*This image was taken directly from the lecture.
This lecture pod was about one thing and one thing only – Data Types. There are four different types of data listed in this lecture. First was Nominal, then Ordinal, Interval and lastly Ratio.
Nominal data (pertaining to names) is named categories that can be counted and used to calculate percentages but you cannot take averages from them. The best example of this would be a supermarkets section (e.g. dairy, produce, canned & frozen), each category would be classes as nominal data.
Ordinal data is all about order. This type of data can be counted and be used to calculate percentages. There is also a debate about whether this data can be used to take averages, but I won’t go into it (my opinion is yes, it can). This data has no true mathematical value with numbers being assigned to make data analysis easy. The numbers just need to be in order. The best example of this data would be calculating which line will get you out of the supermarket quickest (long line, medium line, short line etc.).
Interval data refers to data in which the interval between each point is fixed. This type of data is numeric. The best example of this would be time, there is always a set amount of seconds in each minute making it interval data. In this type of data, 0 does not mean that nothing is there (0˚C does not mean that there is not temperature).
Lastly there is ratio data. This type of data is numeric, like interval data but differs from it in one key aspect; it has a meaningful 0 point. To put it clearly, in this type of data, 0 is the absence of anything. Examples of this data would be height, age, weight and money.
I believe however that the most important point made in this lecture was made early on, about whether data types matter. The answer is obviously yes, otherwise we wouldn’t use them but the reason for this is exceptionally simple. We use data types to prevent mistakes, like a postcode being interpreted as a pin code or something like that. This is the most important point because it drives home one simple point to me, it is easy to make mistakes, but it is easier to avoid them if you label your data correctly.
Waterson, S. (2016). DataVis POD02- Data Types. Retrieved from https://vimeo.com/176274669