## Analyzing Data

When we talk about analysing data we usually mean ‘doing statistics’ but it’s always wise to evaluate your findings before you do any formal statistical analysis. For that reason we suggest the following steps when you have some data ready to analyse:

**plot the data**in an appropriate format. If you are comparing sample means then a box plot is best, especially if it shows the 95%CL of the means; for cause and effect (correlation, regression) make a scatter plot; to look at the distribution of data, make a histogram or bar graph**evaluate the pattern**by examining the plot and assessing whether there is a strong or weak effect. Are the differences between means likely to be biologically important? Does the independent (predictor, X-axis) variable have a strong effect on the dependent (response, Y-axis) variable? Is the distribution of data normal?**perform the appropriate statistical analysis**to answer the question of interest. See here for a guide to the kinds of analysis you might want to perform. See below for a simple guide to actually performing those analyses.**interpret your results**in light of step 3 and especially step 2, above. The P-value and/or confidence limits will give you some indication of the confidence you should have in the magnitude of the effects that you could see when you evaluated the pattern (step 2). When P≤0.05 we usually say that the results are statistically significant, and thus that we have reasonably good confidence that the size of the effect that we see is reasonably accurate. This does not mean that the effect is biologically (or economically, or medically) significant, as that is for you to decide.

**Worked examples**

On the pages listed below, we provide simple biological examples that walk you through some of the the steps listed above to give you a better idea as to what we are talking about.

**Some simple statistical tests**

On separate pages we provide some details and resources on how to perform and interpret the results of some simple statistical tests. Click on each procedure or test to go to its page.