Case Studies 2008

Case Studies in Data Analysis sessions will be held at the Joint SSC-SFdS Meeting in Ottawa May 25-28, 2008. The case studies are intended for teams of graduate and senior undergraduate students working either with or without faculty mentors. (See the Guidelines for Mentors.) Each participating team will choose to analyse one of the two data sets described below and the teams will present summaries of the methods they used and the results of their analyses in a poster presentation session at the meeting.

New this year:
One award will be presented for the best case study poster in each of the two case studies. The Case Studies Awards Committee will consider the quality of both the analysis of the data and the presentation of the results in reaching its decision. The value of the award for each case study in the 2008 competition is $500. This is to be shared equally among the participants of each winning team. The Case Studies Awards Committee reserves the right to decline to make an award for a given case study if numbers of entries are insufficient. For further information, please contact Kevin J. Keen.

All presentations this year will be posters. A subject-matter expert will prepare a poster introducing each case study, so teams presenting their analyses do not have to repeat the background information.

If you would like to participate, contact David Haziza for Case Study 1 and Alison Gibbs for Case Study 2. We will need to know the list of names for your team and the university that you represent. We will also need to know if you are giving any other oral or poster presentations at the meeting so that we can avoid scheduling conflicts. The deadline for contacting David or Alison is March 15, 2008.

You will doubtless have questions once you start working with the data. You can ask us; you don’t have to guess the answer! You can direct your questions to us and we will answer what we can and forward other questions to the data donors. We will put answers to frequently-asked questions on the web pages. You are free to submit questions in either English or French.

Case Studies

Click on the titles below for full information about the case studies.

Case Study 1: Handling Item Nonresponse in Surveys

Organizers:
David Haziza
Département de mathématiques et de statistique
Université de Montréal
Telephone: (514) 343-6705
E-mail: david.haziza@umontreal.ca

Gordon Kuromi
Statistics Canada
E-mail: gordon.kuromi@statcan.ca

Despite the best efforts to maximize response, it is almost certain that some degree of nonresponse will occur in large scale surveys. The purpose of this case study is to find imputation strategies that can reduce the nonresponse bias as much as possible and to consider the problem of variance estimation in the presence of imputed data.

Case Study 2: Effects of Genetic Variation on the Relationship between Diet and Cardiovascular Disease Risk

Organizer:
Alison Gibbs
Department of Statistics
University of Toronto
Telephone: (416) 946-7589
E-mail: alison.gibbs@utoronto.ca

Cardiovascular disease is the most common cause of death in Canada. High levels of LDL cholesterol and low levels of HDL cholesterol are known to be associated with increased risk. In this case study, students will analyse data collected during a recent clinical trial of subjects with type 2 diabetes. The purpose of the case study is to examine the effects of genetic variation in the subjects on the relationship between changes in diet and blood cholesterol level.

Chair of the Case Studies Awards Committee:

Kevin J. Keen, Ph.D., P.Stat.
Department of Mathematics
University of Northern British Columbia
Telelphone: (250) 960-5014
E-mail: keenk@unbc.ca

Guidelines for Mentors:

The SSC Case Studies in Data Analysis exist for participation by teams of undergraduate and/or graduate students. All participating teams, if they desire, may recruit a non-student mentor, for example, a faculty member. In order to have the ability to fairly evaluate the work of the student teams, student teams and non-student mentors are encouraged to consider the following guidelines for their participation.

1. Role of the student team: The approach to the case study should be planned, developed, and executed primarily by the student team. The student team also has responsibility for the organization and presentation of its findings.
2. Role of the mentor: The mentor should encourage and support the student team. This may include listening to ideas, suggesting references, and asking questions to stimulate consideration of additional aspects of the case study. To be fair to all participating teams, the mentor should not take an active role in the development of the statistical analyses.