/-/https/www.sciencebuddies.org/cdn/Files/21464/5/sb-logo.png){kind=logo}
/-/https/www.sciencebuddies.org/cdn/Files/21464/5/sb-logo.png){kind=logo}
Sample Size
Moderators: AmyCowen, kgudger, bfinio, MadelineB, Moderators
-
jeloki29
- Posts: 4
- Joined: Mon Nov 08, 2010 7:16 am
- Occupation: Student: 8th grade
- Project Question: How does using a pole to help balance increase the ability to balance?
- Project Due Date: January 18, 2011
- Project Status: I am conducting my research
Sample Size
I'm pretty sure that this question goes into this forum, but anyway I have a question about my sample size. I am doing he Balancing Act experiment and wanted to know if I should just test myself and have someone else help me record my results or should I have test subjects and have myself record the results? If I should have test subjects how many should I have?
-
deleted-71603
- Former Expert
- Posts: 533
- Joined: Wed Jul 25, 2007 7:59 pm
Re: Sample Size
Hi jeloki29.
An interesting topic! Good balance is crucial in so many things we do. I am involved in ballroom dancing, and maintaining awareness of your center of gravity is critical to being a good dancer. I can imagine this applies to several activities.
It is always better to have more than one sample in your experiment. The results of your experiment (the sample outcome) hopefully reflect the true outcome (the population outcome) if you were to test every individual in the world, which, of course, is impossible. The larger the sample size, hopefully the more accurate your results will be compared to the population outcome. There are statistical methods for determining required sample size, but that is well beyond middle school level math. For your experiment, I would recommend a sample size of 3 as a minimum. Recruit your friends and classmates. The larger the sample size the better!
The topic page recommends some variations to the experiment, such as comparing the results with a sample size of more than one (involving your friends / classmates) to the results if it were just you that did the experiment. Are you thinking about doing this? If you recruit your friends / classmates, do you think everyone's results will be the same? If you don't think they will be the same, why not? What will cause the difference? You could tie this question into your initial test question. For example, is there a difference between boys/girls or tall/short people, type of shoe you are wearing (heels, sneakers, barefoot)? If you wish to conduct this type of experiment involving a variable (gender, height, shoe type, etc), then you will have to choose your testers wisely. If you wish to test the difference between boys and girls, for example, then you will want to recruit several of each gender (so, a minimum sample size of 3 now applies to minimum of 3 boys and 3 girls).
I hope this helps. Certainly write back if you have any other questions. If you wish to do an experiment involving a variable, certainly write back with your variable and how you wish to set up your test, if you would like some feedback.
Good luck!
An interesting topic! Good balance is crucial in so many things we do. I am involved in ballroom dancing, and maintaining awareness of your center of gravity is critical to being a good dancer. I can imagine this applies to several activities.
It is always better to have more than one sample in your experiment. The results of your experiment (the sample outcome) hopefully reflect the true outcome (the population outcome) if you were to test every individual in the world, which, of course, is impossible. The larger the sample size, hopefully the more accurate your results will be compared to the population outcome. There are statistical methods for determining required sample size, but that is well beyond middle school level math. For your experiment, I would recommend a sample size of 3 as a minimum. Recruit your friends and classmates. The larger the sample size the better!
The topic page recommends some variations to the experiment, such as comparing the results with a sample size of more than one (involving your friends / classmates) to the results if it were just you that did the experiment. Are you thinking about doing this? If you recruit your friends / classmates, do you think everyone's results will be the same? If you don't think they will be the same, why not? What will cause the difference? You could tie this question into your initial test question. For example, is there a difference between boys/girls or tall/short people, type of shoe you are wearing (heels, sneakers, barefoot)? If you wish to conduct this type of experiment involving a variable (gender, height, shoe type, etc), then you will have to choose your testers wisely. If you wish to test the difference between boys and girls, for example, then you will want to recruit several of each gender (so, a minimum sample size of 3 now applies to minimum of 3 boys and 3 girls).
I hope this helps. Certainly write back if you have any other questions. If you wish to do an experiment involving a variable, certainly write back with your variable and how you wish to set up your test, if you would like some feedback.
Good luck!
Deana
-
deleted-71940
- Former Expert
- Posts: 55
- Joined: Tue Oct 12, 2010 8:25 am
- Occupation: Student: 11th grade expert
- Project Question: n/a
- Project Due Date: n/a
- Project Status: Not applicable
Re: Sample Size
Hello!
This sounds like a great project, and you already have a lot of great advice. I defiantly agree that the larger the sample size the more accurate your results, you will be able to weed out anomalies with a larger testing base. I would suggest asking friends, teachers, and family members to participate in your project.
Good Luck, Hope this helps!
Nithin T
This sounds like a great project, and you already have a lot of great advice. I defiantly agree that the larger the sample size the more accurate your results, you will be able to weed out anomalies with a larger testing base. I would suggest asking friends, teachers, and family members to participate in your project.
Good Luck, Hope this helps!
Nithin T