Science Buddies: "Ask an Expert"

My student's project involves interviewing people and asking them a question for their science fair, collecting their responses, and analyzing the results.

How many people should be interviewed in order to make the data tables as relevant as possible? 10? 30? 50? 100?
Also, if a student is analyzing data based on age, how many people should be interviewed in each category?

Any help you can offer would be greatly appreciated!

Hi,

There is not a specific number of subjects you need in order to have relevant or valid results. However, I would recommend above 30-40 subjects, if possible. This would account for the outliers. In research, the more subjects, the better!

Also, you do not need to balance each age group. If you try to have a certain number of subjects in each category, you will not have a representative conclusion. This means you should choose your subjects at random, not matter their ages. Do not try to make a perfect balance in each category.

I hope this helps and good luck!

Cheers,
Jon

Hello,

It is true that choosing random subjects will yield the best, most accurate results...However based on your explanation, it seems like you will be asking the same question to different people and organizing the results using age groups...So if you are trying to analyze how different age groups respond to the same question, in other words, seeing how age affects the answer, then you do want to have roughly equal amounts of subjects in age group and as many subjects as possible in each age group to get the best results...and of course within each age group, try to pick random subjects which will yield the most accurate results...
I hope this helps!
If you have any questions/clarifications, please feel free to post!