Describe your data with metadata

In the course of your working life as a professional Psychologist you will routinely generate data. By now you will have already generated varying types and amounts. Some will be quantitative and some will be qualitative, some might be from traditional experiments, observational studies or clinical work.

The common thread that links the data you generate is that it’s often rendered unusable in a frighteningly short amount of time. This is irrespective of the perceived importance at the time you collected it, did an analysis and wrote it up!

Data can become unusable for a variety of reasons, e.g. data losses through hard drive failure, accidental deletion, computer upgrades, software that can read the data becomes obsolete or file formats change. However the most common reason is that you, as the person who generated the data in the first place, forget what it means! Or rather you forget how the data was coded which means you can no longer make sense of it. This is especially true if you are looking at the data in isolation without the written interpretation or publication that went alongside it.

With the big push to reuse and share data more widely this is becoming of paramount importance. If no one can understand your data it will be devalued. Remember it might be you a few years down the line that wants to reuse the data you collect today!

By describing your data with metadata you effectively annotate it so that you give it context and meaning. Metadata to all extents and purposes is data about data. Doing this isn’t as complicated as it sounds and it’s better to do it as you go along rather than see it as an additional onerous chore that’s done at the end of a project.