Will the M make the &E redundant in the age of big data?
Fonte: evaluation.nu .
Monitoring and evaluation, or M&E, are two approaches to assess public interventions. Evaluation is commonly understood to be retrospective, whereas monitoring is on-going during the implementation of the intervention.
In the dawning time of big data following the invention of the internet, unimaginable large quantities of data are produced every second. Through tweets, Facebook updates, google searches, credit card transactions, electronic communications etc. private companies, governments and citizens can follow real time developments in flue outbreaks, consumer spending, public opinion etc.
There is no doubt that big data are here to stay , but where does that leave M&E and the evaluator profession, if one can speak of one such?
Big changes will of course not happen over night. But they are likely to happen much faster and change governance patterns and the need for information dramatically. Think back on when you got your first smart phone. It is probably not that many years ago and the technology is quickly becoming available in all corners of the world.
Slowly, businesses and governments will have more and more real time accurate data on key indicators and the need for retrospective evaluation will diminish as a consequence. Because, why take the trouble of evaluating, when the results have already been provided by the monitoring of e.g. satisfaction of target groups, the reading skills of pupils, the health indicators of citizens etc.?
So, what will be of us – the evaluator? First, large quantitative data or quantifiable qualitative data need to be mined and structured. It is likely that this will be a job for the IT-guy and statisticians. The old fashioned qualitative data such as the interview or focus group are likely to slowly be redundant as target groups are simply asked to fill in online surveys on their devices, allow cookies and apps to monitor their behavior or interact on social media. The old fashioned evaluator might be the guy called upon to set it all up and figure out the indicators in the first few years ahead. But down the line, standard indicators are applied by the programmer in algorithms most of the time. In science, interviews will still be applied to collect qualitative contextual data. But in the evaluation trade, the demand will decrease as will the willingness to pay for expensive qualitative data collection like face-to-face interviews.
Do you agree that the future of monitoring will be glorious at the expense of evaluation? Throw a comment or read more about the subject in Robert Kirkpatrick’s interesting contribution in the European Evaluation Society’s newsletter Connections from March 2013.