About Population Estimates

The Alaska Department of Labor and Workforce Development's Research and Analysis Section (R&A) releases population estimates each year through July 1 for all boroughs, census areas, cities, and census designated places. The estimates are used for research and planning, distribution of resources by public and private agencies, and the updating of population projections.

The population estimates begin with data from the last decennial census (e.g., Census 2000 for the annual 2001-2009 estimates), which R&A adjusts for known errors and post-census changes in geography. For estimates each year following the census, R&A incorporates administrative data that can be used as indicators of population change or revised to create the latest series, or “vintage.” Indicator data include Permanent Fund Dividend applications, data from military and group quarters surveys and research, and a state total population estimate from the U.S. Census Bureau. After release of data from the subsequent decennial census (e.g. Census 2010 for the 2001-2009 estimates), R&A makes final adjustments for the intercensal years.

Why do estimates differ from site to site?

Population Estimates for Alaska are produced by the U.S. Census Bureau, R&A, and some municipalities. These different series may be similar or they may differ. There are several reasons for the differences, and the user should weigh all of the following factors in picking estimates.

Definitions of Residency: The U.S. Census Bureau and R&A both estimate permanent full-time residents for Alaska. The number of people temporarily in the state can vary by as much as 20 percent, and some communities may be much more seasonal because of the tourism, fishing, timber, and construction industries. If an individual's usual place of residence is in another state but that person works or lives part of the year in Alaska, that person is not counted as an Alaska resident in these estimates. Some communities may count such people in their estimates.

Different Geography: Different agencies making estimates may define a place with the same name in different ways. The most common form of census error is in defining a geographic boundary accurately or locating individuals relative to a map boundary. Sometimes boundary changes such as city annexations are not corrected before the deadline for producing an annual estimate.

Different Timing of Indicators: Because of the timing of the collection of administrative record indicators, the indicators used in estimates may vary in how current they are. Some U.S. Census Bureau estimates, for example, are made with indicators that are a couple of years old because of the time it takes for the information to reach the bureau from the collecting agencies. Estimates may obviously be made at different times of the year. Because population is constantly changing, estimates produced at different times of the year may differ.

Different Estimate Methods: As indicated, most estimates are produced using administrative record indicators of population change. Different series of indicators used to produce estimates will give somewhat different information. Some estimates, however, are generated by sample survey techniques or may be based on estimating population using housing and households. Some estimates for small places are produced using complete count information.

Different Purposes for Estimates: Most of the time, estimates are produced for planning purposes, and the intent is to produce the most accurate estimates possible. Once in a while, estimates or projections will be produced or commissioned to promote a particular project or community. As a result, the user should always consider the purpose of the estimates.

How are population numbers from the U.S. Census Bureau, R&A's population estimates, population projections, and the American Community Survey different?

The decennial census, conducted by the U.S. Census Bureau, attempts to count each person in the United States every 10 years. The last census was in 2010, and it collected information on age, race, ethnicity, sex, and household status for a range of geographical areas.

R&A produces population estimates for each year between censuses by incorporating administrative records, including Permanent Fund Dividend applications and surveys of special populations (such as military and college students). Estimates are prepared for each of Alaska's boroughs, census areas, cities, and census designated places. Estimates for boroughs and census areas also include age, race, ethnicity, sex, and household status information.

R&A produces population projections by extrapolating data on the components of population change (births, deaths, and migration) by age and sex into the future. There is much uncertainty in population projections because it's impossible to predict the future, but projections based on reasoned assumptions are an important tool for planners and policy makers. The further into the future projections go, the greater the uncertainty.

The U.S. Census Bureau produces detailed social and economic data (such as income level and educational attainment) through the American Community Survey. The ACS is based on an annual survey of households and group quarters (such as prisons and dorms) that's adjusted to match Census Bureau population estimates. Though the ACS provides detailed and often useful data, it can be unreliable for many areas, so margins of error are necessary to use it effectively.