Example Four: Issues at Middle Stages of Research

Confidentiality and Other Ethical Dilemmas

It is relatively easy to assert that all research partners should be involved in the data interpretation stage.  Clearly it is important that all be involved, but partnerships are also beginning to reflect on the confidentiality problems that can emerge.  Consider the following:

     A dissertation student in psychology doing collaborative research on depression and anxiety disorder in a very small tribal nation in the Northeastern United States.  The tribe is very concerned about the devastation these psychological problems are having in the community and tribal leaders wanted this research to be done.  The researcher has worked closely with the tribe in designing the research but she is not Indian and she did not grow up on the reservation.  She knows how important it is that the data be collected and interpreted in a culturally appropriate manner.  But, the tribe is small enough that everyone is likely to know what others have written and said in the study.  What to do?

     In the house-to-house study in the Cambodian and Laotian communities, the interviews were carried out by individuals who—if they didn’t know each other—knew friends of friends.  Because the interviewers had intimate knowledge of the culture they were able to interpret subtle nuances of responses and views.  But, confidentiality was more difficult to maintain because these same interviewers knew the community.  What to do?

     In that same study, families wanted to participate together.  They didn’t want a single person in the family to represent their views.  They talked about their responses together.  So, confidentiality was not maintained within the family in the way the data were collected.  But, this was also the way that the families wanted to provide the data.  What to do?

     Again, in that same study, people were often insulted when they were asked not to the survey and that the survey would be anonymous.  People ask whether they were seen as not good enough to be in the study or to have their names attached to the study results.  What to do?

Now, not all of the above points have to do with confidentiality as it arises in interpretation.  But all of these examples speak to this general concern.  Research partnerships (as opposed to studies carried out just be researchers) are likely to raise more issues about confidentiality at all stages of the research cycle.  Many partnerships are addressing this issue very directly.  

     They are talking together about best ways to ensure confidentiality with insulting participants

     They are discussing together the dilemma of information being inadvertently shared throughout the community.

     They are developing protocols for the handling of sensitive information.

     They are talking with other partnerships about the strategies that they are using and whether those strategies have worked.