My project is who makes a better eye witness, boys or girls. My question relates to the scientific method. You have to do an experiment three times to be valid. But in my experiment I have shown a picture to two different college classes. Then they have answered questions about the picture. There were 60 people in the first class and 46 in the other class. This is 106people. I know I should do at least 100 people to get a 10% margin of error. My question is do I need to do the experiment in one more class for my three times, or do I consider that I have done the experiment 106 times?
Also, I have a higher number of girls than boys, is this a problem and how would I account for the difference?
arh225
Who makes a better eye witness, boys or girls?
Moderators: AmyCowen, kgudger, bfinio, MadelineB, Moderators
-
- Former Expert
- Posts: 404
- Joined: Tue Sep 18, 2007 3:27 pm
- Occupation: Research Assistant
- Project Question: Neuroregeneration
- Project Due Date: N/A
- Project Status: Not applicable
Re: Who makes a better eye witness, boys or girls?
Hi!
Sounds like a really interesting project! More people in your sample is always a better thing, because it makes your results more accurate. Don't get too caught up in the exact number of trials you have to do, just keep in mind that you want to sample as many classes as possible. If possible, you should really try to get equal amounts of girls and boys in your study because that will make comparison more even and fair between the two groups. If not, what you could do is that when you discuss your results, you could mention that there may be some error due to the uneven balance between the size of different gender groups. Hope this helps, good luck!
Sounds like a really interesting project! More people in your sample is always a better thing, because it makes your results more accurate. Don't get too caught up in the exact number of trials you have to do, just keep in mind that you want to sample as many classes as possible. If possible, you should really try to get equal amounts of girls and boys in your study because that will make comparison more even and fair between the two groups. If not, what you could do is that when you discuss your results, you could mention that there may be some error due to the uneven balance between the size of different gender groups. Hope this helps, good luck!
"There is a single light of science, and to brighten it anywhere is to brighten it everywhere." -Isaac Asimov
-
- Former Expert
- Posts: 1019
- Joined: Fri Oct 28, 2005 11:43 am
- Occupation: Research Hydrologist
- Project Question: n/a
- Project Due Date: n/a
- Project Status: Not applicable
Re: Who makes a better eye witness, boys or girls?
Hi arh225,
Sounds like you've done a good job with your project. It is not necessarily a problem to have different numbers of females versus males. For example, you can report your results as percentages, which are independent of the sample size, and you can use statistical tests that account for the sample size. It will depend on what type of data you have collected.
Good luck!
Chris
Sounds like you've done a good job with your project. It is not necessarily a problem to have different numbers of females versus males. For example, you can report your results as percentages, which are independent of the sample size, and you can use statistical tests that account for the sample size. It will depend on what type of data you have collected.
Good luck!
Chris
-
- Former Expert
- Posts: 895
- Joined: Tue Sep 06, 2005 3:59 pm
- Occupation: Professor
- Project Question: How do different animals adapt to their environment?
- Project Due Date: N/A
- Project Status: Not applicable
Re: Who makes a better eye witness, boys or girls?
Hi arh225,
As the other experts pointed out, you have a very interesting and well conducted study! I wanted to chime in to answer your last question. Your major concern was the number of "replicates" you had - the number of times you repeated your experiment with different individuals. Technically, your "replicates" are the individual students you surveyed, but it would not be 106; it would be the number of females in your group (as females were the smaller group).
As Chris noted, you can used percentages (or proportions) to compare the results from your two groups without worrying about the unequal sample sizes.
Good luck!
Heather
As the other experts pointed out, you have a very interesting and well conducted study! I wanted to chime in to answer your last question. Your major concern was the number of "replicates" you had - the number of times you repeated your experiment with different individuals. Technically, your "replicates" are the individual students you surveyed, but it would not be 106; it would be the number of females in your group (as females were the smaller group).
As Chris noted, you can used percentages (or proportions) to compare the results from your two groups without worrying about the unequal sample sizes.
Good luck!
Heather