David Hufnagel
Economic Forcasting
October 31, 2007
The assignment
Beauty and the Regressions
used data collected from a study
comparing characteristics of instructors and the resulting effect on test scores.
The
study of Teacher Ratings used showed multiple variables, including variables prepared
from; gender, physical appearance, age, minority status, number of credits the course
consists of, native language, and the score given from an individual teaching evaluation.
The paper will be concerned with the following variables explicitly for the assignment;
gender, a teaching evaluation score, and a rating of the instructor by attractiveness.
The paper will examine the effects of beauty (attractiveness) and gender on the variable
score.
Before gender is also considered, a regression analysis of the variables score
and beauty is considered.
Again, the variable score is the score earned on a teaching
evaluation with results scaled from onetofive, five being excellent.
The variable beauty
is a rating, with mean set to zero, of the instructor’s physical appearance.
Regression of Variables Score and Beauty
==================================================================
RESPONSE: SCORE
Mean: 3.9983
NOBS: 463
DF: 461

Variable
Coefficient
Std.Err.
TStat
90.0 % Conf.Interval
Intercept
3.9983
.0253
157.7272
(3.9565, 4.0401)
BEAUTY
.1330
.0322
4.1334
(.0800, .1860)

Std.error of the regression: 0.5455
Sum of squared residuals
: 137.1556
Rsquared: 0.0357
Rbarsq.: 0.0336
F(1,461): 17.0847
Pvalue: 0.0000
By looking at the coefficient of the variable beauty, the regression shows a
positive relation between the variables score and beauty.
From this data we can
conclude that when the teacher evaluation score raises one unit, the value of the
variable beauty raises by .1330. The mean value of the variable beauty is averaged to
zero, while the values of variable score is a number between one and five. The scales
between the two are not standard units, as the case with a measurement like income.
The variation of beauty, approximately .1330 when the variable score raises by one unit,
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
This is the end of the preview.
Sign up
to
access the rest of the document.
 Spring '08
 Various
 Statistics, Regression Analysis, Standard Deviation, nobs, Variable Coefficient Std.Err

Click to edit the document details