# PNB 3XE3 Study Guide - Quiz Guide: Homoscedasticity, General Linear Model, Multicollinearity

52 views11 pages

Published on 9 Mar 2020

Department

Psychology, Neuroscience & Behaviour

Course

PNB 3XE3

Professor

Chapter 12:

Homework 5

1. We are interested in whether ___ is a good fit to the observed data.

a) our model, for which parameters have been calculated based on our data

b) b0 + b1Xi, where b1 is a parameter describing the relationship between our independent

variable and the outcome or dependent variable

c) b0 + b1X1i + b2X2i + … bnXni

d) most generally, an equation that summarizes the pattern of observed data

e) any of the above answers are possible good answers

2. The following graph shows a violation of which assumption?

a) homoscedasticity

b) no perfect multicollinearity

c) additivity and linearity

d) independent errors

3. What does additivity mean in the additivity and linearity assumption?

a) you can add one model to another

b) you can add equations together

c) you can add outcome variables together

d) if a number of variables affect the outcome, then their combined effect is best described

by adding their effects together

4. With the equation outcomei = (b0 + b1Xi) + errori …

a) the parameters are estimated based on your collected data

b) for each parameter, you can calculate a test statistic

c) for each parameter you can calculate a p-value

d) all of the above

5. The equation outcomei = (b0 +b1Xi) + errori is called

a) Bayes’ theorem

b) the general linear model

c) the equation

d) the hypothesis test

6. In which example would errors be independent?

a) You have people watching a movie and then interview them to get their ratings. After

watching the movie together, they line up to wait until you call them one by one.

b) You are measuring reaction time while observers are alone in your lab. Several students

who enroll are in the same stats class.

c) You are measuring Netflix binging as an effect of household income. A married couple

enrolls in your study.

d) You are interested in people’s ratings of the new psychology curriculum. You increase

your sample size by having enrolled participants bring their friends.

7. Which of the following is false?

a) you use a statistical model to predict the outcome for a given case (i)

b) a model is derived from a specific data set

c) a model is always imperfect

d) a model is always linear

8. In the equation outcomei = (b0 + b1Xi) + errori which is the predictor variable?

a) b1Xi

b) outcomei

c) Xi

d) errori

9. A ___ is a numerical characteristic of a sample and a ___ is a numerical

characteristic of a population.

a) distribution … variable

b) variable … distribution

c) parameter … statistic

d) statistic … parameter

10. In the following figure, each blue dot represents what?

a) the error

b) parameters

c) the level of oxytocin measured in a particular person

d) the residual

11. If the relationship between the predictor variable and the outcome variable can be

best described by a straight line, you have satisfied which of the following?

a) homoscedasticity

b) no perfect multicollinearity

c) linearity

d) predictors are uncorrelated with external variables

12. You measure vocabulary in a group of 2 year olds, a group of 10 year olds and a

group of 18 year olds. You find little variance among the 2 year olds, a bit more

among the 10 year olds, and a great deal of variance among the 18 year olds. This

data set violates which assumption?

a) additivity and linearity

b) independent errors

c) homoscedasticity

d) all of the above

13. The difference between the value of the outcome predicted by the model and the

value of the outcome actually observed is called the…

a) error

b) residual

c) variance

d) A and B only

## Document Summary

After watching the movie together, they line up to wait until you call them one by one: you are measuring reaction time while observers are alone in your lab. Several students who enroll are in the same stats class: you are measuring netflix binging as an effect of household income. A married couple enrolls in your study: you are interested in people"s ratings of the new psychology curriculum. You find little variance among the 2 year olds, a bit more among the 10 year olds, and a great deal of variance among the 18 year olds. A researcher wanted to know exactly what it was about being a professor that created this stress and subsequent burnout. She recruited 75 professors and administered several questionnaires that measured: burnout (high score = burnt out), perceived. The outcome of interest was burnout, and cooper"s (1988) model of stress indicates that perceived control and coping style are important predictors of this variable.