drastic difference when using 2013 vs 2014 data
I created a model of nonresident traveler spending in our state using 2013 IMPLAN data, then recreated the model using 2014 IMPLAN data. Of course I don't expect things to be identical between the two models, but I'm seeing some large differences that I can't explain, and I'm hoping you can provide some insight.
Direct employment in the model using 2013 IMPLAN data is 39,137 versus 47,001 using 2014 data. Likewise, total employment is 55,386 versus 63,542.
One specific sector in which I see an unexplainably large difference is 492, Independent artists, writers, and performers. With the 2013 data, employment in this sector is 2,213. With 2014 data, it is 9,246.
Something seems to not be right about this, but I could sure use your help figuring out what is going on. Is there, perhaps, a glitch in the new data?
I've attached a spreadsheet with just the Total Impact Summary and Top Ten Industries Affected for comparison between the two data years.
Thanks very much,
Kara
-
HI Kara! Thank you for your forum Post. So that we can replicate your impacts, would you mind please letting us know the study region and Sector that you are impacting. You could also send the information in an email to info@implan.com Thanks! -
The study region is the state of Montana. I've attached a spreadsheet with the sectors. There are quite a few, since we're looking at many types of travelers spending in the state! The spreadsheet lists the expenditure category and the sector or sectors I've used in the model. Let me know if you need the dollar amount spent in each sector. The dollar amount per sector is the same in each of the two models. Thank you. -
Hi Kara! In 2014, we developed a method to estimate W&S Employment separately from Proprietor Employment (which will allow for better wage per worker interpretations and will allow us to split the State/Local Government tax impacts into further government level detail, which is currently in development). As a part of this process, we incorporated some new data sources (e.g., Census Non-Employer Statistics and CBP Organizations with Employees by Ownership Type) and more involved processes for estimating the proprietor count. These changes are therefore an improvement over past data. We apologize for the concern and confusion; while we try to maintain a certain level of consistency over time, we also try to improve our estimates when possible. Thanks IMPLAN Staff!! -
Thank you. Of course better data and estimation methods are always an improvement, and people will be happy to see higher employment numbers. I'm just sure there will be some criticism from people who see our report that the increase is too good to be true. I just wanted to make sure there weren't any mistakes and that I had a good reason to explain what has contributed to the large increase. Thanks, Kara
Please sign in to leave a comment.
Comments
6 comments