Big Data and marker analysis. As methodologies, big data and marker analysis are similar in their
facility to predict implementation failures in program interventions that are intended to continue
over extended periods. Evaluators using big data can identify likely failure factors in later phases
from large data files even after a continuing, ongoing program has commenced. Similarly, evaluators
can also employ marker analysis predictively in the planning of subsequent interventions that
follow initial ones. On the other hand, there are also noteworthy differences in the application of
these two methods. The use of big data involves stakeholders after risk factors have been identified
as statistical predictors. Marker analysis, in contrast, involves stakeholders and other local people as
collaborators much earlier, particularly in the initial identification of likely failure factors. Finally,
strong statistical skills seem a comparatively more necessary qualification for successful big data
prediction than for a marker analysis while expertise in collaboration, teamwork, and methods promoting
a collective commitment are more requisite for success in the latter method.