Editing Data Into Insights (Spring 2021)/Story Time

== The goal ==

For this assignment, you will be telling a data story - but without any real data!

== 1. Brainstorm ==

I want you to imagine that you work for your favorite brand. This could be a website, a clothing store, a pizza chain, etc. Think about the kind of data that this business might have and the sort of insights that you might get from it.

Make up an insight that you might get from this type of data, and imagine that you have been asked to give a presentation to the CEO and board about what you think the insight means and what the company should do about it.

For example, maybe you love Doritos and you start with the (completely made up) insight that sales have been falling in the Southwest and you want to argue that this is because of an increasing Hispanic American population in that region, and that Doritos should introduce flavors and marketing aimed at this population.

== 2. Think of counterfactual explanations ==

Think of reasons that your insight might not be true, and imagine the kind of data that you would like to have in order to show whether or not your causal story is true. Brainstorm at least five counterfactual explanations for your (again, completely fictional) insight.

== 3. Identify a story ==

Assemble the pieces of the story into a coherent whole. You may want to consider the storytelling arcs from earlier in the semester. One simple and very effective arc is demonstrated and discussed in EDS, pages 277-291. You may also want to read EDS Chapter 4, especially pages 86-93 and the list on pp. 104-105.

In short, you want to make sure that each slide and each visualization contributes to a linear, clear story that provides credibility to your central insight. Plan out what your story will be, and then create slides.

== 4. Create "visualizations" ==

In the real world, data and money are limited, and we can't always gather the data needed in order to figure out what is going on for sure. However, for this assignment you don't have real world constraints. You just get to imagine that whatever dataset you wish existed actually exists. For example, for our Doritos presentation I could just make up the results from a focus group that gave feedback on a new set of flavors.

You can do this however you want - you can make up data and visualize in ggplot or Excel, you can draw a fake graph in Adobe, etc. Make sure to use the principles you've been learning about to create visualizations that are clear, honest, and advance the story.

== 5. The final product ==

You will produce:
* A slide deck (Powerpoint, Keynote, [https://rmarkdown.rstudio.com/lesson-11.html R Markdown], etc.)
** Should include at least three data visualizations
** Should tell a compelling data story, aimed at the CEO of the brand/company
* A 6-8 minute video presentation of your slides. I suggest using [https://obsproject.com/ OBS] and putting a video of yourself at the bottom of the screen but you can do this however you think is best.