Hi,
I am aware of the issues and pitfalls of an online behavioral testing - i.e. collecting data in a cognitive experimental (visual search) task throughout the internet. Yet, I decided to do a little test:
I want to assess the same visual search task in a controlled lab setting and using an online platform. I am not planning to do a repeated measures setting.
In the lab setting, around 30 participants will do (based on my previous experiments).
Here's the big question: If I want to validate the online method how many participants do I need online? Shall I go with the same sample size or should I aim for as many respondents as possible?
Thanks in advance!
Andras
Thanks for the answer Ajit.
I would prefer to see some evidence based on previous research on online testing to back up the chosen sample size. E.g. de Leeuw, J. R., & Motz, B. A. (2016). Psychophysics in a Web browser? Comparing response times collected with JavaScript and Psychophysics Toolbox in a visual search task. Behavior Research Methods, 48(1), 1-12.
Hello! I think that in this case the samples should be equal. However, there is a question ... Why were there 30 participants in the first experiment? Sample (participants) size should be calculated on the basis of statistical analysis.
That will depend on the nature of variation within sample.As for an example if there is no variation from sample to sample then a single participant will suffice.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size