General Visualization Study

continuous

variable for which, in theory, there are an infinite number of values between any two values

Pie vs Donut

A better alternative for pie charts in general is the donut chart, which is just a pie chart with a hole in the middle. Even though the pie chart (left) and donut chart (right) in Figure 4 depict the same data, removing the data where the donut's hole is helps us to instead see the chart as a circular stacked bar chart. That is, our brain shifts from trying to make sense of angles and areas in the pie chart to seeing comparative lengths around the image in the donut chart. In a quest to convey maximal insight with visualizations, a simple change in a visualization type like this could make a big difference in comprehension for users.

Categorical or Continuous Data:

Categorical or Continuous Data: Categorical data, also known as discrete data, refers to information that has a finite number of categories. Categorical variables can be further classified into nominal and ordinal types. Nominal variables refer to categorical data that fall into a defined number of categories, but there is no inherent ordering to the categories. Ordinal variables have an order to them. For example, "type of packaging" might have the options "glass," "plastic" and "aluminum." The type of packaging, then, is a nominal variable, because there is no particular way to order "glass," "plastic" and "aluminum." But if you had "container size," and the options were "small," "medium" and "large," you would have an inherent ordering to the categories and would thus have an ordinal variable. Continuous data, on the other hand, refers to data that theoretically is infinite, is not categorized and is part of a continuous whole. Continuous variables can be further classified into interval and ratio types. Interval variables refer to data that is in a specific order and there is a known difference between the values. Ratio variables are essentially interval variables, but also have a known zero point. Temperature, for example, is an interval variable, because there is a set order to the data (i.e., 40 degrees Celsius is greater than 30 degrees Celsius) and the scale is consistent (i.e., the difference between 40 and 30 degrees is the same as between 52 and 62 degrees), but there is no actual zero point (i.e., you cannot have "no" temperature). The length or weight of something, however, is a ratio variable, because there is a set order to the data (i.e., 200 pounds is greater than 100 pounds) and the scale is consistent (i.e., the difference between 202 and 242 pounds is the same as between 3 and 23 pounds), and there is a zero point (i.e., there is a theoretical minimum with weight that ends at zero).

Quantitative

Data that is in numbers

qualitative data

Information describing color, odor, shape, or some other physical characteristic

Quantitative or Qualitative Data:

Quantitative or Qualitative Data: As their names suggest, quantitative data is concerned with numerical quantities, while qualitative data is concerned with qualities or non-numerical attributes. Business data almost always involves both kinds of data. A typical sales report, for example, is loaded with both quantitative data (e.g., number of units sold, number of units left in inventory, revenue, prices) and qualitative data (e.g., geographic regions, names of products, names of sales managers, descriptions and names of products).

distinctions between nominal and ordinal variables

The distinctions between nominal and ordinal variables and between interval and ratio variables are important for knowing the level of statistical sophistication available to you when you attempt to analyze data. More advanced statistical analyses — t-tests and regressions, for instance — require interval and ratio-level data. Continuous data can also sometimes be transformed into categorical data through a process known as "binning" or "bucketing," which opens up those data points to different types of visualizations. For example, time is a continuous, ratio-level variable. Say you are tracking all visitors to your website, and note how long it took from first arrival on a homepage until the time someone completed an online purchase. You would have a potentially large and widely varying dataset that could be visualized in a scatter plot or some other visualization type suitable for continuous data. However, it may be more effective to bin the data into a few buckets (e.g., "less than a minute," "one to five minutes" and "longer than five minutes") and visualize this now-categorical, ordinal data in a bar chart or donut chart. Depending on the distribution of the data and the insights the analysis reveals, a scatter plot may be more or less effective than a bar chart at visualizing time spent on a website.

categorical

absolute; without exception