Still having problems...Am I missing something obvious?

#1
This is essentially the same as my earlier post, but i'm still no further ahead. I am in the process of completing all my course work for my degree and I have an experimental proposal to write up. I am investigating the germination of Berberis wilsonii seeds under varied conditions. For this particular part of the work my hypothesis is that there will be a significant association between the length of time the seeds are cold stratified and the germination rate. I have 6 petri-dishes, each with 50 seeds in. one is control, the others are subjected to 2wks, 4wks, 8wks, 12wks and 16wks refrigeration before being germinated...

So, the independant variable here is continuous ratio, the dependant variable discrete count.

I was pointed towards both Pearsons & Spearmans correlations, but as far as I can see, neither will fit my data type.

Any takers??
 

hlsmith

Less is more. Stay pure. Stay poor.
#2
Sounds like a fun experiment. This seems like a basic study design that you could see in agriculture. I personally haven't done basic studies like this before, so won't be a great help. What kind of values did you see in your outcome? This may dictate if you are working with a Poisson or linear regression.
 
#3
I haven't run the experiment yet, its merely a proposal. Based on other Berberis spp. though, I would guess 12wks would produce the highest germination.
 
#4
This is from a similar experiment on another species, only with two treatments:​

Real germination rate

Control


0.0c​

3 months stratification in perlite


0.0​

3 months naked stratification


0.0​

4 months stratification in perlite


73.8​

4 months naked stratification


4.8​

5 months stratification in perlite


75.1​

5 months naked stratification


7.1​
 
#5
Sounds like a fun experiment. This seems like a basic study design that you could see in agriculture. I personally haven't done basic studies like this before, so won't be a great help. What kind of values did you see in your outcome? This may dictate if you are working with a Poisson or linear regression.
Does this data help at all? I would expect my results will not be dissimilar.
 
#6
Going to have to get some sleep now, (11pm GMT here in the uk). Will be keeping my fingers crossed that somebody out there can help me with this!

Thanks in advance people.
 
#7
im going Probit/logistic regression on this one, I don't see any obvious pairing/ correlation in the experiment?
dictate if you are working with a Poisson or
These are probably option as well, especially Poisson, I think may have issues becaue it is bounded on top at 50 as im reeding it?

Good luck, and my the stats gods shine their countenance upon this experiment!
 
#8
im going Probit/logistic regression on this one, I don't see any obvious pairing/ correlation in the experiment?

These are probably option as well, especially Poisson, I think may have issues becaue it is bounded on top at 50 as im reeding it?

Good luck, and my the stats gods shine their countenance upon this experiment!
Nice one fed2, much appreciated. I did wonder about regression, but I haven't had any experience with it, so was scared off. I'm going to run with it though I think. I'm pretty confident with the rest of my paper, so if i lose a couple of marks because I picked the wrong test, i'll just have to live with it I guess.