This preview shows page 1. Sign up to view the full content.
ECO 544
Assignment 3
1. Column 5 and column 6 indicate the regression results including explanatory variables. In column 5
the coefficient of Spending per enrolled student is 0.072, which means that if spending per pupil
increases by 1 unit ($1,000), NAEP average score increases by 7.2% of a standard deviation, and this
estimate is statistically significant at the 95% level. In column 6 the coefficient of Pupilteacher ratio is
0.03 and statistically significant. This means that if pupilteacher ratio increases by 1, NAEP average
score decrease by 3% of a standard deviation.
2.
(1) Hawthorne effects mean that the “treatment” group works in some way to prove the treatment
effective: students in small classes responded to the fact that they were part of an experiment, rather
than a true causal effect of small classes themselves. And it is John Henry effects that the “control”
group works extra hard to overcome the deficit of being assigned to a small class: the effect sizes
might actually be larger than measured by the STAR experiment because students in regular classes
This is the end of the preview. Sign up
to
access the rest of the document.
This note was uploaded on 02/24/2012 for the course ECON 544 taught by Professor Li during the Fall '10 term at SUNY Buffalo.
 Fall '10
 Li
 Economics

Click to edit the document details