Data mining in the survey setting: why do children go off the rails?

Abstract

Data Mining is relatively new in the field of statistics, although widely used elsewhere. Is it a good idea to discard the model-based methods in favour of Data Driven methods? Data driven methods produce a high degree of accuracy, but very little interpretability. Model based methods are interpretable, but lack accuracy. Data mining techniques are commonly used where the data collection has been automated. I will show these methods are also useful in the large survey setting.