I'm hoping to find somebody that can give me some advice with my Statistics.

I carried out a study on birds diet. My results show significant annual differences in the diet composition (based on isotopic signatures of Nitrogen and Carbon). I have 30 random samples for each of 4 years (120 samples in total for each isotope).

I'm trying to explain the differences I found by looking for correlations to environmental data (i.e. temperature and prey availability) and to the reproductive success of the whole population on those years.

However, while I have 30 values of each isotope for each year, I only have one value per year for the reproductive success and prey availability (total number of chicks that survived per year and prey biomass respectively) and 2 temperature values per month (24 per year).

My question is if I should average all isotope values per year and correlate each to the yearly temperature averages and the yearly single values for prey biomass and chick survival, or if there is another test I could use. Since the timeframe is 4 years, I'd only have 4 points to say if data is or not correlated.

Could anyone help me please? :)

PS: I tested for significant differences in temperatures between years and I couldn't find any, which I attribute to low resolution of temperature (0.5ºC) ]]>

I then imported the digitized points to STATA and ran

stset tmonths, fail( St)

streg, d(weibull)

From the results I used:

Haz. Ratio of _cons as lambda and haz.ratio of /ln_p as gamma and got the transition probability for each respective cycle in Excel with the formula =EXP(lambda*timeinmonths^gamma).

I just wonder if this is the right approach as I feel I'm messing something up. ]]>

I am planning a one sample study of 230 patients. I am thinking of using the one sample t-test and the binomial test to perform univariate analysis of the continuous and dichotomous variables, against a dichotomous outcome (death following surgery). I am considering using a binary logistic regression to test significant variables from the univariate analysis. Would it be possible to do this with a one sample study or is there another method which is used?

Thank you. ]]>