I personally find graphs very easy. This is because I enjoy mathematics, I have my whole life. I know why I like mathematics and that is because its all logic, and there is either a right or a wrong answer, and there is only one way to get to the right answers. Anyway, I was looking at figures to do with Roger Williams University. I actually found a lot of interesting figures about Roger Williams University. For example, that there was only 30 people to graduate with a marine biology degree in the class of 2010. Also that Roger Williams University is 252nd ranked for the average full-time faculty salary.

Anyway, I decided to make a table of the current demographics. Which I found to be very un-balanced. I made a table out of the figures that I have found and here it is.

You can see that the gender seems to be almost the same, if you look at the total of men and the total of women there is only a difference of 73. However, when you look at the difference between race, the numbers are great. I find it quite shocking that there are only 12 American Indian / Alaskan native.

I then started to look at what graphs I could do. Obviously my first graph was a graph with all this information, and then I thought that why put in the totals? Its only increasing the cognitive load. So then I took out the total from the bar graph. And then I also thought, that putting the women and men in separate bars would also increase to cognitive load, but its also should a clear representation of the population in Roger Williams university. So I decided this is a appropriate bar graph. However, to show the demographic population, the gender isn’t that important if they are almost identical.

So then my next question was what graph would be appropriate for showing the demographic population. The population can be considered as a whole, or 100%. So the races of the population can be separated into percentages of the whole population, therefore the appropriate graph for the demographic distribution would be a pie chart or a stacked bar chart. However for this, a pie chart would be better because there it would be clear and have a smaller cognitive load.

My next thought was thinking about showing the data labels so that one can get a more accrete idea of the distribution. However when adding the data labels, its decreases the size of the pie chart and it makes everything look and feel more clustered. So then I thought about just putting the percentages, and this the same problem. So I decided not to add any of the data labels. Another reasons for not adding these labels is that with a pie chart, you can pretty much estimate the percentage given by the area that it covers. For example, just by looking at it, I can see that the white non Hispanic population is around 75% of the total.