Tag Archives: pattern analysis

Importance of Census Tracts in Data Analytics

census tracts are important for many reasons. It is easy to misidentify or misunderstand patterns and characteristics within cities, counties and metros which become obfuscated using these higher level, more aggregate, geographies. Many cities and counties that might be experiencing demographic-economic decline will often have bright spots that are groups of a few or many census tracts.

Patterns of Percent Population with Bachelor’s Degree
— by Census Tract; Los Angeles Metro
The following graphic shows percent population age 25 years and over with bachelor’s degree by census tract based on ACS 2014 5 year estimates for a portion of the Los Angeles metro. Accommodating different demographic-economic thresholds/patterns, different legend color/data intervals are used. The pattern layer is set to 80% transparency enabling a view of earth features. Click graphic for larger view, more detail and legend color/data intervals. This map illustrates the geographic level of detail available using census tract demographics and the relative ease to gain insights using geospatial data analytics tools.

– View developed using CV XE GIS and related GIS project.

Get a Custom Map for Your Area of Interest
Use this form to request a no fee map graphic similar to the one shown above for a county of interest. Enter the request with county name and state in the text section; e.g., “Requesting social characteristics tract map for Cook County, IL.”

This section reviews reasons for the importance of census tracts in data analyyics. See related Web sections on tools, resources and methods that you can use to access, integrate and analyze U.S. by census tract general demographics data. The U.S. national scope Census Tracts Demographic-Economic Dataset contains approximately 600 subject matter items tabulated for each census tract organized into four subject matter groups:
General Demographics
Social Characteristics
Economic Characteristics
Housing Characteristics

Importance of Census Tracts for Data Analytics
Census tracts are important for many reasons.  A partial list of reasons is provided below.
• Covering the U.S. wall-to-wall, census tracts are the preferred “small area” geography for superior data analytics.
• The Census Bureau now produces annual tract demographic-economic data from the American Community Survey;  there is an evolving time-series at the tract level creating new analytical opportunities.
• Originally developed to equivalence neighborhoods, many still do.
• Defined by the Census Bureau in collaboration with local groups, tracts typically reflect boundaries meaningful for local area analysis.
• Defined generally for use with each new decennial census, most tract boundaries are stable and non-changing for ten years and many much longer.
• Designed to average 4,000 population, there are more than twice as many census tracts (73,056) than ZIP code areas (33,129).
• Tract boundaries are well-defined; unlike ZIP code areas which are subject to multi-sourced geographic definitions.
• Many data developers (e.g., epidemiologists) use census tract geography to tabulate their own small area data enabling more effective use of those data with Census Bureau census tract data.
• As a statistical geographic area (in contrast to politically defined areas, census tracts are coterminous with counties; data at the census tract level can be aggregated to the county level.
• Small area estimates for tracts are typically more reliable than for block groups.
.. census tracts are comprised on one or more coterminous block groups.
.. on average, a census tract is comprised of three block groups.
• Census tracts are used by many Federal, state and local governments for compliance and program management.

The annually updated American Community Survey provides “richer” demographic-economic characteristics for national scope census tracts. While Census 2010 provides data similar to those items in the General Demographics section, only ACS sourced data provide details on topics such as income and poverty, labor force and employment, housing value and costs, educational participation and attainment, language spoken at home, among many related items. The approximate 600 items accessible via the tract dataset are supplemented by a wide range of additional subject matter. ACS census tract data are updated annually in December of each year.

Weekly Data Analytics Lab Sessions
Join me in a Data Analytics Lab session to discuss more details about using census tract geography and demographic-economic data.  Learn more about integrating these data with other geography, your data and use of data analytics that apply to your situation.

About the Author
— Warren Glimpse is former senior Census Bureau statistician responsible for innovative data access and use operations. He is also the former associate director of the U.S. Office of Federal Statistical Policy and Standards for data access and use. He has more than 20 years of experience in the private sector developing data resources and tools for integration and analysis of geographic, demographic, economic and business data. Contact Warren. Join Warren on LinkedIn.

Comparing Census Tract Demographics Over Time

.. it’s about more than census tracts .. this section is about comparing American Community Survey ACS 5-year estimates: 2005-2009 with 2010-2014 … something new and powerful happening this week.

To make good business decisions we need hard data, recent data, trend data … to assess patterns and change and develop reliable, superior plans. Read about the past and then how things have changed for the better.

Imagine that it is 2005. Data from Census 2000 are now 5 years old. There will not be another update for richer demographics for all counties and cities in the foreseeable further. There will not be any update for small area geography such as census tracts or block groups until Census 2010. Businesses are forced to use out-of-date data to assess markets … where and how are opportunities changing? City and neighborhood planners can only make educated guesses to respond to growing needs of various population groups. Federal and state government programs that base funding allocations on demographics are challenged. Changes in the rental vacancy rates for most cities, counties and metros will remain unknown for the foreseeable future.

Fast forward to 2015 and present day reality. The situation is now radically different. First, we can now compare 5-year estimates from the 2009 American Community Survey ACS to those from the 2014 ACS 5 year estimates. Second, we will be able to do that again in 2016 — compare 5-year estimates from ACS 2010 to those from ACS 2015. Health planners can now assess the size and change in special needs population and how that matches up to resources that respond to those needs — rather than guessing. Schools and school districts can better understand how school age population trending and plan for enrollment change. Education agencies are better able to assess how changing demographics among school systems compare to one-another. Businesses can now determine the size of potential markets and how they are trending based on hard data. It is possible to compare changing patterns in rental vacancy rates and rental housing market conditions for all levels of geography down to block group.

The American Community Survey ACS provides a wide range of important statistics about people and housing for every community in the nation. These data are the only source of local estimates for most of the approximately 40 topics it covers for even the smallest communities. It produces statistics for ancestry, language, education, commuting, employment, mortgage status and rent, as well as income, poverty and health insurance. The ACS estimates are tabulated annually as 1-year estimates (e.g., the ACS 2014 1-year estimates) and 5-year estimates (e.g., the ACS 2014 5-year estimates. See a comparison below in this section about scope, advantages/disadvantages, and other usage attributes for the 1-year versus 5-year estimates.

See ACS 2014 5-year main page for additional data access & use details.

Data from the 5-year estimates are available for all geographies down to the block group level regardless of population size. Starting with the ACS 2014 5-year estimates, for the first time, users will be able to compare two non-overlapping five-year periods 2005-09 and 2010-14. Looking ahead, data from the 2006-10 and 2011-15 (available December 2016) will be comparable … and so on. Over several years, a time-series of 5-year estimates, non-overlapping five-year periods, will evolve.

Comparing Geography Between 2005-09 & 2010-14 ACS 5-Year Data
The following graphic summarizes geographic tabulation areas for 2005-09 and 2010-14 ACS 5-year data. Use the corresponding Web table as a reference guide for comparing data over time. Links provided in the table enable you to navigate to selected data access tables. This Web-page table updates with new links; bookmark the page for re-visits.

Updates
Posts later this month will provide updates on this topics; new data and new data analytics tools. Join me in a Data Analytics Lab session to discuss use of these data using analytical tools and methods applied to your situation.

About the Author
— Warren Glimpse is former senior Census Bureau statistician responsible for innovative data access and use operations. He is also the former associate director of the U.S. Office of Federal Statistical Policy and Standards for data access and use. He has more than 20 years of experience in the private sector developing data resources and tools for integration and analysis of geographic, demographic, economic and business data. Contact Warren. Join Warren on LinkedIn.