Hello,
For the purpose of a grant proposal I need to estimate the rate of attrition in a longitudinal study. In a prior wave we collected about 300 families with a child at the age of 5. Now 4 years after we return to these families and propose to collect two more waves. I am trying to find a source to base the estimated attrition on.
Thank you!
The attrition rates vary according to the length of time between cycles and the resources available to track down participants.
I've used the National Longitudinal Survey of Children and Youth (NLSCY), with data collected every two years by Statistics Canada, our national statistical agency, for eight cycles on Canadian children and their parents.
Sample sizes over the eight cycles were:
Cycle 1 22,831
Cycle 2 20,025
...
Cycle 5 9,202
As you can see, from Cycles 1 to 2 (two years later), the attrition rate was 12%, while from Cycles 1 to 5 (eight years later) the attrition rate was 60%. I've never used data beyond Cycle 5 because of this high attrition rate. Also keep in mind that Statistics Canada has more time and resources than you do to track down participants, so for your purposes these attrition rates are minimum values.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Waves