TEAM D: POLITICAL ATTITUDES AND ACADEMIC MAJOR

Matt Vela
Movses Musaelian

GENERAL COMMENTS:

* Unfortunate about the lack of bcc on the origianl email invitations

* Given that your survey tries to relate political attitudes and
  major, it's a shame you didn't prestratify (e.g. by college).  Some
  of what you did [e.g. slide 6] sounds awfully close to a
  pre-stratification design; it might not have been much more work to
  go ahead and pre-stratify.

* Under-representation of CFA suggests you may need to post-stratify
  to get results that are representative across campus. 

* Note that it is probably not a good idea to intensively sample CFA
  to make up for representation, because you have not designed the
  survey to be a pre-stratified survey.

* It was too bad today [Tue Apr 03] to hear that you lost key
  information on your first 73 respondents.  As we discussed in class,
  imputation is not really an option, and recontacting the 73 initial
  respondents does not respect their privacy or time.  So your sample
  size starts over with the new survey (using closed-response question
  on major).

  I think you should probably keep the survey going a few extra days,
  but then accept that you probably won't get the full sample size you
  aimed for.

* Except for those first 73 respondents, think about other ways to
  handle missing data than just complete case analysis.

I'm giving you both 90's on this presentation.

GRADING SCHEME FOR PRESENTATIONS

Introduction:                                         10/10
   What’s it about?
   Research Question & Motivation
   Questionnaire, or at least a couple 
      sample questions from it

Body:                                                 40/50
   Describe the work you have done so far
   How sample was selected... 
      Any advice for future 303 students?
   How survey was set up and respondents contacted
   Glitches?  How did you fix ‘em?
   What is the nonresponse or refusal rate so far? 
      How will you decrease it?
   What are the distinguishing features of 
      nonresponders/refusers? 
   Successes? Tell us.
 
Conclusion:                                            10/10
   Future work – what’s left to do?
   When will you / did you cut off data collection?
   What post-processing and/or analyses will you 
      perform? (nonresponse, poststratification, 
      answering your research questions, etc.) 
   Anything else?

Individuals:                                           30/30
   Know what you're talking about. 		       (for both)
   Sound practiced, not rushed.
   At least 2 real slides per person

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 un-excused no-shows will get 0. 

Matt Vela
Movses Musaelian