Lab 7: Organ Donation

Background: Organ donation saves lives, and surveys show that many Americans approve of organ donation, but recruiting organ donors is difficult. Some people prefer not to make an active decision about organ donation becaue the topic can be unpleasant, but perhaps wording the question differently could affect people's willingness to become a donor. Johnson and Goldstein (2003) describe a study where researchers wrote an online survey that asked people to imagine they have moved to a new state and are applying for a driver’s license.  As part of this application, they were to decide whether to become an organ donor (the response variable).  Some people were randomly assigned to be told one of three default options (the explanatory variable):

  • “Opt in”: Subjects were told that the default option in the state was not to be a donor (these people would then need to “opt in” to the organ donation program if they wanted to participate),
  • “Opt out”: Subjects were told the default option in the state was to be a donor (these people would need to then “opt out” if they did not want to participate in organ donation),
  • “Neutral”: There was no prior default option, the group was just asked whether they would like to be an organ donor (the “neutral” group).

Researchers counted the number of individuals in each group that said they would be organ donors. In fact, many states/countries have recently changed their recruitment process based on lessons learned in this and similar studies.

Goals:  In this lab, you will explore

  • Different statistics for measuring the association between two categorical variables in a two-way table.
  • The “shuffle to shuffle” variation in these statistics under random shuffling.
  • The chi-square distribution as an approximation to the null distribution of the chi-square test statistic.

Word file: lab7.doc

Data file: donor.txt

When you are ready, click Start to begin.

Start Lab