Product

I made a bar graph that shows that every sleeping bag weekend, the weather in Pittsburgh improves dramatically, from higher temperatures to clearer skies. I used Excel to make the graph, then touched it up with a couple of images of sun and shine. I colored the days that students were on campus gold to make it seem warmer, and more like sunlight, while the days they weren't there were a colder blue. I also used just the highs of the days students were on campus, when on the other days I considered the highs and lows. Pictured in the graph are the day before, the days of, then the day after the sleeping bag weekend, in that order.

Intention

It's a widely known myth that the weather improves when prospective students are visiting, but since my theory takes it one step further, I wanted to support it with solid data (or seemingly solid data).

Context

I tried to make this in the same vein as the Fox News infographics that use partially true or related data to paint a targetted outcome. 

Process

I started by trying to build a graph using google charts. This took up many hours and wouldn't compile, so I switched to Excel. Unfortunately, I have no images from the google charts fiasco because I couldn't get it to show up. 

Excel gave me less freedom, so I tried to rely less on the graph and more on manipulations that I could do after it was done. In the end, I decided to highlight the relevant days to make them seem warmer, and throw in some sunny skies and rainy days. The Rain is thrown over any day that's "cloudy or partly cloudy" and the sun is thrown over days that are "partly cloudy or sunny".

Reflection

Overall, I think I flopped on this project. I really liked my idea, but I didn't execute it well in the time given for the assignment. Next time, I won't spend so much time using a program I couldn't figure out how to use.

I did learn that there's a lot that goes into fudging the numbers, and if I had more time, I'd radically change my graph. I'd want to show more misleading "vs" data, and have more data in general. Some "temperature vs. power output from Newell-Simon" or even "predicted temperature vs observed," would have been nice. Also, I would've loved to find another way to build the graph that was more aesthetically pleasing.

Attribution

Sun image: http://www.thebigboot.com/sat-25th-yippee-sunny-day/

Rain image: http://www.softicons.com/web-icons/vector-stylish-weather-icons-by-bartosz-kaszubowski/cloud-rain-icon