You’ve probably seen the headlines lately regarding the impact of COVID-19 on Census operations.
In fact, just a couple of weeks ago, the U.S. Census Bureau announced that it would hire more employees to accelerate the completion of data collection and apportionment counts by its December 31, 2020, deadline. Even at the time of this being published, there has been discussion by Congress to fund the Census more heavily in order to make up for lost time caused by COVID-19.
We decided it’d be a good time to sit down with one of our data providers, Synergos Technologies, Inc. (STI), to learn how (and if) they’re adjusting their efforts based on the challenges in collecting Census data.
Here’s what STI had to say about their PopStats quarterly population estimates and the potential impact:
STI: The Census Bureau has a big advantage in 2020 that it didn’t have in 2010: the progression of technology. People can much more easily respond via the internet now, and that could certainly help with higher response rates at a time when people aren’t as reachable due to social distancing and quarantine requirements.
It’s yet to be determined how much of an undercount there might be, but we know we’ll get to reliable population statistics regardless because the Census Bureau knows for the most part what households exist right now—they just do not know how many people live in each household. They’ll use their protocols to make their best estimates if they cannot confirm that information—like double-checking problems areas that they have unsure estimations. Undercounting has been raised as an issue since the 1990 Census, and an issue will be raised even more so now with an ongoing pandemic.
Regardless of the problems the Census Bureau faces, as far as Synergos Technologies is concerned, we use U.S. Postal Service Zip+4 data in addition to Census data to build our estimates from the bottom up. This helps us smooth out any potential undercounts.
STI: It doesn’t; however, it might impact our five-year numbers in the future. That being said, our five-year forecast figures will remain unimpacted. When the 2020 Census comes out in late 2021 or early 2022, there will be a rebuild of block groups and past history. Even then, the model will stay the same for the most part.
In return, this does raise an issue we’ve experienced recently pre-COVID. The level of past growth cannot accurately represent the amount of future growth between the number of aging, the level of immigration, and the slowing rate of births. Simply speaking, a lot can happen in five years. For example: in 2008, Clark County was expected to experience an explosive growth rate all the way through 2020. A year later in 2009, the county experienced a negative growth rate.
Bottom line: even with the serious tragedy that struck our country and the world this year due to the pandemic, this will not hinder the ongoing long-term growth that our country is expecting.
STI: We do not. The Census will have its own identification of areas with poor response rates. We were incredibly close to the 2010 Census with our estimate last time around. Even closer than the Census’s own preliminary estimates. The areas where we had a greater error rate where those with Native populations, such as Alaska and Arizona. This was due to the lack of use of the Postal Service there, and that has certainly improved some over the last decade. We have also made adjustments to hopefully better capture these areas. We do look at how Census data stacks up against the U.S.P.S. data (Zip+4) as well as data from social services that tells us about vacancy.
STI: Our five-year forecast has a simulation model that relies on trending to provide forecasts. Our forecasts have proven to be very accurate. In 2010, we checked out five-year population forecast from 2005. Our forecasted population proved to be more accurate than our competitors’ 2010 current-year estimates.
STI: At this point, we feel it is too early to determine whether the 2020 Census data will be less accurate because of the impact of today’s events. The Census Bureau has been impressing us since 1990 with their methodologies and how they’ve combatted challenges. We have full confidence the bureau will be able to make an accurate count again this year.
STI: We have a detailed explanation on our website that outlines what our customers can expect with the release of the 2020 census. Basically, there will be two versions of STI: PopStatsTM for at least two quarters: one using 2010 Census boundaries and one using the new 2020 Census boundaries. We do this so that there is an easy and smooth transition from 2010 boundaries to 2020 boundaries. Details and a timeline can be found on our website.