UnderGround Forums
 

AcademicGround >> need help w/ statistical analysis


5/19/04 3:44 PM
Ignore | Quote | Vote Down | Vote Up
jscorbett
27 The total sum of your votes up and votes down Send Private Message Add Comment To Profile

Edited: 19-May-04
Member Since: 01/01/2001
Posts: 1487
 
I am assisting a student with a project that evaluates differences in photosynthesis rates of genetically altered and standard plants. We need some help evaluating the data for significant differences, since the hypothesis is that the photosynthesis rates are similiar. I will post a link to the paper ASAP which contains the data we need to analyze. thanks
5/19/04 10:01 PM
Ignore | Quote | Vote Down | Vote Up
Slash
19 The total sum of your votes up and votes down Send Private Message Add Comment To Profile

Edited: 19-May-04
Member Since: 01/01/2001
Posts: 2595
inferences can be very advanced. post away
5/20/04 12:11 PM
Ignore | Quote | Vote Down | Vote Up
jscorbett
27 The total sum of your votes up and votes down Send Private Message Add Comment To Profile

Edited: 20-May-04
Member Since: 01/01/2001
Posts: 1489

Here is a link to the paper:

http://www.valdostamac.com/research%20paper.doc

The data is in the results chapter.  Let me know what you think.

 

thanks!!!

 

5/20/04 3:54 PM
Ignore | Quote | Vote Down | Vote Up
Slash
19 The total sum of your votes up and votes down Send Private Message Add Comment To Profile

Edited: 20-May-04
Member Since: 01/01/2001
Posts: 2601
first, i'm not sure which variable would be a good indicator for photosythesis rate (FO, FM, FV/FM). based on the data and available information, i would use a two-way crossed treatment model. compare the contrast of the mean of the photosynthesis rate of each type of plant (normal, genetically eng.) as well as the variance. 6th page and on: http://www.utstat.toronto.edu/craiu/332/Lectures/class5.pdf again, it is complicated and i'm even forgetting the material i learned just last semester. there is really no other way around it, you must use experimental design analysis, which is another level above regression analysis. you can contact a statistical consultant that can run it in SAS (stats program) and he will definately give you precise numbers and inferences. try my prof, radu craiu at craiu@utstat.toronto.edu. i'm not sure if he still does statistical consulting
5/28/04 5:14 AM
Ignore | Quote | Vote Down | Vote Up
RTWu
39 The total sum of your votes up and votes down Send Private Message Add Comment To Profile

Edited: 28-May-04
Member Since: 01/01/2001
Posts: 432
I didn't read the paper at all, just looked at the data. If you are looking for differences between the normal plant and geneticaly altered plant, you should run a 2-way ANOVA with repeated measures and a between group variable. Then run post-hoc analysis if you find significant differences to determine where the differences lie. It is very easy to set up using R or SPSS, I could run it for you in 5 min if you wanted.
5/28/04 8:25 PM
Ignore | Quote | Vote Down | Vote Up
jscorbett
27 The total sum of your votes up and votes down Send Private Message Add Comment To Profile

Edited: 28-May-04
Member Since: 01/01/2001
Posts: 1502
We found a plant researcher who helped us. We ran single factor ANOVA for each measure; FO, FM, FV/FM; and found no significant difference. Thanks for the offers of help! This really isn't my field. I will try to post the finished paper later. thanks again.
8/1/04 9:26 AM
Ignore | Quote | Vote Down | Vote Up
jscorbett
27 The total sum of your votes up and votes down Send Private Message Add Comment To Profile

Edited: 01-Aug-04
Member Since: 01/01/2001
Posts: 1719
The student submitted the project and is a National Finalist for the Agriscience Student of the Year program. I was told his application/paper placed at least 50 points ahead (out of 200 max) of his competitors. He will compete in the interview phase (counts 100 points max) of the competition at the National FFA Convention in late October. Hopefully, he will win. I will post his results afterwards.

Reply Post

You must log in to post a reply. Click here to login.