The North Carolina Cities Data Portal is a work in progress. It aims to fill a gap in knowledge on the changes occurring in North Carolina cities by making data on North Carolina cities easily available and comparable. One of the biggest challenges we face with this project is finding data sources that are reliable, easily attained, and updated on a regular basis. The data currently included here is just the beginning. The long-term goals of this project are to collect data on an expanded range of indicators of community well-being. As part of the next steps in the process we will assess the feasibility of collecting and displaying data on a wide range of indicators. We welcome suggestions for data and indicators to be included in the future.
Data for this report was compiled from the U.S. Census Bureau, including the 1980, 1990, 2000, and 2010 censuses, as well as the three-year American Community Survey (ACS) estimates, starting with the 2005-2008 timeframe. ACS estimates are available as one-, three-, and five-year estimates. One-year estimates are available for areas with populations of 65,000 or more. The one-year estimates offer the most current data but are drawn from a smaller sample and therefore are less precise than the three- and five-year estimates. Only a handful of North Carolina cities meet the population threshold so this wasn’t a good fit for this project.
The five-year estimates are the most reliable of the three ACS estimates, but also the least current since it draws from 60 months of collected data. Because the Census Bureau strongly discourages the comparison of data from overlapping periods, we would have to wait several years to be able to compare the 2005-2009 data with the 2010-2014 data. For these reasons we ruled out using the five-year data at this point in time.
The three-year data offer a happy medium, as it is more current than the five-year data and more reliable than the one-year data. It was available for 44 cities in the state when we began the project and more cities will be included as they exceed the population threshold of 20,000 for inclusion in this dataset. New three-year data will be released annually, allowing us to regularly update the Data Portal.
There are a couple of reasons to be cautious when comparing data over time. First, it is important to keep in mind that the ACS data are based on samples of the population and thus are estimates. Because they are estimates, small changes from year to year could be attributable to the margin of error for the estimate rather than actual changes. Changes could also be the result of modifications to the wording or ordering of questions, which can impact the data. The bottom line is that minor differences between cities and timeframes are not of great importance and we should focus on larger differences or changes that trend over time.
In addition to updating the site when new ACS three-year data is published, we will investigate the availability of data for cities for additional indicators on topics including: housing (including homeownership, foreclosures, and home values); health (infant mortality, diabetes); transportation (length of commute, use of public transit); education (graduation rates); employment (quality, job numbers), environment (air quality, water quality); fiscal health (debt, bond ratings); and more-detailed income and poverty data. We will also assess the feasibility of adding indicators that will not be updated at the same time as the ACS data.
Kevin Park, a doctoral candidate in the Department of City & Regional Planning has played a critical role in this project. Kevin developed the website, including the interactive charts and maps for each topic and helped with the data acquisition. Allison Bullock, Maire Dekle, Sarah Kirk, and Kevin Neary, all graduate students in the Department of City & Regional Planning also contributed to this project.