysis that compares the means across groups, normally used in analysis of experimental data.

The best linear predictor of a dependent variable using more than one independent variable.

Making a prediction using a fitted model (particularly regression) far from the data used to fit the model.

Statistics examining the relationships between multiple (more than two) variables at the same time.

Phenomenon in which an independent variable is a linear combination of two more of the other independent variables.

Multivariate Statistics

An adjusted version of R-squared that takes into consideration the number of independent variables. Technically, an unbiased estimator of the population R-squared - the proportion of the dependent variable variance explained by all the independent variables in the population.

A statistical analysis that compares the means across groups, normally used in analysis of experimental data.

The best linear predictor of a dependent variable using more than one independent variable.

Making a prediction using a fitted model (particularly regression) far from the data used to fit the model.

Statistics examining the relationships between multiple (more than two) variables at the same time.

Phenomenon in which an independent variable is a linear combination of two more of the other independent variables.

3. Please Match the Term to Its Definition *This question is required.

Space Cell

Model predicting the log odds of an event.

The predicted difference in the probability due to a specified change in the relevant independent variable.

A variable defined as the product of two other variables, usually used to empirically measure an interaction.

Ordinary least squares regression model in which dependent variable is a dummy variable and predicted values of the dependent variable are interpreted as probabilities.

The effect of an independent variable on a dependent variable, before or without its moderation by another variable (interaction).

Method that estimates the pattern of relationships between variables in a presumed causal structure.

Marginal Effect

Model predicting the log odds of an event.

The predicted difference in the probability due to a specified change in the relevant independent variable.

A variable defined as the product of two other variables, usually used to empirically measure an interaction.

Ordinary least squares regression model in which dependent variable is a dummy variable and predicted values of the dependent variable are interpreted as probabilities.

The effect of an independent variable on a dependent variable, before or without its moderation by another variable (interaction).

Method that estimates the pattern of relationships between variables in a presumed causal structure.

Logistic Regression

Model predicting the log odds of an event.

The predicted difference in the probability due to a specified change in the relevant independent variable.

A variable defined as the product of two other variables, usually used to empirically measure an interaction.

Ordinary least squares regression model in which dependent variable is a dummy variable and predicted values of the dependent variable are interpreted as probabilities.

The effect of an independent variable on a dependent variable, before or without its moderation by another variable (interaction).

Method that estimates the pattern of relationships between variables in a presumed causal structure.

Interaction Variable

Model predicting the log odds of an event.

The predicted difference in the probability due to a specified change in the relevant independent variable.

A variable defined as the product of two other variables, usually used to empirically measure an interaction.

Ordinary least squares regression model in which dependent variable is a dummy variable and predicted values of the dependent variable are interpreted as probabilities.

The effect of an independent variable on a dependent variable, before or without its moderation by another variable (interaction).

Method that estimates the pattern of relationships between variables in a presumed causal structure.

Linear Probability Model

Model predicting the log odds of an event.

The predicted difference in the probability due to a specified change in the relevant independent variable.

A variable defined as the product of two other variables, usually used to empirically measure an interaction.

Ordinary least squares regression model in which dependent variable is a dummy variable and predicted values of the dependent variable are interpreted as probabilities.

The effect of an independent variable on a dependent variable, before or without its moderation by another variable (interaction).

Method that estimates the pattern of relationships between variables in a presumed causal structure.

Main Effect

Model predicting the log odds of an event.

The predicted difference in the probability due to a specified change in the relevant independent variable.

A variable defined as the product of two other variables, usually used to empirically measure an interaction.

Ordinary least squares regression model in which dependent variable is a dummy variable and predicted values of the dependent variable are interpreted as probabilities.

The effect of an independent variable on a dependent variable, before or without its moderation by another variable (interaction).

Method that estimates the pattern of relationships between variables in a presumed causal structure.

Path Analysis

Model predicting the log odds of an event.

The predicted difference in the probability due to a specified change in the relevant independent variable.

A variable defined as the product of two other variables, usually used to empirically measure an interaction.

Ordinary least squares regression model in which dependent variable is a dummy variable and predicted values of the dependent variable are interpreted as probabilities.

The effect of an independent variable on a dependent variable, before or without its moderation by another variable (interaction).

Method that estimates the pattern of relationships between variables in a presumed causal structure.

4. Please Match the Term to Its Definition *This question is required.

Space Cell

Correlations between the observed variables and the underlying and unobserved factors.

Multivariate method that groups many variables (indicators) into a smaller set of clusters or underlying factors.

Factor analysis in which the number of factors and how items correlate with factors are discovered by the procedure rather than specified in advance by the researcher.

Factor analysis in which theory is used to impose both the number of factors and which variables load onto each factor.

Factor analysis in which the researcher does not use theory to impose a structure but lets the computer choose the number of factors and estimate how the items correlate with each factor.

Factor Analysis

Correlations between the observed variables and the underlying and unobserved factors.

Multivariate method that groups many variables (indicators) into a smaller set of clusters or underlying factors.

Factor analysis in which the number of factors and how items correlate with factors are discovered by the procedure rather than specified in advance by the researcher.

Factor analysis in which theory is used to impose both the number of factors and which variables load onto each factor.

Factor analysis in which the researcher does not use theory to impose a structure but lets the computer choose the number of factors and estimate how the items correlate with each factor.

Exploratory Factor Analysis

Correlations between the observed variables and the underlying and unobserved factors.

Multivariate method that groups many variables (indicators) into a smaller set of clusters or underlying factors.

Factor analysis in which the number of factors and how items correlate with factors are discovered by the procedure rather than specified in advance by the researcher.

Factor analysis in which theory is used to impose both the number of factors and which variables load onto each factor.

Factor analysis in which the researcher does not use theory to impose a structure but lets the computer choose the number of factors and estimate how the items correlate with each factor.

Confirmatory Factor Analysis

Correlations between the observed variables and the underlying and unobserved factors.

Multivariate method that groups many variables (indicators) into a smaller set of clusters or underlying factors.

Factor analysis in which the number of factors and how items correlate with factors are discovered by the procedure rather than specified in advance by the researcher.

Factor analysis in which theory is used to impose both the number of factors and which variables load onto each factor.

Factor analysis in which the researcher does not use theory to impose a structure but lets the computer choose the number of factors and estimate how the items correlate with each factor.

Exploratory Factor Analysis

Correlations between the observed variables and the underlying and unobserved factors.

Multivariate method that groups many variables (indicators) into a smaller set of clusters or underlying factors.

Factor analysis in which the number of factors and how items correlate with factors are discovered by the procedure rather than specified in advance by the researcher.

Factor analysis in which theory is used to impose both the number of factors and which variables load onto each factor.

Factor analysis in which the researcher does not use theory to impose a structure but lets the computer choose the number of factors and estimate how the items correlate with each factor.

Factor Loadings

Correlations between the observed variables and the underlying and unobserved factors.

Multivariate method that groups many variables (indicators) into a smaller set of clusters or underlying factors.

Factor analysis in which the number of factors and how items correlate with factors are discovered by the procedure rather than specified in advance by the researcher.

Factor analysis in which theory is used to impose both the number of factors and which variables load onto each factor.

Factor analysis in which the researcher does not use theory to impose a structure but lets the computer choose the number of factors and estimate how the items correlate with each factor.

5. Please Match the Term to Its Definition *This question is required.

Space Cell

Multivariate method for estimating models in which observed indicators represent latent variables and also latent variables are related to each other in a presumed causal (structural) manner similar to path analysis.

Using data from times series in the past to predict future values of the dependent variable(s).

Models that describe relationships between variables at different units of analysis. Also known as hierarchical models.

The effect, outcome, prediction, or response from a cause or independent variable - the variable the researcher is trying to explain.

Method to predict the length of time until some event.

Structural Equation Modeling (SEM)

Multivariate method for estimating models in which observed indicators represent latent variables and also latent variables are related to each other in a presumed causal (structural) manner similar to path analysis.

Using data from times series in the past to predict future values of the dependent variable(s).

Models that describe relationships between variables at different units of analysis. Also known as hierarchical models.

The effect, outcome, prediction, or response from a cause or independent variable - the variable the researcher is trying to explain.

Method to predict the length of time until some event.

Multilevel Models

Multivariate method for estimating models in which observed indicators represent latent variables and also latent variables are related to each other in a presumed causal (structural) manner similar to path analysis.

Using data from times series in the past to predict future values of the dependent variable(s).

Models that describe relationships between variables at different units of analysis. Also known as hierarchical models.

The effect, outcome, prediction, or response from a cause or independent variable - the variable the researcher is trying to explain.

Method to predict the length of time until some event.

Forecasting

Multivariate method for estimating models in which observed indicators represent latent variables and also latent variables are related to each other in a presumed causal (structural) manner similar to path analysis.

Using data from times series in the past to predict future values of the dependent variable(s).

Models that describe relationships between variables at different units of analysis. Also known as hierarchical models.

The effect, outcome, prediction, or response from a cause or independent variable - the variable the researcher is trying to explain.

Method to predict the length of time until some event.

Dependent Variable

Multivariate method for estimating models in which observed indicators represent latent variables and also latent variables are related to each other in a presumed causal (structural) manner similar to path analysis.

Using data from times series in the past to predict future values of the dependent variable(s).

Models that describe relationships between variables at different units of analysis. Also known as hierarchical models.

The effect, outcome, prediction, or response from a cause or independent variable - the variable the researcher is trying to explain.

Method to predict the length of time until some event.

Survival Analysis

Multivariate method for estimating models in which observed indicators represent latent variables and also latent variables are related to each other in a presumed causal (structural) manner similar to path analysis.

Using data from times series in the past to predict future values of the dependent variable(s).

Models that describe relationships between variables at different units of analysis. Also known as hierarchical models.

The effect, outcome, prediction, or response from a cause or independent variable - the variable the researcher is trying to explain.

Method to predict the length of time until some event.

6. ___ is a data reduction method that does not impose or assume a structure. *This question is required.

A. Exploratory factor analysis

B. Confirmatory factor analysis

C. Path analysis

D. Both A and B

7. ____ often consist of repeated measures of the same variable from the same individuals over time *This question is required.

Panel data

Time series data

Cross sectional data

Forecasted data

8. A simple regression is run with the log of annual income in dollars as the dependent variable and years of education as the independent variable. The coefficient of education will provide the following information: *This question is required.

The percent change in income predicted for a one percent increase in education

The percent change in income predicted for a one-year increase in education

The dollar change in income predicted for a one percent increase in education

The dollar change in income predicted for a one-year increase in education

None of the above

9. In a survey, an individualâ€™s health insurance status is reported as being in one of the following categories: Medicare, Medicaid, Employer provided insurance, Individual private insurance, Other insurance or Uninsured. If health insurance status is to be used as an independent variable in a regression, the following are acceptable combinations of dummy variables *This question is required.

A. Medicare, Medicaid, Employer, Individual, Other, Uninsured

B. Medicare, Medicaid, Employer, Individual, Other

C. Medicare, Medicaid, Employer, Individual, Uninsured