EMET2007 Lecture 9 for Wattle

51 per hour less than men ie the dierence between the

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: olding other factors constant, women are expected to earn $2.51 per hour less than men, i.e. the di¤erence between the mean wage of men and that of women is $2.51. Discussion It can easily be tested whether di¤erence in means is signi…cant The wage di¤erence between men and women is larger if no other things are controlled for; i.e. part of the di¤erence is due to di¤erences in education, experience and tenure between men and women Lecture 9 (qualitative variables) EMET2007/6007 8 st May 2013 14 / 53 Further example: E¤ects of training grants on hours of training \ hrsemp = 46.67 +26.25 grant (43.41 ) (5.59 ) 0.98 log (sales ) (3.54 ) 6.07 log (employ ) (3.88 ) n = 105 R 2 = 0.237 hrsemp = Hours training per employee grant = Dummy indicating whether …rm received training grant This is an example of program evaluation Treatment group (= grant receivers) vs. control group (= no grant) tgrant = 4.70 so we can conclude that giving grants does e¤ect the level of training provided Having a grant increases the expected hours of training by 26.25 hours Is the e¤ect of treatment on the outcome of interest causal ? Lecture 9 (qualitative variables) EMET2007/6007 8 st May 2013 15 / 53 E¤ects of training grants on hours of training (cont) \ hrsemp = 46.67 +26.25 grant (43.41 ) (5.59 ) 0.98 log (sales ) (3.54 ) 6.07 log (employ ) (3.88 ) n = 105 R 2 = 0.237 Having a grant increases the expected hours of training by 26.25 hours Is the e¤ect of treatment on the outcome of interest causal ? employ is the number of employees The coe¢ cient on log (employ ) implies that a 10% increase (+0.1) in the number of employees will lead to a decrease in expected number of hours training workers of 0.61 hours Lecture 9 (qualitative variables) EMET2007/6007 8 st May 2013 16 / 53 Using dummy explanatory variables in equations for log(y) c log (price ) = 1.35 + 0.168 log (lotsize ) + 0.707 log (sqrft ) (0.65 ) (0.093 ) + 0.027 bdrms + 0.054 colonial (0.029 ) n = 88, colonial (0.038 ) (0.045 ) R 2 = 0.649 Dummy indicating whether house is of colonial style Lecture 9 (qualitative variables) EMET2007/6007 8 st May 2013 17 / 53 As the dummy for colonial style changes from 0 to...
View Full Document

This note was uploaded on 06/15/2013 for the course EMET 2007 taught by Professor Strachan during the Two '13 term at Australian National University.

Ask a homework question - tutors are online