Data Analysis and Graph Design with Adobe Illustrator

We have discussed quantitative marketing research, designed online/web/Internet surveys with PollDaddy and added the surveys to your tumblr websites, and gathered quantitative marketing research survey data.

We didn’t use the best sampling method for our quantitative marketing research procedure… which means that it’s probably not appropriate to use inferential statistics. But, marketing academics love to talk about statistics and I think it’s important that you understand the link between your PollDaddy survey questions and a statistical test. Statistics involve lots of assumptions and mathematical modeling, and to be good at statistics you have to be good at matrix algebra. We are not going to focus on the matrix algebra in this class.

So, let’s link our survey data to statistical analysis and design cool graphs.

I don’t have the Excel “Analysis ToolPak” Add-on which would be great for this class. We’ll use SPSS in the lab. SPSS is the statistical software I used in grad school and it’s what a lot of marketing academics use. It will be like shooting a small animal with an elephant gun, but just in case you ever get asked by a marketing professor to help with a study, now you’ll be familiar with SPSS.

Most of the statistical analyses you do are probably going to be descriptive statistics–describing what your sample looks like. Let’s use SPSS to calculate central tendency (mean, median), make histograms (distribution), and to do crosstabs (cross tabulation, contingency table) on your data.

Some of your statistical analyses might be inferential statistics–making generalizations about a larger population based on your sample. Let’s use SPSS to do a t-test (examine the difference between the average of two groups/categories), ANOVA (examine the difference between the averages of two or more groups/categories), do a correlation analysis (examine the relationship between two variables measured at the interval/ratio level of measurement), and a regression analysis (examine how multiple independent variables influence one dependent variable, all measured at the interval/ratio level of measurement).

Here’s a diagram that simplifies how to select which statistical test is appropriate to analyze your research model:

*crosstabs can be used to calculate a chi-squared statistic that has a p-value associated with it…

Now that we have analyzed our data, let’s look at how we can make cool graphs with Adobe Illustrator. Adobe Illustrator is what most of the pros use to design graphs. If you are a die hard Microsoft Excel or Apple iWork Numbers fan, then it’s fine if you spend time in the lab next week designing cool graphs with Excel or with Numbers, but you will have to find your own tutorials.

In order to graph your data you first have to think about which type of Adobe Illustrator graph is appropriate to summarize your data. For example, line graphs show trends over time, scatter plots show the relationship between two variables, pie charts show proportions, and bar charts show values across categories.

Here is a list of Adobe Illustrator graph tutorials that you can follow depending on what type of graph you want to design.