Exercise 15: Understanding Multiple Linear Regression
1. What are the assumptions for linear regression?
Assumptions for linear regression include:
independent variables are measured with minimal error
variables are treated as interval- or ratio-level measures
residuals are not correlated
dependent variable scores are normally distributed
scores are homoscedastic, equally dispersed about the line of best fit.
y scores have equal variance at each value of x, thus difference scores are random and have
homogeneous variance
2. Was multiple regression analysis the appropriate analysis technique to conduct in the
Frank et al. (2014) study? Provide a rationale for your answer.
Yes, multiple regression was appropriate. The findings were significantly correlated (r>=
0.25, p<0.1). Also, many of the reporting’s meet the criteria for multiple linear regression
1. What are the assumptions for linear regression?
Assumptions for linear regression include:
independent variables are measured with minimal error
variables are treated as interval- or ratio-level measures
residuals are not correlated
dependent variable scores are normally distributed
scores are homoscedastic, equally dispersed about the line of best fit.
y scores have equal variance at each value of x, thus difference scores are random and have
homogeneous variance
2. Was multiple regression analysis the appropriate analysis technique to conduct in the
Frank et al. (2014) study? Provide a rationale for your answer.
Yes, multiple regression was appropriate. The findings were significantly correlated (r>=
0.25, p<0.1). Also, many of the reporting’s meet the criteria for multiple linear regression
