Wooldridge Source: Michigan Department of Education, www.michigan.gov/mde Data loads lazily.
data('meap01')A data.frame with 1823 observations on 11 variables:
dcode: district code
bcode: building code
math4: percent students satisfactory, 4th grade math
read4: percent students satisfactory, 4th grade reading
lunch: percent students eligible for free or reduced lunch
enroll: school enrollment
expend: total spending, $
exppp: expenditures per pupil: expend/enroll
lenroll: log(enroll)
lexpend: log(expend)
lexppp: log(exppp)
https://www.cengage.com/cgi-wadsworth/course_products_wp.pl?fid=M20b&product_isbn_issn=9781111531041
This is another good data set to compare simple and multiple regression estimates. The expenditure variable (in logs, say) and the poverty measure (lunch) are negatively correlated in this data set. A simple regression of math4 on lexppp gives a negative coefficient. Controlling for lunch makes the spending coefficient positive and significant.
Used in Text: page 18
str(meap01)
#> 'data.frame': 1823 obs. of 11 variables:
#> $ dcode : num 1010 2070 2080 3010 3010 3010 3020 3020 3020 3030 ...
#> $ bcode : int 4937 597 4860 790 1403 4056 922 2864 4851 881 ...
#> $ math4 : num 83.3 90.3 61.9 85.7 77.3 ...
#> $ read4 : num 77.8 82.3 71.4 60 59.1 ...
#> $ lunch : num 40.6 27.1 41.8 12.8 17.1 ...
#> $ enroll : int 468 679 400 251 439 561 442 381 274 326 ...
#> $ expend : num 2747475 1505772 2121871 1211034 1913501 ...
#> $ exppp : num 5871 2218 5305 4825 4359 ...
#> $ lenroll: num 6.15 6.52 5.99 5.53 6.08 ...
#> $ lexpend: num 14.8 14.2 14.6 14 14.5 ...
#> $ lexppp : num 8.68 7.7 8.58 8.48 8.38 ...
#> - attr(*, "time.stamp")= chr "25 Jun 2011 23:03"