For our dataset, we chose to do arrest data in Pittsburgh. We separated the data into neighborhoods and race and age. The first stacked bar is of all the options. There is a drop down menu to choose between different comparisons. Second on the drop down is Men and Women. This elaborates which gender commits the most crimes. Each age is stacked on top of each other to get a feel for how old people are that are arrested. I wanted this part of the graph to show that women do get arrested and they are not always young and reckless. The scatter plot shows a better representation of age, but doesn’t tell the same story as the stacked bar graph. The scatter plot shows a very good distribution of data and lets you see the offense to each dot. The next breakdown is women by race. Since Pittsburgh is an urban area, it is important to examine real data on race and crime. The same goes for men and race. I decided to do a breakdown of the non-white races to get a closer look at the smaller ones. In women and men of color, I wanted it to be noted that the other races are barely accounted for, but this could be because of the urban environment and its high black population.
I chose to do the graph using plotly because it is what I am most comfortable working with. Our original goal was to have the user choose the x and y axes, but the data did not really support that. We were working with data that came in forms that were not numbers. I think this posed a very big challenge for me because my last project was a relatively small data set with only integer numbers. I had to adapt to this and problem solve for how to get so much data into a way that I could handle it. It is very frustrating when the data is not numbers because that can skew how you want to present it. It is also difficult choosing which categories to pair with each other. I was interested in comparing race or age with time, but once Alexa had parsed the dates from the times, it was still difficult to access them.
While I was waiting on other things to be done, I started to play around with the mapping of the latitude and longitude. I did run into some problems because plotly does not have enough options for selecting the base map. I tried changing the range of the lat and long like they did in an example, but it was not working for me. I did learn that plotly and mat plot lib have a lot to offer. They are very good at taking data and making something of it. I was surprised that it knew when each data point was from the same neighborhood. For some reason I thought I would have to separate it out.