Team 2-B		97/100

BUTLER, EMILY LYNN        Senior HSS STA  L     9.0  elbutler
MAKRIS, CHRISTOPHER PETER Senior HSS STA  L     9.0  cmakris  
TODD, THOMAS RANDAL       Senior HSS STA  L     9.0  trt      
HAYES, MEGGIE ROCHELLE    Senior HSS HSS  L     9.0  mrhayes 
HECKMANN, DOUGLAS R       Soph   HSS STA  L     9.0  dheckman

OVERALL:  You've done a good job tightening up both topics.  I'd
prefer you to do the first one (alcoholic energy drinks).

1. College students' attitudes toward alcoholic energy drinks.

   A. Interest.	       10/10

      The question is shifting in a direction that is more interesting.


   B. Research Questions.	10/10

      You've raised three different kinds of research question here:
        - student attitudes
	- student habits/use
        - effects of media on each of the above

      Please think carefully about constructing survey questions that
      get at each of these areas.  Note that it may be necessary to
      limit or eliminate one of the three areas to have a survey that
      is not too long for people to fill out. 
      
   C. References.    5/5

      Fine.


   D. Sampling Frame		8/8

      This is a good initial description of a sampling plan that does
      not involve an explicit sampling frame. 

      The idea of combining "man on the street" interviews with
      stratification is a good one.  The stratification will set
      sample size targets for certain kinds of individuals, that you
      will have to try to meet.

      It might be a good idea to do some sampling experiments to be
      sure that the times and locations you choose will give you the
      kinds of students you seek.  This will also give you better
      insight about coverag of the target population with your
      sampling methods.

      In these experiments you can also practice selecting
      interviewees: e.g. every third person (no exceptions!) and then
      on what criteria would you *not* accept the person into your
      sample?

      Etc.
      

   E. Target Population		7/8

      I think "all undergrads at public and private schools" is too
      broad a population, and not one for which you can say much about
      coverage by your sample here in pgh.  Better to just have the
      target population(s) be the undergrad student bodies of Pitt and
      CMU. 

      As we will see in class, it probably is not necessary to sample
      in proportion to the two school's sizes.  However sampling im
      proportion to strata within schools will make things simpler.

   F. Mode of Data Collection	4/5

      Sampling every 3rd person (or a similar rule) is not really
      "random" but it is arbitrary in a good way -- it will tend to
      force you not to incorporate your biases about what kind of
      people you like to approach, into the sampling method.  

      It will be important later in the project to try to check to see
      that the results you get will be "similar" to a random sample.
      For example, do you get similar proportions of various types of
      students, as would be listed in the CMU FactBook (or similar
      Pitt publication)?

   G. Important Variables	4/4

      These are fine.  See also B. above.


2. Students' Reactions to Carnegie Mellon's Fiscal Expenditures



   A. Interest.	       10/10

      This seems to make more sense this time around -- good.

      Some of the research I suggested before:

      (a) the university budget generally, and in particular how
          student tuition, and fees, are actually spent  [you have
          some of this in your references now]

      (b) the financial sources for major CMU campus building /
          landscaping / art initiatives (e.g. are these paid for from
          the same "pot" as tuition, or are they paid for by special
          gifts, donations, and fund drives/capital campaigns?).
          E.g. research several specific cases of recent changes to
          the campus.

      (c) Motivations for alumni giving (e.g. contact someone in the
          alumni relations or donor development offices to get a
          profile of what works to get small doners to donate money,
          what works for large donors to donate money, etc.)  Also,
          what kinds of undergrads go on to become donors?

       will be important for constructing a good questionnaire that
       could produce interesting results for this survey.

   B. Research Questions. 	  10/10

      Would it be interesting to compare students' impressions of the
      source of funds for various visible projects on campus (such as
      those you list) with the actual sources?   I can imagine a two
      stage survey: (a) how do you think these things are funded and
      how does that make you feel (e.g. about becoming a donor to
      CMU); (b) here's how they were actually funded.  does that
      change your feeling about becoming a donor to CMU some day?      

   C. References.   	     10/10

      OK.

   D. Sampling Frame		8/8

      Cbook. OK.

   E. Target Population		8/8

      good comments.

   F. Mode of Data Collection	4/5

      I am guessing you would use email to invite respondents to a
      website for the actual survey.

   G. Important Variables	4/4

      fine.  See also B. above.