This is more than a little can of worms, so it is hard to know where to start. It sounds like you identified all butterflies at one location for 6 years. To keep it simpler, you used one sampling method and I will assume that it was not quantitative like line-transect sampling.

The first approach is biodiversity analysis. Given your data, how many species are expected in this area? A program like Estimates will help. http://viceroy.eeb.uconn.edu/estimates/

You can then detail any special finds, rare or endangered species, location records, and that sort of thing. Make a table of first, last, peak emergence dates for each species. You could also sort the species by host plant or feeding guild.

That is a minimum. The rest depends on what else you can bring to the table. Do you have local temperature measures? Do you have measures at one time point per year or every day in the year? Are you trying to turn a pile of numbers into a story, or did you have some initial questions that need to be answered?

With temperature measurements you could build degree day models. The emergence for each year aligns better using degree days as opposed to date. The alternative is to use a species with a narrow emergence date as a benchmark and adjust accordingly.

Without some other question there isn't much further that you can take this. You could look "before and after" some environmental disturbance event (flood, fire, logging come to mind). Six years is a long time, but not enough to consider looking for a climate change effect. Is a yearly comparison meaningful?

Dear Megan Evans ,

Timothy A Ebert is right. You need to start with a research question. I'm not so familiar with your context so I'll not comment on your study design apart from saying that it is very important to determine whether you are randomly sampling or if you data is time series dependent.

I have written a really simple guide to quantitative data analysis which you might like to read.

For learning statistics I would recommend the Statstutor repository (www.statstutor.ac.uk) and the CAST ebooks from Massey University. The former is bite sized. The latter uses java applets to demonstrate statistical concepts with simulations. And both are good quality. The worst thing to do is to Google for statistics videos on YouTube and most textbooks are not great for learning the subject from scratch.

you can use SPSS, STATA or ANOVA, Chi square