Oh, all the wonderful data you will collect! Data is all around us, you just need to know where to look. If you wanted to, you could download a dataset online for free, and play around with thousands of data points. You might also want to collect and categorize your data for analysis.
Most scientific experiments start with data collection. Understanding the class of data you are collecting is an important step in designing your study. Here, we will summarize the different types of data and define them for you, with examples.
The nominal class of data possesses qualitative aspects. It could be something like hair color, favorite book, or hometown. Remember, if you are collecting nominal information, you could be at risk of introducing bias. Make sure you don’t use leading questions during your data collection process.
You can not use the same quantitative methods with nominal data. However, you will be able to calculate the mean and mode. You will still be able to use these variables in comparisons between two different groups.
How confident are you that you could win a wrestling match with a gorilla? You could say:
1 – Very confident
2 – Confident
3 – Unsure
4 – Not Confident
5 – This is not going to end up well!
Sure, there are some numbers involved but the numbers themselves do not carry any value. However, the numbers do have some intrinsic order to rank confidence. We just do not know the size of the difference between 1 and 2 or 3 and 5. But we do know that 1 is more confident than 2, and 5 is significantly less confident than 3. You can determine the mode and median of this data to give you an idea of averages.
The interval scale is as self-explanatory as they come. It features measurements that tell us both the order and exact difference between variables. Think of measuring temperature in Ireland. You can use the Celsius scale to track how warm it is outside.
You know that 10 degrees Celsius is smaller than 20 degrees Celsius. You also know that the difference between 10 and 20 degrees Celsius is the same as the difference between 30 and 20 degrees Celsius. But where these measurements fail is giving us ratios, providing meaning for a value of zero. This means that we can not say that 20 degrees Celsius is twice as hot as 10 degrees Celsius.
This is the first type of variable that we have encountered where you can calculate mean, median, and mode.
Height, weight, and time are classical ratio data measures. Not only is it an ordered numerical scale where we know the magnitude of differences, but we also have a true zero. Having a true zero lets us compare data using ratios as well as opening up a whole array of inferential statistics we can perform during further analyses.
If you want to design a good experiment, you will need to determine what kind of data you need to collect. Each class of data has unique type features and nuances, as well as weaknesses. The data you collect will also determine how you measure central tendency and your later statistical analyses.