More experiments, more data

Well, I finally made it to the point that I’ve started running experiments. Only two 2.5 months. 🙂 But, at least I now have some data. Which, of course, means I now have to figure out how to analyze it.

Every time I finish analyzing an experiment, I tell myself that I should really think more about the data analysis next time. So that I don’t have to scramble once I end up with all the data to figure out how to analyze it. But, each time I never seem to follow through.

It’s not that I don’t think about it. I actually spend a lot of time thinking about it. It’s just that my experimental design usually goes through so many changes, that by the time we actually run the experiment, I’m confused. I know what we’re trying to do. I know what we want to test. I just have to figure out how I use the data to do that.

This time, I’m approaching it in a smarter way than I have in the past. Or, at least I think it’s smarter. Instead of looking at the data and trying to find a statistical technique, I’m starting with the questions that we wanted to answer. And then looking for the specific data that we’ve collected that can be used to answer that question. Then, based on the type of data I have, I’m looking at what statistical technique makes sense.

So far, I’ve got a one-way ANOVA to answer one question and a T-Test and two-way ANOVA for the second. Of course, once I start examining my data more carefully, I’ll see if these actually make sense. But, the good thing is, is that I actually have a plan of attack. I know what I’m going to try to do, and I can see what happens. And it actually makes sense to me (which is more than I can say a lot of the time).

In fact, I’m actually looking forward to having all the data and seeing if this analysis works out. It’s going to require me to “massage” the data a bit to get it into the correct form. But, once that’s done, the actual tests should be easy. This might be the first time I’m not completely terrified by the analysis step. I just hope it actually works out how I’m thinking it should.


Join the discussion

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s