What are data sets?

It is a collection of related sets of information that is composed of separate elements but can be manipulated as a unit by a computer, so that will be the answer.

A data set, as you would imagine, is a body of information. If in one table, there is generally a unit of measurement, such as individuals, communities, states, products, etc. An identifier for the individual unit (such as the OP’s fluther name @RedDeerGuy1) would be listed in the first column and their unique information (such as date of birth, address, level of education completed, etc.) would be in subsequent columns. Sometimes there are multiple records for an individual unit, such as visits to your doctor. Those would be labeled with your identifier and then also an identifier for each encounter (such as date of doctor visit), followed by information related to that encounter (such as your height and weight). If there are multiple tables, then they each list the individual’s identifier so that they can be linked together for reports. Some data sets are simple and can easily be interpreted by an individual just by looking at it. Others are highly complex and require automated interpretation and reporting. Data sets can also hold exclusively numerical data, exclusively qualitative (or word) data, or a mix of the two. For example, I have datasets based on interviews, which have a few numerical fields (such as the date and duration of interview), but are otherwise long word responses.

My response comes from a health research perspective, so my examples are related to my work. But the examples could easily be converted to represent other fields, such as marketing/sales or engineering, for example.

Does that answer your question?