# Linear regression

Six Sigma – iSixSigma › Forums › Old Forums › General › Linear regression

This topic contains 5 replies, has 5 voices, and was last updated by Darth 13 years, 9 months ago.

- AuthorPosts
- January 4, 2006 at 11:54 pm #41895
I was analyzing a linear regression with fish mass on the x-axis and nutrient excretion rate on the y-axis. I was wanting to determine the p-value of this data to see if there is a significant difference between the excretion rate of small fish versus large fish (indicated by the slope of the line). Could someone help me with how to go about this.Thanks.

Jeff0January 5, 2006 at 2:47 pm #131928You seem to be asking two questions……Is there a relationship between fish mass and how much fish poop is produced. Regression is used for that. Then you ask is there a difference between the poop rate of small versus large fish. You would use Hypothesis Testing, specificially a two-sample t or some equivalent to answer that. If you did a scatterplot of the data (fish mass vs poop rate) but distinguished small vs large, you could get a visual picture based on the plot. The best way to do all of this is with some statistical software, particularly Minitab.

0January 5, 2006 at 3:34 pm #131932

Jered HornParticipant@HornJM**Include @HornJM in your post and this person will**

be notified via email.A scatterplot with fist poop…sounds fun.

Isn’t it obvious that the big guys would put out more???0January 5, 2006 at 4:18 pm #131933The options that you have is to first of all establish that there is a correlation between the excretion rate and the fish mass. You can do it by finding the coefficient of correlation between the two. if r is significant then you do a hyp test. You can use excel for these tests. These question has been answered several times on this forum.

https://www.isixsigma.com/forum/showmessage.asp?messageID=83883

https://www.isixsigma.com/forum/showmessage.asp?messageID=82809

https://www.isixsigma.com/forum/showmessage.asp?messageID=82234

0January 5, 2006 at 4:46 pm #131934I agree with Jeff..

Maybe someone is having fun with this discussion.0January 5, 2006 at 5:27 pm #131935Yes, the bigger they are the greater the volume of poop. Look at Stan and Carnell for affirmation of that relationship. But, what is the relationship? Is it linear? Twice as big gives you twice as much? Or is there some other relationship? In my home fishtank this would be useful to predict so that I know how many catfish to put in to keep the tank clean….knowing the amount of FP they can consume.

0 - AuthorPosts

The forum ‘General’ is closed to new topics and replies.