Outdated methodology impairs Census Bureau’s count of Black population

by Peter Wagner, May 3, 2004

Census 2000 showed a number of rural counties in the West, Midwest and Northeast that more than doubled their Black populations over the previous decade. Is this some sort of reversal of the great migration that saw millions of Blacks leave the rural South for Northern cities? Is a new economic opportunity drawing Blacks to leave cities for rural places? Not quite.

Most of the counties shown by the Census Bureau to have the fastest growing Black populations (see counties marked in purple in the first map below), are counties with new prisons with large incarcerated Black populations. (Compare with second map below.)

The Census Bureau counts incarcerated people as if they were residents of the prison town, even though prisoners have no contact with the outside community and are not there by choice. This methodology has staggering implications for how and where Black citizens are counted. On Census Day, 2.5% of Black Americans found themselves behind bars. Twelve percent of Black men in their 20s or early 30s are incarcerated. These figures are 7 to 8 times higher than the corresponding statistics for Whites. The Census Bureau’s method of counting the incarcerated disproportionately counts Blacks in the wrong place.

Census Bureau map showing growth of Black population in each county in the U.S. 1990-2000

Figure 1. Many counties (in dark purple) report significant Black populations doubling in size from 1990 to 2000. In many of these cases, the cause is not willing Black migration, but the construction of new prisons. (Map source: U.S. Census Bureau, Mapping Census 2000: The Geography of U.S. Diversity.)

Percent of Black population in each U.S. county that is incarcerated, excluding counties with less than 100 Black people incarcerated

Figure 2. Many counties report that a large percentage of their Black population is incarcerated. Actually, these Black populations live somewhere else but are counted as county residents by the Census because they are incarcerated in the county.

Assigning the incarcerated to the prison address is an outdated method of collecting data that reduces the value of Census information about the racial makeup of our communities.

Brown County, Illinois, is the most extreme example. According to the Census, Brown County is 18% Black, more than all but 4 Illinois counties. Yet all but 5 of the 1,265 Blacks reported by the Census in Brown County are incarcerated residents of somewhere else. The large Black population of Brown County is a statistical fiction.

Brown County may have more of its Black population behind bars than any other county in the country, but it is not a unique glitch in the data. We found 256 counties that that have more than a quarter of their Black population behind bars. In 173 counties, more than half of the Black population is incarcerated. (See table below for list of counties. These figures deliberately understate our results, as we excluded from the above an additional 75 counties that have less than 100 Blacks incarcerated.)

This does not necessarily reflect a bias in counties’ criminal justice systems; rather it reflects a geographic bias in where prisons are located. Prisons are often located far from the communities where most of the prisoners come from. By relying on the racial and geographic bias of the prison industry to determine the residence of the incarcerated, the Census introduces a racial and geographic bias into its data.

The Census Bureau has been counting people at the prisons since the first Census in 1790, long before the data was used for demographic analysis or legislative redistricting. The Supreme Court didn’t coin the “One person one vote” principle, or require that state legislative districts comply with its requirement to draw equally sized districts until 1964. Congress didn’t mandate the Census Bureau to assist with state redistricting until 1975. High incarceration itself is quite new, having interfered only with the last two U.S. Censuses in 1990 and 2000. (See Figure 3.) Times have changed.

graph of US incarceration rate 1925-2001 showing incarceration stable until after the 1980 census

Figure 3. The 1980 Census was the last Census taken before incarceration rates began to skyrocket.

As the country and its needs for data evolved, the Census has regularly updated its counting methodology for other special populations like students and the military. The Census Bureau’s slogan is “Helping you make informed decisions”. If the Census Bureau wants to help policy makers make informed decisions about the size and growth of the Black population of the United States, it needs to change how it counts the incarcerated.

 

Counties with more than 50% of their Black Population incarcerated (excluding counties with less than 100 Black people incarcerated)

County State 2000 population Black population Black population incarcerated Percent of Black population incarcerated
Graham Arizona 33,489 625 391 63%
Izard Arkansas 13,249 191 168 88%
Del Norte California 27,507 1,184 1,081 91%
Lassen California 33,828 2,992 2,728 91%
Amador California 35,100 1,359 1,052 77%
Tuolumne California 54,501 1,146 964 84%
Imperial California 142,361 5,624 2,939 52%
Logan Colorado 20,504 420 313 75%
Kit Carson Colorado 8,011 139 133 96%
Lincoln Colorado 6,087 302 268 89%
Delta Colorado 27,834 146 116 79%
Chaffee Colorado 16,242 257 245 95%
Fremont Colorado 46,145 2,464 2,357 96%
Crowley Colorado 5,518 389 376 97%
Bent Colorado 5,998 219 181 83%
Huerfano Colorado 7,862 216 173 80%
Holmes Florida 18,564 1,208 791 65%
Liberty Florida 7,021 1,294 678 52%
Lafayette Florida 7,022 1,009 689 68%
Union Florida 13,442 3,070 1,776 58%
Gilchrist Florida 14,437 1,010 540 53%
Habersham Georgia 35,902 1,610 829 51%
Tazewell Illinois 128,485 1,131 813 72%
Pike Illinois 17,384 260 251 97%
Clinton Illinois 35,535 1,391 1,037 75%
Randolph Illinois 33,893 3,147 2,139 68%
Fulton Illinois 38,250 1,378 1,250 91%
Brown Illinois 6,950 1,265 1,260 100%
La Salle Illinois 111,509 1,723 895 52%
Livingston Illinois 39,678 2,053 1,727 84%
Logan Illinois 31,183 2,045 1,581 77%
Lee Illinois 36,062 1,772 1,318 74%
Edgar Illinois 19,704 362 295 81%
Christian Illinois 35,372 758 641 85%
Montgomery Illinois 30,652 1,143 1,039 91%
Fayette Illinois 21,802 1,064 1,038 98%
Crawford Illinois 20,452 927 785 85%
Bond Illinois 17,633 1,306 850 65%
Perry Illinois 23,094 1,851 1,430 77%
Hardin Illinois 4,800 132 117 89%
Johnson Illinois 12,878 1,825 1,789 98%
Parke Indiana 17,241 370 276 75%
Hendricks Indiana 104,093 1,162 789 68%
Putnam Indiana 36,019 1,057 707 67%
Sullivan Indiana 21,751 928 891 96%
Perry Indiana 18,899 274 250 91%
Jones Iowa 20,221 361 308 85%
Jasper Iowa 37,213 309 200 65%
Henry Iowa 20,336 302 156 52%
Page Iowa 16,976 282 178 63%
Norton Kansas 5,953 241 236 98%
Ellsworth Kansas 6,525 232 217 94%
Lyon Kentucky 8,080 543 427 79%
Oldham Kentucky 46,178 1,943 1,025 53%
Boyd Kentucky 49,752 1,267 641 51%
Morgan Kentucky 13,948 611 593 97%
Floyd Kentucky 42,441 546 314 58%
Lee Kentucky 7,916 300 265 88%
Clay Kentucky 24,556 1,178 831 71%
West Feliciana Louisiana 15,111 7,633 3,877 51%
Washington Maryland 131,923 10,247 5,199 51%
Allegany Maryland 74,930 4,006 2,139 53%
Baraga Michigan 8,746 436 422 97%
Marquette Michigan 64,634 853 600 70%
Gogebic Michigan 17,370 305 287 94%
Luce Michigan 7,024 528 515 98%
Alger Michigan 9,862 603 593 98%
Schoolcraft Michigan 8,903 145 132 91%
Iron Michigan 13,138 144 117 81%
Chippewa Michigan 38,543 2,127 2,020 95%
Crawford Michigan 14,273 214 170 79%
Manistee Michigan 24,527 399 342 86%
Arenac Michigan 17,269 315 297 94%
Montcalm Michigan 61,266 1,330 1,127 85%
Gratiot Michigan 42,285 1,572 1,468 93%
Lapeer Michigan 87,904 720 521 72%
Ionia Michigan 61,518 2,807 2,642 94%
Branch Michigan 45,787 1,206 1,016 84%
Carlton Minnesota 31,671 308 251 81%
Pine Minnesota 26,530 341 241 71%
Sherburne Minnesota 64,417 550 303 55%
Swift Minnesota 11,956 322 306 95%
Rice Minnesota 56,665 741 396 53%
Waseca Minnesota 19,526 441 302 68%
Webster Missouri 31,045 359 287 80%
DeKalb Missouri 11,597 1,028 979 95%
Livingston Missouri 14,558 339 193 57%
Cooper Missouri 16,670 1,493 793 53%
Moniteau Missouri 14,827 561 482 86%
Pike Missouri 18,351 1,682 993 59%
Washington Missouri 23,344 578 459 79%
St. Francois Missouri 55,641 1,126 856 76%
Pershing Nevada 6,693 358 340 95%
White Pine Nevada 9,181 380 344 91%
Carson City Nevada 52,457 946 657 69%
Hunterdon New Jersey 121,989 2,743 1,607 59%
Torrance New Mexico 16,911 280 155 55%
Clinton New York 79,894 2,863 2,043 71%
Franklin New York 51,134 3,389 3,147 93%
Essex New York 38,851 1,092 960 88%
Washington New York 61,042 1,785 1,543 86%
Greene New York 48,195 2,664 1,705 64%
St. Lawrence New York 111,931 2,664 1,816 68%
Cayuga New York 81,963 3,272 1,855 57%
Orleans New York 44,171 3,230 1,710 53%
Seneca New York 33,342 758 508 67%
Livingston New York 64,328 1,938 1,247 64%
Wyoming New York 43,424 2,395 2,290 96%
Avery North Carolina 17,167 598 525 88%
Williams Ohio 39,188 283 206 73%
Marion Ohio 66,217 3,805 2,146 56%
Union Ohio 40,909 1,149 886 77%
Madison Ohio 40,213 2,511 1,792 71%
Noble Ohio 14,058 940 923 98%
Pickaway Ohio 52,727 3,391 2,907 86%
Hocking Ohio 28,241 259 161 62%
Warren Ohio 158,383 4,327 2,424 56%
Alfalfa Oklahoma 6,105 256 244 95%
Woods Oklahoma 9,089 216 123 57%
Woodward Oklahoma 18,486 204 159 78%
Beckham Oklahoma 19,799 1,098 762 69%
Hughes Oklahoma 14,154 634 348 55%
Greer Oklahoma 6,061 532 331 62%
Yamhill Oregon 84,992 721 367 51%
Malheur Oregon 31,615 387 289 75%
Wayne Pennsylvania 47,722 757 395 52%
Clearfield Pennsylvania 83,382 1,239 1,079 87%
Northumberland Pennsylvania 94,556 1,439 903 63%
Union Pennsylvania 41,624 2,878 2,499 87%
Schuylkill Pennsylvania 150,336 3,147 2,401 76%
Huntingdon Pennsylvania 45,586 2,342 1,873 80%
Somerset Pennsylvania 80,023 1,275 1,166 91%
Greene Pennsylvania 40,672 1,585 1,356 86%
McKean Pennsylvania 45,936 860 745 87%
Yankton South Dakota 21,652 252 167 66%
Johnson Tennessee 17,499 424 356 84%
Lake Tennessee 7,954 2,481 1,406 57%
Morgan Tennessee 19,757 440 412 94%
Hickman Tennessee 22,295 1,009 553 55%
Bledsoe Tennessee 12,367 458 280 61%
Wayne Tennessee 16,842 1,145 1,019 89%
Pecos Texas 16,809 738 645 87%
Hartley Texas 5,537 451 432 96%
Childress Texas 7,688 1,083 754 70%
Dickens Texas 2,762 226 171 76%
Dawson Texas 14,985 1,297 860 66%
Mitchell Texas 9,698 1,242 962 77%
Jack Texas 8,763 486 404 83%
Wise Texas 48,793 600 308 51%
Parker Texas 88,495 1,586 926 58%
Jones Texas 20,785 2,392 1,810 76%
Medina Texas 39,304 866 597 69%
Karnes Texas 15,446 1,667 1,377 83%
Frio Texas 16,252 792 743 94%
Live Oak Texas 12,309 301 271 90%
Bee Texas 32,359 3,203 2,641 82%
La Salle Texas 5,866 208 180 87%
Willacy Texas 20,082 439 323 74%
Buchanan Virginia 26,978 708 470 66%
Bland Virginia 6,871 288 216 75%
Russell Virginia 30,308 934 721 77%
Grayson Virginia 17,917 1,217 833 68%
Clallam Washington 64,525 545 308 57%
Mason Washington 49,405 587 383 65%
Walla Walla Washington 55,180 930 535 58%
Randolph West Virginia 28,262 302 184 61%
Monroe West Virginia 14,583 872 736 84%
Jackson Wisconsin 19,100 433 339 78%
Crawford Wisconsin 17,243 233 157 67%
Marquette Wisconsin 15,832 545 499 92%
Sheboygan Wisconsin 112,646 1,224 661 54%
Columbia Wisconsin 52,468 460 339 74%
Dodge Wisconsin 85,897 2,142 1,916 89%

Source: Rose Heyer and Peter Wagner, Too big to ignore: How counting people in prisons distorted Census 2000; Peter Wagner and Rose Heyer Thirty-Two Years After Attica: Many More Blacks in Prison but not as Guards; Bureau of Justice Statistics, Prison and Jail Inmates at Midyear 2000; and additional research by Rose Heyer and Peter Wagner.



Stay Informed


Get the latest updates:



Share on 𝕏 Donate