Skagit County Trends Newsletter


United States Census Bureau Logo

The American Community Survey (ACS), from the U.S. Census, provides estimates for the nation, all 50 states individually, various counties and states across the nation. These estimates include various measures of well-being, income, population, diversity, housing, age and more. These estimates are used to keep track of important trends throughout each county.

Each year as every ACS-sourced indicator on the Skagit County Trends site receives an update, the numbers may show a significant change, a small change, or anything in-between. All ACS indicators are estimates based on a national survey. In prior years there was no way to tell the importance of that change, since the Institute does not have access to the source data.

For the 2016 annual ACS updates, the “More Information” section now includes the results of a statistical significance test, making it easier to determine how significant the change in data is.

First, we collect the estimates and Margin of Error (MOE) for the base year (the first year recorded for the trend), then we add the most recent estimates and MOE. An algorithm provided by the ACS determines whether the change is statistically significant or not. We simply enter the estimates and MOE’s. If the estimate is found

to be "statistically significant", it means the most recent estimate is different from that of the base year. This is important because this will show that the trend is changing.

An example of an ACS indicator is the Median Age of the Population, this indicator shows a trend of the median age of the population throughout the combined counties, the state and the nation. For each location, a significance test is performed and we can track the movement of the median age for the nation, the state and the county. If the test results come back as "not statistically significant", it means the median age has not really changed over time. On the other hand, if the test results are "statistically significant", it means over time the trend has seen change due to real factors. For example, if the median age of the population has gotten older this could be due to medical advances and a better standard of living.

Most every ACS-sourced indicator allow statistical significance tests but not all. One example, some indicators use a combination of more than one ACS table and the test cannot be carried out. If test results are offered, the test was able to be performed. If no results are offered, the test was unable to be performed.

With the addition of the statistical significance test results, we can now draw more accurate conclusions than previously available about the estimates offered for each ACS indicator. Knowing if changes are actually significant will help the community better understand what’s truly going on around in our two counties.