ST445 Class Exercise 


A few years ago, I downloaded some information from Yahoo! Sports,
specifically, the results of the Women's Two-Man Bobsled. (sic?) Each team of two women were timed 
on four runs or heats, and the winner was determined by the fastest times.
The data was first placed in a spreadsheet and then stored as a comma-delimited file (w2bobsled.csv) 
and as a tab-delimited file (w2bobsled.txt).

The order of the variables:
result (char)
county (char)
team members (char)
time for first heat (sec) (DNS=Did not finish)
time for second heat (DSQ=disqualified)
time for third heat
time for fourth heat
total time (hh:mm:ss.s)

The names of the team members are separated only by spaces and this could be a large character field. 
My suggestion is to allow for plenty of room with a LENGTH statement like: length athletes $30 ;

Note that the total time is in a format that's new to you, so I will suggest reading with the informat time10. 
so that SAS will translate '00:02:12.0' to 132 seconds.

0) Decide how you want to handle the teams with incomplete data -- either did not finish (DNS) or disqualified (DSQ).
1) Check the data. Notice the heat times are given to hundredths of a second, while the total time just to tenths. 
   Does it look like the heat times correctly add up to total? Are they rounded or chopped?
2) Find the mean, median, and variance for the four heat times. Which heat was the fastest? slowest?
3) Suppose that the winner was determined not by the fastest total time, but that a team's slowest time could be dropped. 
   Under this new scoring, who would be the top six finishers?
4) For the top six finishers (under either rules) plot the times vs heat in a y*x=z plot with team as 'z' identifier.
   Submit your output and log and answer any questions.