This preview shows pages 1–3. Sign up to view the full content.
MATH 423 Final Examination December 18, 2007
Student Name:
Student Number:
McGill University
Faculty of Science
FINAL EXAMINATION
MATH 423
Regression and Analysis of Variance
December 18th, 2007
9 a.m.  12 Noon
Calculators are allowed.
One 8.5”
×
11” twosided sheet of notes is allowed.
Language dictionaries are allowed.
You must do BOTH of the ±rst two questions (Questions 1 and 2) and TWO of the remaining
THREE questions (Questions 3 through 5) . You will receive points from the two questions
with the highest marks from Questions 3, 4, and 5.
Question 6 is worth 10 bonus points. You will not lose points on the exam for any work on this
question, you can only add points for any correct work that you provide.
There are 12 pages to this exam. The total number of marks for the exam is 100, although it
is possible score as high as 110 due to the bonus question.
Examiner: Professor Russell Steele
Associate Examiner: Professor David Stephens
1
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Question 1: (30 points)
Average public teacher pay and spending on public schools per pupil in 1985 for 50 states and
the District of Columbia were reported by the Albuquerque Tribune. Output from regression
analyses are reported at the end of the exam and questions about the analysis of the data are
below. The variables in the data set are:
•
PAY: Average public school teacher annual salary (in $)
•
SPEND: Spending on public schools per pupil (in $)
•
AR: Region corresponding to
East
,
Central
, and
West
The primary research question of interest is to determine the extent to which the spending
per pupil and area are associated with the average public school teacher salary. Refer to the
code and output at the end of the test for the following questions.
(a) Test for signiFcance of both covariates individually (not in the presence of the other).
Refer clearly to the part(s) of the output that you are using for your tests.
(b) Give an interpretation for each of the coe±cient estimates in
model4
.
(c) Test for a statistically signiFcant interaction between area and spending for modelling
teacher pay. Refer clearly to the part(s) of the output that you are using for your tests.
(d) Comment on the appropriateness of the model assumptions and the stability of
model4
.
Should the researcher have any concerns about the validity of inferences made from this
model? Why or why not?
(e) Choose what you feel is the “best” of the four models, not taking into account either
observation in²uence or the appropriateness of model assumptions. DeFne clearly what
you mean by “best” and explain how you arrived at your conclusion.
(f) One of the 6 calls to the
This is the end of the preview. Sign up
to
access the rest of the document.
This note was uploaded on 01/15/2010 for the course MATH 423 taught by Professor Steele during the Spring '06 term at McGill.
 Spring '06
 STEELE
 Math, Variance

Click to edit the document details