Show The Graduate Center Menu

The 2005-09 ACS: Implications for 2012 NYS Legislative Redistricting

 
 

The 2005-09 ACS: Implications for 2012 NYS Legislative Redistricting

NOTE: The maps and analysis below have been superseded by updated maps using official 2010 population counts. Please click here for the new information. Thanks!


By 2012, the New York State and Assembly will be redistricted. How the lines are drawn -- and whether the new districts tend to concentrate or disperse populations that tend to vote Republican or Democrat, among other characteristics -- will have a substantial impact on state policies for the next 10 years, and for the next round of redistricting in 2021-22.

The analysis below, based on the American Community Survey (ACS) population data for 2005-2009, indicates that New York City would gain a State Senate seat while the state's upstate region would lose a Senate seat. In the Assembly, it appears that Long Island would gain a seat while New York City would lose an Assembly seat.

The 2005-09 ACS

Last week the US Census Bureau released a data set that offers a glimpse into how the current Senate and Assembly districts may need to change. On Tuesday, Dec. 14, the Bureau published population estimates from the American Community Survey covering a five year period, from 2005 through 2009. It was the first time the Bureau provided detailed population data inbetween decennial censuses at the local level, and for a variety of other levels of geography including state legislative districts.

Based on this 2005-09 data, the "ideal" (or average) size of a State Senate district would be 313,289 people (assuming the number of Senate districts remains at 62). In the State Assembly, the "ideal" (or average) size would be 129,493 people. In 2000, the average size of a State Senate district was 306,072 people, and the average size of an Assembly district was 126,510 people.

The Census Bureau's ACS population estimates for the 2005-09 period by legislative district (see the American FactFinder for access to this data) enables us to see how each current district's population compares with the expected "ideal" size based on the 2005-09 American Community Survey.

Mapping the data

In order to determine how the distribution of Senate and Assembly seats may change within regions and across the state, the Center for Urban Research mapped how far each district's 2005-09 estimated population differs from the "ideal" size. In New York State, legislative districts within plus or minus 5% of the ideal size are considered acceptable from an overall population perspective. (There are other demographic considerations that need to be addressed in redistricting, but this analysis focuses on total population.)

The maps below show which districts are within 5% of the ideal size in the State Senate and the Assembly, and which ones have a population 5% or more, 0r 5% or less, than the ideal size. Districts shaded red have "too many" people relative to the other districts, based on the 2005-09 ACS estimates, and would need to give up population to other districts. The districts shaded blue have too few people. Portions of them would need to be combined with others so that all districts are within +/- 5% of the target size.

State Senate


State Assembly

The following table summarizes these population estimates on a regional basis. It compares each of the three major regions of the state -- Long Island, New York City, and upstate -- regarding:

  • current number of districts,
  • current regional population based on the 2005-09 ACS (for the State Senate, one district in New York City extends into Westchester County, slightly modifying the regional population totals),
  • average district population size based on the 2005-09 ACS,
  • "ideal" district population size based on the 2005-09 ACS, and
  • a calculation of the expected number of districts (dividing the 2005-09 regional population total by the ideal district population size), and how this expected number differs from the current district distribution.

Here is a link to the spreadsheet [XLS]. Below are links to spreadsheets listing the district-by-district 2005-09 population estimates and the percent differences from the ideal size:

Other Resources

The office of New York State Senator Martin Malavé Dilan has posted several maps and data tables on "demographic trends across New York, including data on Congressional districts, State Senate districts, county population changes as well as prisoner populations from 2000-2009."

Cautionary Notes

This analysis is based on several assumptions that may change. In particular, the population data is based on a 5-year average from 2005 to 2009. The data do not reflect trends, and may differ substantially from the actual population counts based on the 2010 decennial Census. The 2010 decennial population data will be the official data that is used to determine "ideal" district size and eventual redistricting.

This analysis also does not account for a change in how New York's prison population will be counted for the purposes of redistricting. In 2010, NY Governor Paterson signed a law (Chapter Laws of 2010, Chapter 57, Part XX) to require New York's redistricting task force to reallocate prison populations back to verifiable “homes of record” where the prisoner resided prior to his or her incarceration for state legislative and local governmental redistricting. The task force will obtain the prison count population data from the New York State Department of Correctional Services. Until then, we are not able to accurately estimate the ideal district size based on this revised approach to counting the prison population.

Also, we assume (as noted above) that the total number of State Senate and Assembly districts remain the same (62 Senate seats, 150 Assembly seats). Finally, we assume that district boundaries in the next round of redistricting will respect the regions (LI, NYC, upstate) used in the tables above. If, for example, a Long Island Senate district is allowed to extend into New York City, or vice versa, the analysis above would need to change.