Scribd
Upload
5 views

Problem Set 1

The document presents a problem set for an Econometrics course, focusing on three different models to analyze the effects of various factors on outcomes such as GPA, test scores, and teaching evaluations. Each model includes questions about potential unobservable factors, their correlation with independent variables, and the appropriateness of using simple linear regression for causal inference. The data sources mentioned include a university dataset, California's STAR dataset, and course evaluations from the University of Texas at Austin.

Uploaded by

agueyesall
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
5 views

Problem Set 1

The document presents a problem set for an Econometrics course, focusing on three different models to analyze the effects of various factors on outcomes such as GPA, test scores, and teaching evaluations. Each model includes questions about potential unobservable factors, their correlation with independent variables, and the appropriateness of using simple linear regression for causal inference. The data sources mentioned include a university dataset, California's STAR dataset, and course evaluations from the University of Texas at Austin.

Uploaded by

agueyesall
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Econometrics I

Problem set 1

1. Consider a simple model to estimate the e¤ect of personal computer (PC)


ownership on college grade point average (GPA) for graduating seniors at
a large public university:

GPAi = 0 + 1 PCi + ui , i = 1; :::; n;

where PCi takes the value 1 if individual i owns a PC, 0 otherwise, and
ui is the model error (it captures unobservable components). Answer the
following questions:

a. Which type of factors might be contained in ui ?


b. If the collected data are observational, might these factors be corre-
lated with PCi ?
c. Depending on your previous answers, do you consider the simple
linear regression model to be an appropriate tool to uncover the mean
causal e¤ect of personal computer ownership on college performance?

2. The California Standardized Testing and Reporting (STAR)) dataset con-


tains data on test performance, school characteristics and student demo-
graphic backgrounds. The data used here are from all 420 K-6 and K-8
districts in California with data available for 1998 and 1999 (so the sample
size is 420, that is n = 420). Test scores are the average of the reading and
math scores on the Stanford 9 standardized test administered to 5th grade
students. The student-teacher ratio used here is the number of students
divided by the number of full-time equivalent teachers in the district. All
of these data were obtained from the California Department of Educa-
tion (www.cde.ca.gov). Speci…cally, our data set contains the following
variables:

testscri : average test score (= (read_scr+math_scr)/2 ) in district i;


stri : student-teacher ratio in district i.

We are primarily interested in analyzing the relationship between the


student-teacher ratio and the average test score. In principle, we sus-
pect that, other things being equal, a higher student-teacher ratio should
worsen the academic performance of students, but, of course, we need to
assess this question carefully. In order to analyze this issue, we propose
the following model

testscri = 0 + 1 stri + ui ; i = 1; :::; n:

Answer the following questions:

1
a. Which type of factors might be contained in ui ?
b. Might these factors be correlated with stri ?
c. For a given constant value a, calculate

E ( testscri j stri = a) E ( testscri j stri = a 1)

d. Depending on your previous answers, do you consider the simple


linear regression model to be an appropriate tool to uncover the mean
causal e¤ect of the student-teacher ratio on students’performance?

3. The purpose of this question is to analyze the e¤ect of the instructor phys-
ical appearance on the teaching evaluations he or she receives. For this
purpose, we will employ data used by Professor Daniel Hamermesh of the
University of Texas at Austin, which were used in his paper with Amy
Parker “Beauty in the Classroom: Instructors’Pulchritude and Putative
Pedagogical Productivity,”Economics of Education Review, August 2005,
Vol. 24, No. 4, pp. 369-376. These data corresponds to course evalua-
tions, course characteristics, and professor characteristics for 463 courses
for the academic years 2000-2002 at the University of Texas at Austin.
We propose the model

course_evali = 0 + 1 beautyi + ui ; i = 1; :::; n;

where:

course_evali : course teaching evaluation score for each course, on a scale


of 1 (very unsatisfactory) to 5 (excellent).
beautyi : variable representing a rating of course i’s instructor’s physical
appearance. This rating was made by a panel of six students, averaged
across the six panelists and shifted to have sample mean zero (the highest
value this variable takes in the sample is 1.97, whereas the lowest, -1.45).

Answer the following questions:

a. Which type of factors might be contained in ui ?


b. Might these factors be correlated with beautyi ?
c. In this model and with the available data it is assumed that E (ui jbeautyi ) =
K (where K is a constant). Do you think this is a sensible assumption
in the present framework? Justify your answer.

You might also like