I have data from so many different sources. How do I use them together to improve our learning from them?
In our experience, the need to combine data from different sources has increased greatly. For example, fusing survey data with tracking data is likely to lead to better learning and decisions than using only one of them.
However, this kind of data fusion often comes with several problems including those related to missing data. Our data integration library vary from the common imputation methods like regression to those that take advantage of Markov Chain Monte Carlo (MCMC) simulation and Maximum Likelihood Estimation algorithms.
Our Data Integration services ensure that your decisions are based on synthesizing all the databases at your disposal, and not by looking at each data source individually.
We also use meta-analysis methods to pool results from multiple studies in the hope of identifying patterns among study results, sources of disagreement among those results, or other interesting relationships that may come to light in the context of multiple studies.