kscrane
Posts: 2
Joined: Thu Feb 28, 2013 12:38 pm
Occupation: Parent of 5th grader

Average for results

Postby kscrane » Thu Feb 28, 2013 2:49 pm

My son is doing the Fruit DNA extraction for his science fair project. He's tested three fruits, 10 times each. Our question is, should he average all 10 results for each fruit, or should he throw out the high and low and average the remaining 8 ? Would this be more accurate? Is there a preferred method? Thanks!

arasson
Former Expert
Posts: 10
Joined: Fri Jan 18, 2013 1:23 pm
Occupation: UCLA PhD student researcher

Re: Average for results

Postby arasson » Thu Feb 28, 2013 3:05 pm

Hi kscrane,

Thats a great project! Unless you think there are outliers for more reason other than natural varience, I think it would be better practice to leave all the data points in your analysis. The only concern I have is about the amount of starting materials. The different fruits have different water weight so even if you are starting with the same weight of each fruit you may not have the same amount of fruit material that you are working with. Is this something that you have taken into consideration? If not, than it will be difficult to confirm that the different amount of DNA extracted was due to the polypoidy of the fruit or if you just started with different amounts of materials.

I hope this helps!

kscrane
Posts: 2
Joined: Thu Feb 28, 2013 12:38 pm
Occupation: Parent of 5th grader

Re: Average for results

Postby kscrane » Sat Mar 02, 2013 4:22 pm

Thank you! We did not take water weight into account, and that may explain his results. The triploid banana just edged out the hexaploid kiwi, and the octoploid strawberry had the smallest average (although they were all relatively close). Do you have to either prove or disprove a hypothesis? Can his conclusion say that it's "inconclusive" and question water weight?


Return to “Grades K-5: Getting Ready for the Science Fair”