Why Doesn’t Emsi Burning Glass Use OES State-Level Staffing Patterns?

The BLS publishes state-level estimates of occupational employment by industry. We researched this data to evaluate the possibility of using it to aid in the process of regionalizing staffing patterns. We use regionalized staffing patterns to convert industry job counts to occupation job counts, so it is important to ensure that our staffing patterns are accurate.

We decided not to make use of the data, for several reasons. First, OES publishes these data as research estimates and cautions users against using them for anything other than research. Since our users expect to use Emsi Burning Glass data to make actionable decisions that impact their economies rather than merely conduct research, we are hesitant to introduce these estimates into our data.

Second, OES has far fewer occupation categories than Emsi Burning Glass does at this step in the process. Emsi Burning Glass has 147 occupation categories, whereas OES offers 30. This means that OES is not showing many occupational categories into which industry employment can be split out. Using OES data would result in a loss of detail.

Third, these research estimates are rounded in some cases, and about 10% of the data is suppressed. In order to use this data, we would have to model it rather heavily, which defeats the purpose of incorporating it. Ultimately this would add complexity to our processes without adding any certainty.

Finally, since these are research estimates, the BLS may decide at any time to discontinue publishing them. We hesitate to create dependencies on data products that have no assurance of their longevity.