We left off from the last post after performing an ANOVA statistical test and showing that there is a difference between treatments. Our next steps are to build a model and then check our model and assumptions we had previously made are correct.

One of my friends asked me what my data science workflow is when I start a new data project. I’ll breakdown my workflow and tools here:

This post is the first of two posts to focus on how to perform an exploratory data analysis (EDA) of the experimental data set, create a hypothesis and perform an analysis of variance (ANOVA) on the hypothesis. The second part will focus on how to build a model and determine if the model is valid. My background is in nanotechnology so this post will focus on a simple experiment where the experimenter would like to test the differences in plasma power on deposition of a thin film on the surface of a silicon wafer. Values are made up and completely my own. All analysis will be performed in Python using the SciPy package and written in Jupyter notebooks.

Hi. This blog is set up to be a collection of discussions on how to use Python for designs of experiments. I’m currently deep diving into this subject matter and will be showing how to use Python with design of experiments dealing with experimental chemistry. In the process I’ll be exploring exploratory data analysis, setting up factorial experiments, statistical analysis of the experiments, visualizations, modeling, and what to do next all in Python! Additionally, I will occasionally add other projects and musing on to this blog.