Data: not just numbers; not always big
Pity the poor arts administrator who has drunk the Big Data kool-aid without fully understanding its strengths, limitations and ramifications.
Who thinks data only means numbers (”But how can I quantify that?”) without knowing the many varied research and evaluation tools available to them. See: An intro to qualitative methods.
Who thinks research always begins with a survey. See: What is a research question?
Who thinks you always have to collect original data yourself. See: What is secondary data?
Who doesn’t take the time to find out if the biggest questions the organization is asking have already been researched and answered. See: How to write a literature review.
Who doesn’t consider the potential ramifications of null (or negative!) results before launching a study. See: Research on why researchers don’t publish when their study turns up nothing terribly interesting.
Who has fallen for the Big Data Gurus who argue that quantity can overcome non-representative methods. See: A very technical article about overcoming bias in large nonrepresentative internet datasets.
Who asks grantees to “prove” the “impact” of their programs without understanding what it truly takes to measure cause and effect in the social sciences. See: Actual news headlines that mistook correlation for causality.
As an arts administrator, you don’t need big data, you need accurate information. You need to ask smart questions and provide guidance to your research/evaluation staff or contractors about what’s important to you, then let them figure out what the best methods are to answer your questions. Before diving into the research, make sure your question hasn’t already been answered. If not, see if there is secondary data available already to answer it. Good quality research takes time and resources - make sure you have enough of both to commit to the project.
Data is only as powerful as the people who wield it. As you would with any sharp instrument, use it with care and respect and it will serve you well.








