research methods psych chapter 9

multiple regression

A statistical technique that computes the relationship between a predictor variable and a criterion variable, controlling for other predictor variables. Also called multivariate regression.

multivariate design

A study designed to test an association involving more than two measured variables.

longitudinal design

A study in which the same variables are measured in the same people at different points in time.

predictor variable

A variable in multiple-regression analysis that is used to explain variance in the criterion variable. Also called independent variable.

mediator

A variable that helps explain the relationship between two other variables. Also called mediating variable.

Onwuegbuzie et al. (1999) examined foreign-language anxiety as a predictor of foreign-language achievement. They performed a multiple-regression analysis and included other predictor variables, including academic achievement, perceived scholastic competence, and perceived self-worth. They found β = −.39 for foreign-language anxiety and .29 for academic achievement. Which of the following explains what this means?

Foreign-language anxiety has a greater effect than academic achievement does on foreign-language achievement. The absolute value of beta is larger for foreign-language anxiety than for academic achievement, so it has a greater effect. See the section "Beta Basics"

Onwuegbuzie et al. (1999) examined foreign-language anxiety as a predictor of foreign-language achievement. They performed a multiple-regression analysis and included other predictor variables, including academic achievement, perceived scholastic competence, and perceived self-worth. They found β = −.39 for foreign-language anxiety. Which of the following explains what this means?

People with lower levels of anxiety about learning a foreign language have higher levels of achievement in foreign language, even after controlling for the other variables. A negative beta means there is a negative relationship between the predictor variable and the dependent variable, so foreign-language achievement decreases with increasing levels of foreign-language anxiety. See the section "Beta Basics"

In a study in which two variables are measured at two different time points, which of the following is an example of a cross-lag correlation?

Variable A at time 1 is associated with variable B at time 2. Cross-lag correlations are used in a longitudinal design to check if a variable measured at one point in time correlates with a second variable measured at a different time point. See the section "Cross-Lag Correlations"

Professor Horvat designs a study to assess the work satisfaction and home-life satisfaction of a group of graduate students. She administers the same measures of work and home-life satisfaction on two occasions, 1 year apart. She finds job satisfaction at the first time point is correlated with job satisfaction at the second time point. Which type of correlation is this?

autocorrelations Each of these is a correlation of a variable with the same variable measured at two different times; thus it is an autocorrelation. See the section "Autocorrelations"

Iva is studying whether students gain weight during their first year of college. She collects weights from all the incoming freshmen in the fall, then collects weights from them again at the end of the spring term. Which is the term for the correlation she finds when she compares the fall and spring weights of the students?

autocorrelations This is a correlation of one variable with itself, measured on two different occasions, or an autocorrelation. See the section "Autocorrelations"

Which type of research design involves measuring the same variables, for the same people, across different points in time?

longitudinal In longitudinal designs, the same variables are studied in the same people across time. See the section "Establishing Temporal Precedence with Longitudinal Designs"

Dr. Samuels does a study that finds children with unusual names are more likely to have delinquency records as adolescents because they were teased more, and the teasing makes them act out. In the relationship between having an unusual name and adolescent delinquency, Dr. Samuels found that teasing plays which of the following roles?

mediator A mediator (in this scenario, teasing) explains why there is a relationship between two other variables (having an unusual name and adolescent delinquency). See the section "Mediation"

When a third variable explains the relationship between two other variables, that kind of third variable is called a _____.

mediator Mediators address the question of why two other variables are related. See the section "Mediators vs. Moderators"

Dr. Aarons does a study that finds having a cognitively demanding job is associated with cognitive benefits in later years, but only among men, not among women. In her findings, in the relationship between having a cognitively demanding job and cognitive benefits later in life, sex plays which of the following roles?

moderator Moderating variables (sex, in this scenario) provide information about when or under what conditions two other variables (cognitively demanding jobs and cognitive benefits later in life) are related. See the section "Mediators vs. Moderators"

When a relationship between two variables depends on the level of a third variable, that kind of a third variable is called a _____.

moderator Moderating variables provide information about when or under what conditions two other variables are related. See the section "Mediators vs. Moderators"

Which type of study design is used when researchers investigate causality by using a variety of correlational studies that all point in a single, causal direction?

pattern and parsimony In cases where researchers cannot establish causality by running an experiment, they can find support for a causal relationship by looking for pattern and parsimony in a variety of correlational studies. See the section "Getting at Causality with Pattern and Parsimony"

Which of the answer pairs correctly completes the following sentence: Although we cannot establish causation with correlational designs, longitudinal designs address concerns of _____ and multiple regression addresses concerns of ____?

temporal precedence; internal validity Longitudinal designs allow researchers to establish temporal precedence by measuring the same variables in the same people across time points. Multiple regression addresses internal validity concerns by controlling for some potential confounding third variables. See the section "Multivariate Correlational Research"

Which of the following phrases in a popular media article suggests that multiple regression was used?

"controlled for" The terminology "controlled for" or "taking into account" suggests that multiple regression was used. See the section "Regression in Popular Media Articles"

control for

Holding a potential third variable at a constant level (statistically or experimentally) while investigating the association between two other variables

cross-lag correlation

In a longitudinal design, a correlation between an earlier measure of one variable and a later measure of another variable.

cross-sectional correlation

In a longitudinal design, a correlation between two variables that are measured at the same time.

autocorrelation

In a longitudinal design, the correlation of one variable with itself, measured at two different times

June is interested in looking at whether social media use predicts feelings of loneliness and found there was a positive relationship between the two variables. However, she finds that after controlling for the amount of time a person spends with friends per week, social media use was no longer associated with feelings of loneliness. Which of the following is the correct interpretation of this finding?

The amount of time spent with friends is a third-variable problem. This result means that social media use and feelings of loneliness were only correlated because of the third-variable problem of amount of time spent with friends. See the section "What If Beta Is Not Significant?"

parsimony

The degree to which a theory provides the simplest explanation of some phenomenon. In the context of investigating a claim, the simplest explanation of a pattern of data; the best explanation that requires making the fewest exceptions or qualifications.

criterion variable

The variable in a multiple-regression analysis that the researchers are most interested in understanding or predicting. Also called dependent variable

Professor Horvat designs a study to assess the work satisfaction and home-life satisfaction of a group of graduate students. She administers the same measures of work and home-life satisfaction on two occasions, 1 year apart. She finds home-life satisfaction measured at the first time point is strongly correlated with job satisfaction at the second time point, but there is no correlation between job satisfaction measured at the first time point and home life satisfaction measured at the second time point. Which of the following explains why Dr. Horvat can't conclude that home-life satisfaction causes job satisfaction?

There are potential third variables that might explain the relationship. Longitudinal studies conducted in this way do not help to rule out possible third variables. See the section "Longitudinal Studies and the Three Criteria for Causation"

A multiple-regression analysis is run to examine third variables as a follow-up to the study indicating that boys with unusual names are more likely to commit crimes. Unusual names, ethnic backgrounds, and geographical regions are all included as predictor variables. Even after controlling for all these variables, unusual names still predict criminal activity. Which of the following explains why this study doesn't establish causation?

There may be some other third variable. Multiple regression controls for any third variable the researcher measures in the study, but it can't control for other potential third variables not included in the study. See the section "Regression Does Not Establish Causation"

Which of the following is true about the difference between beta and r?

Unlike r, beta reflects the independent contribution of the predictor variable, controlling for the contributions from the other predictor variables. Beta reflects the relationship between one predictor variable and the criterion variable controlling for the other predictor variables. See the section "Beta Basics"

Multiple-regression designs help address internal validity in which of the following ways?

by ruling out third variables By measuring possible third variables and using multiple-regression analysis, these third variables can be eliminated as explanations for the relationship between the key variables. See the section "Regression Results Indicate If a Third Variable Affects the Relationship"

Dr. Stevens did a study that found having a cognitively demanding job is associated with cognitive benefits in later years because people who are highly educated take cognitively demanding jobs, and people who are highly educated have better cognitive skills. In the relationship between having demanding job and cognitive benefits in later years, she found that the amount of education plays which of the following roles?

confounding third variable In a third-variable explanation, the third variable (in this scenario, amount of education) explains the relationship between the two variables in the original bivariate correlation (having a demanding job and cognitive benefits in later years). See the section "Mediators vs. Third Variables"

Dr. Russell did a study that found praise provided by supervisors is associated with higher levels of work productivity only because more-motivated employees are praised more often, and highly motivated people are more productive. In her findings, employee motivation is a ___________ in the relationship between praise from supervisors and work productivity.

confounding third variable In a third-variable explanation, the third variable (in this scenario, employee motivation) is external to the two variables in the original bivariate correlation (praise from supervisors and work productivity). See the section "Mediators vs. Third Variables"

Onwuegbuzie et al. (1999) examined foreign-language anxiety as a predictor of foreign-language achievement. They performed a multiple-regression analysis and included other predictor variables, including academic achievement, perceived scholastic competence, and perceived self-worth. Which of the following is the criterion variable?

foreign-language achievement Foreign-language achievement is the dependent variable, or the variable the researchers were most interested in understanding. See the section "Criterion Variables and Predictor Variables"

Professor Horvat designs a study to assess the work satisfaction and home-life satisfaction of a group of graduate students. She administers the same measures of work and home-life satisfaction on two occasions, 1 year apart. She finds home-life satisfaction measured at the first time point is strongly correlated with job satisfaction at the second time point, but there is no correlation between job satisfaction measured at the first time point and home life satisfaction measured at the second time point. Which of the three criteria for causation does Professor Horvat's study fulfill?

covariance and temporal precedence Because the path from home satisfaction to job satisfaction is stronger than the path from job satisfaction to home satisfaction, the experimenter is closer to determining which precedes the other (temporal precedence). However, there is no internal validity because it is not an experimental design. See the section "Longitudinal Studies and the Three Criteria for Causation"

When using multiple regression, which is the term for the variable the researcher is most interested in explaining or predicting?

criterion variable Multiple regression is used to study three or more variables and the variable chosen as one that researchers want to understand is called the criterion variable. See the section "Criterion Variables and Predictor Variables"

Professor Horvat designs a study to assess the work satisfaction and home-life satisfaction of a group of graduate students. She administers the same measures of work and home-life satisfaction on two occasions, 1 year apart. She finds home-life satisfaction measured at the first time point is strongly correlated with job satisfaction measured at the second time point. Which type of correlation is this?

cross-lag When one variable measured at time 1 correlates with another variable measured at time 2, there is a cross-lag correlation. See the section "Cross-Lag Correlations"

Professor Horvat designs a study to assess the work satisfaction and home-life satisfaction of a group of graduate students. She administers the same measures of work and home-life satisfaction on two occasions, 1 year apart. She finds at both the first time point and the second time point there is a strong correlation between work satisfaction and home-life satisfaction. Which type of correlations are these?

cross-sectional These two tests are made a year apart, showing the relationship between the factors of job satisfaction and home satisfaction. This is a cross-sectional correlation. See the section "Cross-Sectional Correlations"