∙
Consider the job training grant application. Using
clscrap
, the change
in the log of the scrap rate, we estimated the mean policy effect to be
about
−
.374. Crudely, this means an average fall in the scrap rate of
about 37.4% – a very large economic effect.
97

This
** preview**
has intentionally

**sections.**

*blurred***to view the full version.**

*Sign up*
∙
We also found that, even with a small number of observations, we can
soundly reject
H
0
:
≥
0 against
H
1
:
0 at the 1% significance
level.
∙
But suppose the estimated sample average had been
−
.040 with
n
1,600 and
S
.637 (the same estimate obtained with
n
19). This
means a standard error of .637/40
≈
.016. The
t
statistic is then
t
−
.040/.016
−
2.50, which is still a very strong rejection of
H
0
.
But the estimated policy effect is much smaller – roughly 4% rather
than 37%.
98