Going-Greek/1037select.do at main · HigherEdData/Going-Greek · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
*** do 1037select

local do "1037"
local tag "$tag/`do'"

 ** Degree of selectivitiy

    local oldvar "degsel"
    local newvar "Ndegsel"
    local varlab "Degree of selectivitiy"
    global addvars "$addvars `newvar'"

    clonevar `newvar' = `oldvar'
    lab var `newvar' "`varlab'"
    note `newvar': [`oldvar'] Degree of selectivitiy based on Carnegie ///
         Classifications. 1 = Inclusive, 2 = Selective, 3 = More selective. ///
         57 cases have missing values. Laura has access to finely grained ///
         selectivity measures that we will need to use--Barron's Index--but ///
         it is restricted access. | `tag'
    tab `oldvar' `newvar', m

    note `newvar': There is the issue of a better measure for selectivity. ///
         IPEDS only has the three-category Carnegie measure. Barron’s is better. ///
         But for some unknown reason it is restricted access NCES. ///
         http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2010331. I have used ///
         the 1992 and 2004 Barron’s and have them in my NCES lab now, but we ///
         would want 2008. Problem is, if it is restricted data---even though it ///
         is not individually identifiable, apparently analyses with it need ///
         to be run in NCES approved labs. | `tag'

    note `newvar': The difference between Carnegie and Barron’s ///
         competitiveness index is relatively slight. Barron’s is a tiny bit more ///
         detailed. Instead of three categories it has four: 1=most competitive, ///
         2=highly competitive plus, 3=highly competitive, 4=very competitive ///
         plus. The difference is at the top end, disaggregating really elite ///
         schools from those that are pretty selective, but not incredibly ///
         selective. I am ok using the Carnegie measure, as we have other ///
         indicators that get to the issue of selectivity (like acceptance rate and SAT). | `tag'

    note `newvar': The good news is that most everyone just uses the Carnegie ///
         selectivity measure that we already have. So that settles that. If we ///
         were to drop a measure of selectivity, it would be SAT score. But right ///
         now I think people will want both. | `tag'