Nick Barrowman at The New Atlantis:
The word data is derived from the Latin meaning “given.” Rob Kitchin, a social scientist in Ireland and the author of The Data Revolution (2014), has argued that instead of considering data as given it would be more appropriate to think of it as taken, for which the Latin would be capta. Except in divine revelation, data is never simply given, nor should it be accepted on faith. How data are construed, recorded, and collected is the result of human decisions — decisions about what exactly to measure, when and where to do so, and by what methods. Inevitably, what gets measured and recorded has an impact on the conclusions that are drawn.
For example, rates of domestic violence were historically underestimated because these crimes were rarely documented. Polling data may miss people who are homeless or institutionalized, and if marginalized people are incompletely represented by opinion polls, the results may be skewed. Data sets often preferentially include people who are more easily reached or more likely to respond.