Split Ticket Voting: Interpreting the Dataset

Refer to the split_ticket.xls file, which you will find here
http://web.mit.edu/polisci/students/bgreen/split_ticket.xls

The article itself (in .pdf format) is here
http://web.mit.edu/polisci/students/bgreen/split_ticket_voting.pdf

The three relevant sheets of the dataset are:

"brg", which is the unsorted data
sheet 1, which has the data sorted by Grofman's ideological sub groupings
sheet 4, which has the data sorted by my own sub groupings.

The variables are as follows:

v3 = congressional vote, Democrat= 0 Republican= 1
v4= presidential vote, Democrat = 0 Republican = 1
v5= split ticket  unified= 0 split =1
v6= number of votes for republican presidential candidate
v7= number of votes for democratic presidential candidate
v8= percentage of the two party vote given to the Democratic candidate
v9= ADA score
group= Grofman ideological grouping; 1= very conservative, 2= conservative, 3= moderate, 4= liberal, 5= very liberal

And on sheet 4:

group 2= my ideological grouping:
1= Democrat wins < 40% of the two party vote
2= 40%<D<45%
3= 45%<D<50%
4= 50%<D<55%
5= 55%<D<60%
6= D>60%

Please ignore the other sheets.

Brendan Green