Group I - biking

* It was defintely "worth a try" to do a phone survey.  

* The random digit dialling would have gone better if someone in your
  group had been able to write a program to efficiently select numbers
  (and perhaps dial them for you).

* Nelson's alternative of using a phone book in the same way as the
  C-book was a good idea
  - it would have been good to start early enough so that you could
    have switched to that method in time to get a reasonable # of
    responses 
  - using the phone book would have some undercoverage problesm for
    the pgh residential population (even after you eliminate non-pgh 
    residents) that would not be present in rand digit dialling, 
    or in using C-book for CMU student population (what would the
    main problem be?).  So this would not have been perfect but would
    have given you more data to work with....

* identifying the survey as a bike survey up front caused a lot of
  people to drop out because they don't own a bicycle.  Can you
  estimate the fraction (# of people who agreed to answer
  questions)/(# of people who heard you introduce the survey), and
  calculate its SE?  That should give you a CI for fraction of people
  in the city who own bikes.  Might be interesting to compare that
  with other estimates, speculate about biases that might be present
  in either estimate, etc.

* totally agree about using a CMU exchange phone rather than a cell
  phone. This was after all an official project for an official CMU
  class!

* Clearly the presentation was not well rehearsed; I think this goes
  to deeper organizational issues in your group.

Introduction:                                         10/10

Body:                                                 40/50
 
Conclusion:                                            10/10

Individuals:                                           /30

Andrew Leroux             aleroux@andrew.cmu.edu       15/30    75
Nelson Mangaali		  nwm@andrew.cmu.edu	       15/30	75
Ben Klahr 		  bklahr@andrew.cmu.edu	       15/30	75

Lower individ scores are due to poor organization and practice for
talk.  Otherwise I though you did well and courageously, as I said in
class. 

Each person will get a score which is the sum of the above (so, 70 pts
from group preparation of slides, 30 pts for individual presentation),
except that no-shows will get 0.