--- ## Part 4: Extending Our Understanding: Bar Charts and Boxplots (22%) Scatter plots can only give us so much context to the data. We need additional charts to start observing trends more easily. One thing we hadn't considered thus far in our data is the number of cameras across time. What if our data is actually biased by having more recent cameras than pre-2000 cameras, for example? Are there other ways to visualize the same trends we've explored above? Let's try to understand the distributions of our data in different ways.. For this part, we need to extend into different chart types other than scatter plots. Make sure to get acquainted with the possible [mark types](https://vega.github.io/vega-lite/docs/mark.html#types) provided by **Vega-Lite**. We'll be focused on two distinct types: Bar Charts, and Boxplots.

Bar Charts v.s. Histograms: Which to Use?

Bar charts and histograms are very useful tools for demonstrating the distribution of data across either a spectrum of data or across aggregated sections of data. There is a distinction between them, however.

Bar charts describe distinct, discrete categories in the data. This can be features such as labels, company names, or seasons of a year. The bins of bar charts can be either nominal or ordinal.
Histograms describe the distribution of data across a spectrum of continuous data. The feature that is binned is usually quantitative and requires some intuition on how to divide the continuous data into equal segments.

For the purposes of this assignment, we'll explore bar charts. However, just know that Vega-Lite offers the ability to generate both kinds of graphs using the markBar() method; it's all about what kind of data you feed into it that determines whether it's a bar chart or a histogram. For the determined, we recommend at least viewing the possible types of encodings useful for bar charts.