puzzled over which test to use

jude

New Member
#1
I have a number of hypotheses relating to temporal trends in wading bird diversity and abundance that need testing. Am I right in thinking that when you have 2 discrete variables and you think there is a causal relationship between them you use a regression analysis?

Also, I believe I have one hypothesis with one discrete variable (abundance of lagoon birds) and one continuous variable (the height of the tide in metres). Can I use regression for this too if I think one influences/causes the other? I thought not, but I'm not sure what else to use..

Is there a reason no one is answering this thread? Does it have to be a yes or no answer?
 
Last edited:

JohnM

TS Contributor
#2
jude said:
I have a number of hypotheses relating to temporal trends in wading bird diversity and abundance that need testing. Am I right in thinking that when you have 2 discrete variables and you think there is a causal relationship between them you use a regression analysis?
Here I would use contingency tables / chi-squared test.

juse said:
Also, I believe I have one hypothesis with one discrete variable (abundance of lagoon birds) and one continuous variable (the height of the tide in metres). Can I use regression for this too if I think one influences/causes the other? I thought not, but I'm not sure what else to use..
Here you could use a simple linear regression, but also just graph the relationship first to see what it looks like.