Data sets are used all the time. Those outside of the world of metrics and computers put little thought into it. However, all processes, actions, and movement provide information in a data set.
By definition a data set is a collection of information organized as a stream of bytes in logical record and block structures for use by IBM mainframe operating systems. The record format is determined by data set organization, record format and other parameters. [Source]
That’s a technical definition, but what does that really mean? Data sets are generally created by three methods.
- The individual creates the data set for a purpose.
- A collaborator created a data set with a question in mind.
- A collaborator created a data set in order to explore it.
From my experience, a data set is a group of information used to either find unknown information or to prove/disprove a question.
Data
If we have data, let’s look at data. If all we have are opinions, let’s go with mine.
—Jim Barksdale
Data is information. What we do with that information means a lot. However, since action is from data, we better understand the quality of that data.
There are degrees of quality data and methods to define quality data.
- Relevance – Data that is useful to support.
- Timeliness – How quickly data is created.
- Precision – The exactness of data.
- Correctness – Data that is free of errors.
- Completeness – Data that is compete relative.
- Credibility – Data from reputable sources.
- Traceability – Data that can be traced to it source. [Source]
With quality data combined to a data set, we can use this to create metrics. One simple example is school grades for a student. The data is the grades provided throughout the course. The combined grades form a data set. We can view this data set and pull out information of our student.
- Which subject is the student excelling or failing in?
- What is the overall grade point average in each class?
- What is the likelihood the student will go to college or trade school, or neither?
The more quality data provided, the more information we can gleam from.
Therefore, it is imperative to gain as much quality data as possible, in order to portray accurate information. This is why I was shocked to learn how much missing data occurs with abortion data sets. You can read my article, Finding Facts on Abortion, to see what I mean.
We just don’t want to focus to heavily on the metrics without acquiring our original goals. The main purpose of data sets, data quality, and metrics is to create an answer to a question or create an positive actionable item.
Recently, I was curious to acquire quality data for the purpose of creating a metric on ABLE illegal alcohol seizures. I required information on amount of illegal busts. I had two reasons:
- A question in mind (how many moonshine stills) and
- explore the data.
Unfortunately, the data is hard to come by. You can read my experience in my article, Recent Moonshine Busted in Oklahoma.
Hopefully you now have a better idea of data sets, quality data, and metrics we can utilize.
Matt Cole has high regard for knowledge share. He has a desire to share critical thinking and information. With a Masters in Information Technology and a wide array of certifications, while not working full-time, he wishes to knowledge share through providing insight, information organization, and critical thinking skills.
#KnowledgeShare | Matt Cole | #infobyMattCole
