Risa Proehl on tracking population and trends

photo by Dina Avila

Though Risa Proehl keeps a low profile, her work can create an enormous financial impact. As a demographic analyst who manages the Oregon Population Estimate Program, a lot of state and federal funding hinges on Proehl’s findings.

She has been a research faculty member with Portland State University’s Population Research Center since 1998. We sat down with her to talk method and trends.

Explain the process for putting together population estimates.

Our Center is mandated under the Oregon Revised Statutes to produce annual population estimates for Oregon, and its counties and incorporated cities (and towns, we have nine incorporated towns). The annual population estimates are as of July 1, and the estimate’s cycle runs from July 1 to June 30. Each summer we send the counties and cities a questionnaire asking for recent information on housing and population change that occurred in their jurisdiction during the past year.

For cities and towns, we use a housing unit method. This method relies solely on the information collected from our annual questionnaire and on baseline counts from the most recent decennial Census. We start with the Census populations, and then we estimate numbers of people residing in new housing units by assuming average numbers of people in households and occupancy rates from the Census. To that household population, we add any changes in the group quarters population.

We prepare the county estimates using multiple data sets that are available only at the county level. We use the questionnaire data in a reasonableness check of the preliminary estimates. To develop the county estimates, we use statistical methods that relate population to data that are symptomatic of population change. In addition to Census counts, some of the data sets we use are on vital stats, school enrollment, DMV driver’s license issuances, employment, Medicare enrollees, state and federal tax exemptions and registered voters.

We update the estimates quarterly to account for any annexations experienced by cities throughout the year. We request cities to send us detailed information about the number of housing units and persons included in the annexations.

What are its strengths and weaknesses?

A strength of our process is that we collect local data, and that we have direct and ongoing communication with local government officials and staff regarding population change in their area. We are able to easily confer with local officials and staff if there is a question about population change or about the data.

A strength of the Estimates Program is the duration of it being housed in the Population Research Center and the consistency associated with that. Our center was created in 1955 and has been producing Oregon’s annual estimates since its origin. We have an efficient system in place for producing the estimates, while also being able to give individual attention to each of the thirty-six counties and 242 cities and towns, as needed.

A weakness of the estimation process is that, although we request annual information from cities and counties, not all comply all of the time. It is not so dire for a county since we rely on other data sets to produce county level estimates, but if a city does not submit recent information, we hold its estimate constant to the previous year, while we know that in some places the population is likely changing, at least slightly.

In the Estimates Program, though, we don’t want to produce an estimate that is solely based on estimated data, and we don’t want to surmise changes have occurred if we don’t have evidence. That said, our response rate for cities and towns is almost 80 percent, which is considered very high for a survey.

What decisions are being made on the state and local level based on your figures?

The estimates determine how state revenue is distributed. The population estimates are produced so that an equitable disbursement of the state revenue to cities, towns and counties can be made.

We also develop estimates of population for age groups. The Oregon Department of Education and Educational Service Districts use our estimates of population ages 4 to 20 for apportionment of school funds, and the Oregon Youth Authority apportions financial aid for court services to counties using the 0 to 17 age estimates.

Other Oregon agencies and departments, such as the State Library, Housing and Community Services, ODOT, Parks and Recreation, Public Health and DEQ, to name a few, are also required to use our population estimates for the statewide distribution of program funds.

Researchers use our estimates to establish demographic trends or to calculate rates, and nonprofit organizations use them for reporting populations in grant applications.

The populations on “welcome” signs in Oregon cities and towns include our numbers, as well.

What do you consider the most important aspect of your job as the manager for key population figures?

To ensure that we are in compliance with the Oregon Revised Statutes and the Oregon Administrative Rules, and that the estimates are carefully prepared with best practices and methods, which includes listening to what local government officials and staff have to say.

Have you ever been approached to change a population estimate to benefit a municipality?

Certainly. The primary reason a city or a county challenges their population estimate is because they think their population should be higher. This is understandable because the higher the number, the larger the share of state revenue they will receive.

In my experience, all inquiries made by jurisdictions in the challenge process have been reasonable. That is not to say, though, that we adjusted the populations from all the challenges. And there have been cases where a challenge was brought forth and based on the new data submitted, either our original estimate was validated, or we even adjusted the population down.

Has your population estimation system ever failed you? Explain.

My first reaction to the question was, “No, not me, personally.” But, on second thought, it depends on what we consider failure. What I am thinking of is not so much a program system failure, but something that is a constant hindrance to producing accurate estimates that we can’t avoid. That hindrance is the occurrence of demographic change that can’t be incorporated into the estimates because we don’t know about it. We don’t know about it because the change is not showing up in the data or the data simply are not available.

What trends among Oregon’s population strike you the most?

Oregon’s population is aging and the white non-Hispanic population is decreasing. These trends are similar to the trends of the nation as a whole.

An aging population means that the shares of population in the older age groups are increasing and the shares in the younger age groups are decreasing. This situation in not unique to Oregon, and it was expected to happen because the Baby Boomers have begun to reach senior citizen status and fertility rates in general have been slightly decreasing. Oregon’s total fertility rate has been below replacement level since at least 2000. Some counties and areas in Oregon twenty or thirty years in the future could lose population due to a natural decrease unless net in-migration of newcomers continues at slightly higher rates than have occurred recently. For Oregon as a whole though, a natural increase and net in-migration together keep annual growth rates positive.

Another main trend is that the share of population that people of color represent has been increasing. The percentage of white non-Hispanics is higher in Oregon than the U. S., but the percentage is declining. This trend is occurring in many counties throughout Oregon, even in areas where the percentages of white non-Hispanics are highest.

What should everyone know about population estimates?

That they are just that—estimates. They are not the result of an enumeration or an actual head count. Although we strive for accuracy when producing the population estimates, and often we are successful in coming extremely close to matching Census counts, the estimates are only as good as the data used. If we don’t have current, decent data, it is difficult to produce an estimate with a high level of precision. Also, I would like people to know that in our program, we depend on and welcome input from local government officials and staff to help improve their accuracy. Developing Oregon’s population estimates is a joint effort.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.