Case 66. False, Fabricated or Misleading Data

Situation

Jim Burke, a graduate student, has been funded for six months as a research assistant from a grant on which Professor Neils Rassmussen is principal investigator. As Burke is preparing slides for a presentation that Rassmussen is scheduled to give in a week, Burke notes very high correlations between many variables. Burke goes back to check the original questionnaires and notices that several questions on which the analyses for the presentation are based have exactly the same responses. Puzzled, he examines other questions and finds that at least 10 percent of all of the questionnaires are strikingly similar, if not identical. Burke shows the questionnaires to a fellow student, Sing Loo, who is working on a different grant with Rassmussen. Loo shrugs and advises, “Keep quiet, it's Rassmussen's problem. Just do your job.” When Burke protests that he can't really do a good job if there are problems with the data set, Loo persists that “your name won't be on it.” Burke debates whether he should talk with Rassmussen about possible fabrication of data by interviewers.

Questions

1.    If Burke suspects that some of the survey data were fabricated by interviewers, what should he do?
2.    What are Rassmussen's responsibilities in this situation?



     Reflect on the above questions and form your own answers before clicking the Discussion key to review the commentary provided with this case.

Discussion

Although Burke's name will not be on the presentation, he should talk with Rassmussen about the possibility that some interviews have been fabricated. Rassmussen has delegated responsibility to Burke as a research assistant to accurately perform tasks to which he has been assigned. Burke might be reluctant to talk with Rassmussen about the issue for fear of appearing to challenge Rassmussen's competence and effectiveness in overseeing a project. As a graduate student, he is dependent on Rassmussen for income and future recommendations and Loo's advice may have led him to believe that Rassmussen would retaliate if threatened. However, Burke contributes to the reporting of fabricated data if he does not bring the matter to Rassmussen's attention.

Even if presentation of false data were not deliberate on Rassmussen's part, he has the responsibility to oversee data collection to prevent fabrication of interviews. This includes training and supervising interviewers or delegating the responsibility for such to competent staff. It also includes use of safeguards such as random call-backs to respondents to ensure that interviewers collect data accurately and responsibly. Rassmussen should evaluate the information Burke provides and make a decision about how to rectify the situation, if he concurs that questionnaires may have been fabricated.