American Community Survey data collection operations were temporarily suspended in certain parts of the country three times in 2017 due to hurricanes Harvey, Irma, and Maria. This temporary suspension of data collection has led to higher margins of error for the impacted areas, but does not pose any major comparability concerns. Users are encouraged to continue to account for the margins of error when comparing ACS estimates over time or across geographies.
In early September 2017 after the lingering effects of Irma and the onset of Maria, outgoing telephone operations were suspended in Puerto Rico. Beginning in late September, all modes of data collection were suspended. They were resumed in January 2018 for all modes. Thus, only about eight and a half months of data were collected from Puerto Rico before operations were suspended. There are noticeable impacts to response rates as well as the margins of error for all estimates. Visit the Sample Size and Data Quality page to learn more about response rates. Since data collection did not resume until 2018, the 2017 1-year estimates and the 2013-2017 5-year estimates only reflect the characteristics of Puerto Rico prior to the hurricanes.
Similar disruptions of data collection activities occurred in both parts of Texas and the entire state of Florida. Data collection in regions in Texas along the Gulf Coast and surrounding Houston were suspended up to four weeks due to hurricane Harvey. The entire state of Florida suspended data collection for up to three weeks due to hurricane Irma. However, the durations were much shorter than in Puerto Rico and data collection was resumed in late September. As a result, the impact on the quality of the estimates is also much smaller and users may notice relatively little difference in the quality of the estimates as compared to last year.