On June 24 we hosted a webinar explaining the background and data behind EMSI’s new job postings analytics. For those of you who were there, thanks for sticking with us through some frustrating technical issues! We were able to complete the meeting, but also received lots of requests for the slides and following Q&A session. The embedded presentation is below, and if you’d like the speaker notes just click the SlideShare icon in the bottom left. Notes can be found under the “Notes on Slide 1” tab.
The Q&A transcript is also included below the PowerPoint, and please feel free to email (email@example.com) or comment with any other questions! We’ll be doing an identical presentation (hopefully without the glitches) on July 9, and another session demonstrating the actual reports on July 10.
Where does the hires number come from?
Hires comes from EMSI’s industry and occupation data, built on multiple government sources including QCEW, OES, and QWI. Quarterly Workforce Indicators provides the basic building blocks for creating hires, separations, and churn for the complete EMSI dataset.
How do you get data on hires by MSA?
EMSI data has been available at the state, metro, and county levels for a number of years. We incorporate sources with different strengths to combine high sector detail and specific geographic regions. In the case of hires, much of the source data is published at the county and MSA levels. Often smaller regions are suppressed in the source data, so we use supplemental sources to arrive at our best estimates.
In the current version of Analyst, is “openings” the same thing as total job postings?
No. The openings number currently in Analyst is an estimate of job openings due to job growth and replacement jobs (workers leaving their occupation for another field, retiring, etc.). Openings are modeled from projections produced by the BLS. The new hires number is a better measure of demand and allows us to compare more closely to job postings.
Will skills data be available?
Yes. Skills data will be available, but not in the initial July release. In-demand skills are an incredibly important value of job postings, and we are working hard to make sure that the skills data is done well. It’s at the top of our list, so stay tuned for announcements on that release.
Would postings possibly also be under-representing labor union sectors?
Most likely, since those jobs wouldn’t likely be posted online in a highly public way. We’ve seen similar issues with certain government jobs and within public schools.
How does this compare to LaborInsight (Burning Glass)? Are you using the same sources?
Without having access to LaborInsight we can’t say exactly how the two compare. On the question of sources, it’s highly likely that there is a lot of overlap between the places BG and other similar tools pull postings from and where EMSI pulls from. Since postings are not collected, sorted, and published by any standard source, everyone is more or less on their own to find them. We also think that postings can only tell you so much, and that comparison to structural LMI is crucial. EMSI is providing both the best LMI available and the ability to compare with postings in one place.
Is this an addition to Analyst or a new tool? Included in a current subscription or for an additional cost?
Job postings analytics will appear as a report within Analyst for an additional cost. We haven’t finalized pricing quite yet, but it will likely be in the $2,500 range.
Have you normalized job titles? In what way?
Yes. We’ve moved the job titles from postings into normalized titles, then connected those titles to their appropriate O*NET and SOC codes to get the appropriate job counts, growth, and earnings figures.
Have you been able to connect postings to industries, in addition to occupations?
Job postings by industry is in the queue, and we’ve made some good progress to that end. Not sure of a release date, but it is something we’re working on.
Can you explain your use of QWI a bit more?
Our recent article on this topic, “Measuring Demand: A Comparison of JOLTS and QWI,” goes into more detail on what the two sources can provide us. The short answer is that QWI uses Social Security numbers connected to mandatory employment records to count the vast majority of all hires and separations in the US economy. This informs our numbers on hires in Analyst, as well as our future churn and separations figures. Since there are still holes to be filled in, like self-employed workers and suppressed numbers, we use QWI as an aid but not the sole source for the new data.
For more on EMSI’s employment data — available at the county, MSA, and ZIP code level — or to see data for your region, email Josh Wright. Follow EMSI on Twitter (@DesktopEcon) or check us out on LinkedIn and Facebook.