The American Community Survey (ACS) is an ongoing survey administered by the US Census Bureau which provides information on people, economy, housing, and more. This interface maps some of the most in-demand variables for Connecticut towns.

All towns have data available as 5-year estimates. This means that the responses were collected during a 5-year interval; for 2020, this is between January 2016 and December 2020. The 2020 data release is the first that allows the comparison of three non-overlapping intervals, 2006-2010 (2010 5-year estimates), 2011-2015 (2015 5-year estimates), and 2016-2020 (2020 5-year estimates). See table below the map for this comparison.

In addition to this interactive tool, we summarize important, high-level findings for the state of Connecticut in this blog post.

The map below shows for all towns in Connecticut, according to 2020 ACS estimates. Darker blues represent higher values, and lighter blues represent lower values.

**Move your mouse** over towns to see the values.
**Click towns** to see where they fall in the distribution
(shown in salmon color), and to filter the towns in the table below the map.

If you have trouble seeing or using the map or the bar chart, we recommend using the most recent version of Firefox or Chrome browsers.

The table below shows by town for three survey releases, ACS 2010 (covering 2006–2010), 2015 (covering 2011–2015) and ACS 2020 (2016–2020), and calculates change in the last two columns.

It is important to note that the ACS is a survey of a sample of residents,
not all residents. Therefore ACS estimates have margins of error which
provide information about the precision of the estimate. Roughly speaking,
we have 90% certainty that the “real” value is around the estimated
value plus-or-minus the margin of error. We show the margins of
error after each estimate in the tables and maps below. Additionally,
we report the statistical significance of changes from the 2006-2010 and 2011-2015
to the 2016-2020 ACS 5-year estimates. If a change is not statistically
significant, this means there is insufficient statistical evidence to
conclude that the actual change is different from zero.

Show me an example!

Margins of error that are larger than ⅓ of the estimate are
shown in red. Changes that are statistically
significant are shown in blue.

For example, median household income in New Haven in 2006–2010
was $46,357 ± $2,163 (expressed in 2020 US dollars).
In other words, we are 90% confident that the true median household
income in New Haven in 2006–2010 is at least $44,194 and at most
$48,520. In 2016–2020, the 2020 ACS estimate was $44,507 ± $2,737.

We can calculate the absolute change between the two non-overlapping periods
of 2006–2010 and 2016–2020:
$44,507 - $46,357 = -$1,850. We can also calculate percent change:
($44,507 - $46,357) ÷ $46,357 × 100 ≈ -4.0%.
Using the formula
to calculate margins of error for percent change, we get ± 7.4%.
The percent change is then estimated at -4.0% ± 7.4%.
In other words, we are 90% confident that the true change is between
-11.4% and +3.4%. Because this interval contains a 0, this change is
*statistically insignificant*.

**Click column headers** to sort the table by that column.

* Note that dollar amounts shown are adjusted for inflation, i.e. converted to 2020 US dollars. 2010 US dollars were multiplied by 1.189763, and 2015 US dollars were multiplied by 1.092577 according to the average annual R-CPI-U-RS values.