Home / Topics / Research Design / Theory and Method / Longitudinal / Anonymity Works Best
Joel W. Grube

Anonymity Works Best

Posted on February 14, 2006

Joel W. Grube (bio) discusses the importance of anonymity in research with adolescents, and a technique for maintaining anonymity in longitudinal research.

 

Q: Why is anonymity important when collecting data from adolescents on sensitive issues?
A: When you're studying sensitive behaviors, like drinking, drug use, or sexual experience, it works best if you keep it anonymous. People report high risk behaviors more frequently in studies that are conducted anonymously, compared to studies that use face-to-face interviews.
Q: When conducting longitudinal research with children, how do I link respondents over time, while still maintaining anonymity?
A: That's not easy. One possibility is to use codes created by the participants themselves, using characteristics that don't change over time, such as sex, birthdate, and initial of mother's first name. I wouldn't recommend using race and ethnicity in these codes because they aren't reported reliably. Each participant will have a unique code which they can use each time they complete a survey. The code should have at least seven elements, and self-generated codes work better with smaller populations, as opposed to very large groups.
Q: What are the disadvantages to self-generated codes?
A: Well, the biggest problem is that people can make mistakes when they write in their codes. But you can use a procedure called "off-one" to compensate for coding errors. First, you pair off all the surveys that match on all elements of the code; then, with the surveys that are left, you pair all those that match on all but one element. Off-one procedures can reduce mismatches to less than 2%.

 

 


Based on personal communication with researcher in May 2005 and published article. Faden, V., Day, N., Windle, M., Windle, R., Grube, J., Molina, B., Pelham W., Gnagy, E., Wilson, T., Jackson, K., & Sher, K. (2004). Collecting longitudinal data through childhood, adolescence, and young adulthood: Methodological challenges. Alcoholism: Clinical & Experimental Research, 28(2), 330-340.

Feedback

Please note that the feedback is viewed only by 4researchers staff and is not intended for communication with individual contributors.

 

Use the form below to submit feedback about this article. If you would like a response, please be sure to include your e-mail address.


More About "Longitudinal"

2:45

Analysis in Randomized Trials

3:03

Bipolar Disorder in Young Adults

Show All...

 

More From Joel W. Grube (bio)

1:41

Asking Teens About Sex

Gathering High-Quality Survey Data

Show All...