PSY 215 Chapter 9
true or false? A multiple-regression analysis can control for more than one variable at a time.
true
true or false? The largest beta in a multiple-regression analysis is the predictor that has the strongest relationship with the criterion variable.
true
does this example represent a third variable, moderator, or mediator? inspiration has a stronger influence on creativity in people who work harder
moderator
does this example represent a third variable, moderator, or mediator? stress is associated with negative health outcomes only when the stress is viewed negatively
moderator
In a(n) ?, two variables measured at the same time are associated. If an earlier measure of a variable is associated with a later measure of the same variable, a(n) ? occurs. In a(n) ?, an earlier measure of one variable is associated with a later measure of a different variable. These three types of correlations are usually found in ? designs.
- cross sectional correlation - autocorrelation - cross lag correlation - longitudinal
Multiple regression can help eliminate problems with ? by controlling for potential third variables. It sets ? variable(s) that influence the ? variable, which is the outcome of interest. Effect sizes in multiple-regression analyses can be represented by both ?, which is a standardized effect size, and ?, which is an unstandardized effect size.
- internal validity - predictor - criterion - beta - b
A(n) ? is conducted over a longer period of time than a typical correlational study to help establish ?. A(n) ? considers more than two variables in the same analysis and can help improve ?.
- longitudinal study - temporal precedence - multiple regression analysis - internal validity
Additional variables may often add nuance or an alternative explanation for a relationship. A ? is a variable that explains how the two main variables are related in different ways depending on the situation. A ? explains why two variables may be related based on an internal factor that connects them.
- moderator - mediatior
Indra wants to investigate the relationship between job status (whether or not they have a job) and grades among college students. She goes to the student center and hands out a survey with two questions about job status and GPA. She finds that at her institution, students with jobs have higher grades than students without jobs. Which of the following variables could be alternative explanations for the association between job status and grades?
- time management skills - conscientiousness
true or false? A multiple-regression analysis that accounts for third variables can establish causality.
false
Iva is studying whether eating meals in the cafeteria is linked to weight gain during the first year of college. She asks all the incoming first-year students about their weights and the number of cafeteria meals they eat each week in the fall, and then she collects the weights and number of cafeteria meals per week from them again at the end of the spring term. Identify the type of correlation that Iva should examine for each goal. Iva wants to find out whether weight in the fall is correlated with weight in the spring.
autocorrelation
Timur has collected some longitudinal data on self-esteem and social media use at two time points 1 year apart. Label each of Timur's findings with the type of correlation it represents. social media use at time 1 was positively correlated with social media use at time 2
autocorrelation
determine the correlation of one variable with itself, measured on 2 different occasions
autocorrelations
show whether the earlier measure of one variable is associated with the later measure of the other variable
cross lag correlations
the variable in a multiple regression analysis that the researchers are most interested in understanding or predicting
criterion variable or dependent variable
Timur has collected some longitudinal data on self-esteem and social media use at two time points 1 year apart. Label each of Timur's findings with the type of correlation it represents. social media use at time 1 was negatively correlated with self esteem at time 1
cross sectional correlation
What is the reason that researchers sometimes rely on pattern and parsimony to investigate causality?
it may not be ethical or practical to manipulate certain variables
Can provide evidence for temporal precedence by measuring the same variables in the sample people at several points in time
longitudinal design
longitudinal study or multiple regression? Sammy wants to find out whether parenting style explains the association between self-esteem and behavior problems.
multiple regression
longitudinal study or multiple regression? Sammy wants to find out whether there is an association between self-esteem and behavior problems even after adjusting for experience with bullying.
multiple regression
Mohamad hypothesizes that lack of sleep before the age of 7 results in substance abuse later in life. He realizes that it would be unethical to conduct an experiment depriving children of sleep and that a correlational study could not establish temporal precedence. Mohamad decides to conduct several different studies to establish multiple association claims that point to his hypothesis. What is Mohamad's approach called?
pattern and parsimony
Onwuegbuzie and colleagues (1999) examined foreign-language anxiety as a predictor of foreign-language achievement. They performed a multiple-regression analysis with other predictor variables, including academic achievement, perceived scholastic competence, and perceived self-worth. The researchers found that the results for foreign-language anxiety in the regression table were β = −0.39, 95% CI [−.59, −.19]. Which statement best explains what the beta for foreign-language anxiety means?
people with lower levels of anxiety about learning a foreign language have higher levels of foreign- language achievement, even after the researchers controlled for the other variables
a variable in multiple regression analysis that is used to explain variance in the criterion variable
predictor variable or independent variable
What is a parsimonious explanation?
the simplest explanation
holding a potential 3rd variable at a constant level (statistically or experimentally) while investigating the association between 2 other variables
control for
DeAndre is studying whether social media use predicts feelings of loneliness, and he finds that there is a positive relationship between the two variables. However, he finds after controlling for the amount of time spent with friends that social media use is no longer associated with feelings of loneliness. Identify the correct and incorrect interpretations of DeAndre's finding - The amount of time spent with friends presents a third-variable problem. - The beta for social media use will have a 95% CI containing zero. - The amount of time spent with friends predicts social media use after DeAndre controls for the other variables. - The amount of time spent with friends is not associated with social media use.
correct: - The amount of time spent with friends presents a third-variable problem. - The beta for social media use will have a 95% CI containing zero. incorrect: - The amount of time spent with friends predicts social media use after DeAndre controls for the other variables. - The amount of time spent with friends is not associated with social media use.
Indra wants to investigate the relationship between job status (whether or not they have a job) and grades among college students. She goes to the student center and hands out a survey with two questions about job status and GPA. She finds that at her institution, students with jobs have higher grades than students without jobs. Which of the three causal criteria are established in Indra's study?
covariance
Iva is studying whether eating meals in the cafeteria is linked to weight gain during the first year of college. She asks all the incoming first-year students about their weights and the number of cafeteria meals they eat each week in the fall, and then she collects the weights and number of cafeteria meals per week from them again at the end of the spring term. Identify the type of correlation that Iva should examine for each goal. Iva wants to find out whether weight in the fall is correlated with the number of cafeteria uses in the spring.
cross lag correlation
Timur has collected some longitudinal data on self-esteem and social media use at two time points 1 year apart. Label each of Timur's findings with the type of correlation it represents. social media use at time 1 was positively correlated with self esteem at time 2
cross lag correlation
Iva is studying whether eating meals in the cafeteria is linked to weight gain during the first year of college. She asks all the incoming first-year students about their weights and the number of cafeteria meals they eat each week in the fall, and then she collects the weights and number of cafeteria meals per week from them again at the end of the spring term. Identify the type of correlation that Iva should examine for each goal. Iva wants to find out whether the number of cafeteria uses in the spring is correlated with weight in the spring.
cross sectional correlation
test to see whether 2 variables measured at the same point in time, are correlated
cross sectional correlations
true or false? Multiple regression establishes temporal precedence as long as it controls for potential third-variable problems.
false
Onwuegbuzie and colleagues (1999) examined foreign-language anxiety as a predictor of foreign-language achievement. They performed a multiple-regression analysis with other predictor variables, including academic achievement, perceived scholastic competence, and perceived self-worth. The researchers found β = −0.39, 95% CI [−.59, −.19] for foreign-language anxiety and β = 0.29, 95% CI [.10, .48] for academic achievement. Which statement is the best interpretation of these two results?
foreign- language anxiety has a greater effect than academic achievement on foreign- language achievement
longitudinal study or multiple regression? Sammy wants to find out whether self-esteem and behavior problems have a mutually reinforcing relationship.
longitudinal study
longitudinal study or multiple regression? Sammy wants to find out whether self-esteem in elementary school predicts behavior problems in middle school.
longitudinal study
does this example represent a third variable, moderator, or mediator? a negative parenting style is the reason that marital violence is linked with childhood aggression
mediator
does this example represent a third variable, moderator, or mediator? friendship explains why social media use and happiness are related
mediator
A statistical technique that computes the relationship between a predictor variable and a criterion variable, controlling for other predictor variables
multiple regression or multivariate regression
a study designed to test an association involving more than 2 measured variables
multivariate designs
The degree to which a scientific theory provides the simplest explanation of some phenomenon
parsimony
Dr. Soulis conducts a study that finds that having a cognitively demanding job is associated with cognitive benefits in later years, but only because people who are highly educated have cognitively demanding jobs, and people who are highly educated have better cognitive skills. In the relationship between having a demanding job and cognitive benefits in later years, amount of education plays what role?
third variable
does this example represent a third variable, moderator, or mediator? the correlation between weight and vocabulary is actually a function of age, not a true association
third variable
does this example represent a third variable, moderator, or mediator? warm temperatures cause ice cream sales to be associated with crime rates, though they otherwise would not be associated
third variable