Lecture 8D: Multiple Linear Regression Flashcards
What is the primary objective of multiple linear regression analysis?
To determine the relationship between one dependent variable and >1 independent variables
What type of variables can be used as predictors in multiple linear regression?
Continuous (preferred) * Categorical (dummy variable)
What is the general regression equation format in multiple linear regression?
Y = b1X1 + b2X2 + b3X3 + … + c
What is a dummy variable?
A variable used to represent categorical data with two or more categories
How would you calculate the predicted 6MW distance for a female using the equation Y = -25.3(gender) + 450.3?
If only two categories (Gender: Male = 0, Female = 1), then Y = -25.3(1) + 450.3 = 425m
What is the significance of using multiple dummy variables?
To represent categorical variables with more than two categories
What is the main method of multiple linear regression covered in this course?
The ‘Enter’ method of forcing a set of predictor variables into the
regression equation
What is the null hypothesis (Ho) when predicting 6MW distance from Berg balance score and quads muscle strength?
Berg balance score and quads muscle strength are not significant predictors of 6MW distance
What is the dependent variable in the example of predicting 6MW distance?
6MW distance (m)
What is the adjusted R square value indicating in the model summary?
It indicates the proportion of variance explained by the predictors, adjusted for the number of predictors
What does the regression equation Y = b1X1 + b2X2 + c represent?
A raw-score regression equation
What is a disadvantage of using the raw-score regression equation?
- Regression coefficients cannot be compared
- Different units for independent variables
What transformation is applied in a standardized regression equation?
Raw scores are transformed to standard scores (z-scores)
What are the advantages of using a standardized regression equation?
- Able to compare beta weights
- Identify which independent variables carry more weight
What indicates that a predictor is non-significant in a regression model?
A high p-value (e.g., p > 0.05)
What is the difference between simple linear regression and multiple linear regression?
- Simple linear regression predicts one continuous DV from 1 IV (1 predictor)
- Multiple linear regression predicts one continuous DV from >1 IVs (>1 predictors)