Fax to Cox from Carraway RE: Edited Joint Appendix

    State of North Carolina 

Department of Justice 
: P. O. Box 629 

MICHAEL F. EASLEY RALEIGH 
ATTORNEY GENERAL 27602-0629 

FAX TRANSMISSION 

10... etl Cow 
NZ / 

FAX NUMBER: | NO. OF PAGES: __ 5) 

  

  

  

FROM: _ "yr au SARE 
  

TELEPHONE NUMBER: (919) 716-6900 FAX NUMBER: (919) 716-6763 

SUBJECT: Luosaskee a (pen Li 
COMMENTS: i wan Chis. 

Ri = an 
Ch ifmedl tee. 

  

  

    

  

  

  

  
A 
2, 

CONFIDENTIALITY NOTE 

  

  

THE INFORMATION CONTAINED IN THIS FACSIMILE MESSAGE IS LEGALLY PRIVILEGED AND 
CONFIDENTIAL INFORMATION INTENDED ONLY FOR THE USE OF THE INDIVIDUAL OR ENTITY 
NAMED ABOVE. IF THE READER OF THIS MESSAGE IS NOT THE INTENDED RECIPIENT, YOU ARE 
HEREBY NOTIFIED THAT ANY DISSEMINATION, DISTRIBUTION OR COPY OF THIS TELECOPY IS 
STRICTLY PROHIBITED. JF YOU HAVE RECEIVED THIS TELECOPY IN ERROR, PLEASE IMMEDIATELY 
NOTIFY US BY TELEPHONE AND RETURN THE ORIGINAL MESSAGE TO US AT THE ADDRESS ABOVE 
VIA THE UNITED STATES POSTAL SERVICE. THANK YOU. 

\WPSINSPLIT\FORM\FAX.SHT 

&3 
10°d 80:01 100. ef In £9.9912676: ed 117 WIJ3dS 99 ON 

 



  

  

[caption omitted for printing] 

DECLARATION OF DR. RONALD F. WEBER 
  

I,"Ronald E. Weber, BPh.D., declare pursuant to 28 U.S.C. 

1746 as follows: 

1. I am currently the Wilder Crane Professor of Government 

in the Department of Political Science at the University of 

Wisconsin, Milwaukee, Wisconsin; President of Campaign and 

Opinion Research Analysts, Inc.; former co-editor of The Journal   

Qf Politics and Chairman of the Department of Political Science 

at the University of Wisconsin, Milwaukee; former Fulbright 

Commission John Marshall Professor of Political Science at the 

Budapest University of Economic Sciences and the Central European 

University, Budapest, Hungary (1996-97); and former President of 

the Southern Political Science Association (1997-98). I received 

my B.A. in Political Science and History from Macalester College, 

St. Paul, MN, . in 1964 and a Ph.D. in Political Science from 

Syracuse University in 1969, with specialties in American state 

politics, voting behavior, and quantitative analyses of political 

data. A copy of my curriculum vitae is attached as Exhibit A. 

* ok 

3. I have been retained as a consultant and expert witness 

in a number of redistricting and voting rights cases and have 

¢0'd 80:01 (QQ. oC I[ 9.991616: Xe ER ol 

 



been qualified as an expert by the U.S. District Courts in [22 

instances] . . . I have testified in a number of Congressional 

and state legislative redistricting cases . .'. . 1 also have 

extensive experience developing redistricting plans for local and 

state government clients and assisting them with preclearance of 

those plans under Section 5 of the U.S. Voting Rights Act of 

1965, as amended in 1982. 

I address the following questions 

(1) whether race was the predominant factor used by the 

state of North Carolina to draw the boundaries of the 

1297 U.S. Congressional districts: 

(2) whether the state of North Carolina in creating the 

U.S. Congressional ‘districting plan of 1997 

subordinated traditional race-neutral districting 

principles, such as compactness, contiguity, respect 

for political subdivisions or communities defined by 

actual shared interests, to racial considerations; 

woo 

(5) whether African-American voters residing in 

Districts il and 12 inithe Act 588 plan of 1997 

participate at lower rates than white voters in recent 

state-wide elections, indicating some evidence that a 

11 WIZ48 5H  



  

history of official discrimination has led to 

politically significant differences in political 

participation in the districts as drawn in 1997; 

(6) whether sufficient levels of white crossover voting 

ex1sts in the northeast and Piedmont regions of North 

Carolina such that fair U.S. Congressional districts 

can be drawn that do not need to be majority African- 

American in voting age population or voter registration 

in order to allow African-American voters a reasonable 

opportunity to elect candidates of choice in U.S. 

Congressional elections; and 

(7) whether race-predominant U.S. Congressional 

Districts 1 and 12 in the 1997 North Carolina plan are 

overly safe from the standpoint of giving a candidate 

of choice of African-American voters an opportunity to 

be elected, thus questioning whether the plan was 

narrowly tailored to satisfy a compelling state 

interest. 

€. The results of my analysis to date will be presented in 

this declaration in the following form: in Section I, I will 

outline briefly the history of Congressional districting in North 

Carolina since 1960; in section II, I will describe the analyses 

conducted to answer the first two questions and set forth my 

70d 60.01 00. C17 $929912616: Xe LI Wide od iy 

 



  

conclusions on those questions; . . . and in section V, VI, and 

VII, I will discuss participation rates of African-American and 

white voters as well as white crossover rates in recent statewide 

elections within two regions of North Carolina and the electoral 

safeness of North Carolina Congressional districts 1 and 12. 

Tables, charts, and exhibits relevant tc my analyses will be 

included within the body of the declaration or as attachments to 

this declaration. 

I. HISTORY OF RECENT CONGRESSIONAL REDISTRICTINGS 
  

IN NORTH CAROLINA 
  

10. As was true following the 1370 Census of Fopulation, 

the 1980 Census of Population revealed that the 11 districts of 

the 1970s were out of population balance. Thus, the North 

Carolina General Assembly had to adjust the populations of the 11 

SU °d 6:07 - 00, S21 $949912616: Xe 4 117 WITS 9d ON 

 



  

  

districts before the 1982 elections. The first plan adopted in 

July, 1981 did not receive pre-clearance under Section 5 by the 

U.S. Department of Justice. The General Assembly followed-up 

with a revised plan that satisfied the Department of Justice’s 

objections in a special session of February, 1982. " For the first 

time in the modern history of North Carolina, it was necessary to 

split four counties in order to balance the populations across 

the districes, Avery, Johnston, Moore, and Yadkin counties were 

each split across two districts. The town of Chapel Hill as well 

as the city of High Point were each split across two districts 

owing to the fact that those two places cross county lines and 

the General Assembly decided to draw the Congressional districts 

using county lines between Orange and Durham counties and between 

Guilford and Randolph counties. These 11 districts were used in 

the elections of 1982-1990. 

12 
L5, I conclude this section by making several observations. 

First, the sub-dividing of counties to achieve equally populated 

Congressional districts in North Carolina is a relatively recent 

occurrence, taking place for the first time with the splitting of 

Just four counties in the early 1980s. Second, no county in 

North Carolina is large enough that it must of necessity be sub- 

divided to comply with the principle of "one-person, one-vote”. 

Mecklenburg County, the largest county in population in North 

90 'd BO: 0% 0, BT TL $9991L616: xe 117 T1348 23 OH 

 



  

Carolina, is slightly smaller than a current Congressional 

district. Third, . . . at the maximum a total of eleven North 

Carolina counties might need to be split to create an equally 

populous twelve district plan, by using whole counties to create 

Congressional districts and then splitting just one county to 

balance the population between each of two districts. Fourth, 

the percentage of African-American persons in the North 

Carolina population declined between 1980 and 1990. 

Finally, as I will consistently point out below, the appropriate 

social science benchmark for comparison of the challenged plan is 

the plan ¢f the 1980s (with 11 districts) and not the 

constitutionally invalidated 12 district plan of the 1990s. 

LL. ANALYSIS OF DISTRIGCTING CRITERIA FMPLOYED BY STATE   

14. The question of whether race was the predominant factor 

used by the state of North Carolina to draw the boundaries of the 

U.S. Congressional districts in 1997 can be addressed by an 

examination of both tabular data prepared by the North Carolina 

Information Systems Division from data compiled by the U.S. 

Bureau of the Census and maps created on the computer facilities 

of the North Carolina General Assembly Legislative Services 

Office Redistricting System. Data from both sources are reported 

20°d BOOT 00. wir 9.991.616: x84 117 W1J345 ad IN A A 

 



in tabular form and on maps to display the use of race as a 

redistricting criterion. 

15. . . . . Table 1 reports for each 1997 North Carolina 

Congressional district the racial composition of the total 

population of the counties that were sub-divided in the creation 

of the plan using data from the County Split Assignments Report 

of March 26, 1997 of the North Carolina Information Systems 

Division. For both Districts 1 and 12 that were created to elect 

an African-American member of Congress, the county splits show a 

typical pattern of African-American total population majorities 

in the largest jurisdictions of each district. 

LIA 34S ad ON  



  

TABLE 1 

Prepeoxtion of Total Populaticn in Spllt Counties 
by Congressional District for 1997 Flan (HB 586--Plan A) 

(abridged) 

African- 
County /City cD Total White 2 auer. 

Forsyth 5 206,766 181,381 87.1 22,297 11.1 

Davidscn e 59,003 57,135 65.2 2,468 4.1 
Guilford 6 211,363 186,33) "83.2 21,541": 10.2 
Rowan 6 77,499 70,819 91.4 5,979 2.3 

Mecklenburg 8 282,808 264,604 90.4 21,0286 7:2 

Iredell 10 54,472 48,438 B&.9S Ne 526. 10.3 

Davidson 12 66, 684 26,161 384.2 0,846 14.8 
Forsyth 12 59,112 15,537 «25.3 43,105 172.9 
Guilford 2 13¢,057 63,253 45.5 90,114: 81.5 
Iredell 12 38,459 28,769 74.8 9,343 24.3 
Mecklenburg 12 218,625 | 100,047) 45.9% 113,442 51.9 
Rowan 12 33,106 21,032 "e3.% 11,754 35.6 

8 

60d 00:07 00a IT[ $9.99T.6T16: Xe 4 JIT WHIT3HS Sid ON 

 



  nm 4a Woh aan as te 

  

* x x 

17. Turning next to District 12, the racial make-up of the 

parts of the six sub-divided counties assigned to District 12 

include three with parts over 50 percent African-American and 

three in which the African-American percentage is under 50 

percent. Almost 75 percent of the total population in District 

12 comes from the three county parts which are majority African- 

American in population. Mecklenburg, Forsyth, and Guilford 

counties which contribute almost 75 percent of the district’s 

total population are located at the extremes of the district. 

The other three county parts (Davidson, Iredell, and Rowan) have 

narrow corridors which were designed to pick up as many African- 

American persons from each of those counties to fill out the 

district to an ideal sized district. A precinct level map of 

District 12 shows that all African-American majority precincts 

but one in those three counties have been assigned to the 

district. Conversely, Table 1 reports that the African-American 

percentage of the total population in the parts of those split 

counties assigned to another district than District 12 is 

consistently lower. For example, the part of Mecklenburg County 

assigned to District 9 is 7.2 percent African-American in total 

population, while the part of Forsyth County allocated to 

District 5 is 1l.1 percent African-American in total population 

and the part of Guilford County assigned to District 6 is 10.2 

01d 80:01 a0, 22.170 2949912616: ~& 117 HIT Si ON 

 



  

  

percent African-American. The county splits as they impact the 

white majority districts adjacent to District 12 in the Piedmont 

are those county splits for the districts where the intent 

usually was to carve out African-American voters so as not to 

endanger the electoral bases of the candidates who might not be 

characterized as candidates of choice of minority voters (e.g. 

Districts 5, 6, and 9). Almost every time there was an 

opportunity to use race as the basis for dividing political 

subdivisions up politically, the North Carolina Congressional 

districting plan does it in the Piedmont part of the state as 

well. 

18. ‘Table 2 provides further confirmation that race was the 

predominant factor in the drawing of the lines for the North 

Carolina Congressional districts in 1997. This table reports the 

exact same data as in Table 1 except that in Table 2 the data are 

organized by county rather than Congressional district. 

When counties were split to achieve population equality, the 

racial composition of the components differ little. When the 

splits are for racial purposes, the differences are large. 

10 

11d 60:0 QO. 9 Nf $9.9912616: <&4 117 W345. 50 

 



  

proportion or Tetald Population in Split Counties 

by County for 19897 Congressional Districting Plan (HB 586--Plan A) 

[abridged] 

African- 

ik 032) Total White kd Amer, 3 

Davidson 6 59,9932 57,135+905.2 2,468 4.1 

Davidson 12 66, 684 56,161 B4.2 0,846 14.8 

Forsyth : 206,768 181,381 "817.7 22.987 231.1 

Forsyth 12 59,132 15,5837 26.3 43,105 "72.9 

Guilford 6 211.363 "385,331 88.2 21,541 0.2 

Guilford 12 136,057 $3,253 48.5 70,334. - 2%.5% 

Iredell 10 54,472 48,438 88.9 S$, 526 10.1 

Iredell 12 38,459 28,769 74.8 9,343 24.3 

Mecklenburg g 292,808 264,604 90.3% 21,026 SR 

Mecklenburg 12 238,825 "i100,047 45.9 113,442 51.8% 

Rowan 6 77,498 70,819 91.4 5,979 7.7 

Rowan 12 23,106 21,032 :63.% 11,794 25%.6 

1: 

ZT °d 60:01 00. oC INL £9.991070: xe 4 LIN HI Eds 94 ON 

 



39. The pattern shown in each of the six counties that are 

split between District 12 and an adjacent district is one in 

which most of the time the sub-division was along racial lines. 

All six counties were split aleng racial lines. The most 

dramatic examples from Table 2 include Forsyth County where 72.9 

percent ¢f the total population allocated to District 12 is 

African-American while only 11.1 percent of the total population 

assigned to District 5 is African-American, and Mecklenburg 

County where 51.9 percent of the total population allocated to 

Pistrict 12 is African-American while only 7.2 percent of the 

total population assigned to District 9 is African-American. 

Similarly, I find Guilford County where 51.5 percent of the total 

population allocated to District 12 is African-American while 

only 10.2 percent of the total population assigned to District 9 

is African-American. A similar pattern holds in the other three 

counties Of District 12. lInieach of those counties, the 

population on the District 12 side of the Congressional di 

line 1s more strongly African-American while being more strongly 

white on the other side of the line in an adjacent district. 

When counties were split to achieve population equality, the 

racial composition of the components differs little. When the 

splits are for racial purposes, the differences are large. 

I can infer from these data that race was a predominant factor in 

the line drawing for Districts 1 and 12 and the adjacent 

01:03. O00. 92 Inf ‘929914616: X 117 WI3348 BE JN  



  [LT — 

  

districts in the 1997 North Carolina Congressional district plan. 

vei. ee 

21. I have also examined data related to city and town 

splits in 199% North Carolina Congressional district plan. 

Using a report titled Places Split by 97 North Carolina 

congressional Districts prepared by Dan Frey, GIS Analyst, with 

the North Carolina General Assembly Information Systems Division, 

I created Tables 3 and 4. 

13 

¥1°d D1:01 00. 92 11 £9.991L616: Xe lI WI 345 90 ! 

 



  

TABLE 3 

Proportion of Total Population in Split Municipalities and Census 
Designated Places by Congressional District for 1997 Plan (HE 586--Plan A) 

irvy/ Tow, 

High Point 
Winston-Salem 

Greensboro 

High Point 
Lexington 

Salisbury 
Spencer 

Thomasville 

Charlotte 

Cornelius 

Davidson* 

Mcoresville 

Statesville 

Troutman 

Charlotte 

Cornelius 

Davidson* 

Greensboro 

High Point 

Lexington 
Mooresville 

Salisbury 

Spencer 

Statesville 

Thomasville 

Troutman 

Winston-Salem 

5 
n
n
 

aN
 

O
Y
 

O
Y
 

ON
Y 

O
Y
 

OW
 

9 

10 
10 
10 
10 

12 
12 
12 
12 
12 
12 
12 
12 
12 
12 
12 
12 
12 

308 

182,419 
2,273 
4,046 

95,080 
32,290 
13,6%¢ 
8,818 

17,837 
3,211 
5,243 
9,006 

74 
54,270 

[abridged] 

73,981 
32,833 
2,522 
4,442 

6 
6,249 

186,112 
304 

0 
0 

3,997 
1,024 

73,535 
1,752 
3,407 

42,236 
15,677 
8,143 
6,687 

10,522 
2,488 
1,2%0 
5,246 

74 
12,272 

Ow
 
O
 

B
O
 

O
y
 

Es
 
W
W
 

~J
 

Ww
 

*City or town is split across a county boundary. 

Source: North Carolina General Assembly, 

N
=
 
O
O
 

A
O
 
W
A
H
L
 

O
D
O
C
 
o
H
 

N
D
F
 

W
U
 

African- 

17, 

—
 

oO
 

(a
e)
 

oo
 

~~
 

~
 Ww
 

38
) 

~J
 

=
 

On
 

~ [e
e]

 

Ne
Y 

N
a
 

~
 
N
E
 

Oy
 

BN
O 

-
 

-
~
 

-
 

-
 

w
w
 

~
~
 

un
 

O
M
 

=
 
w
n
 

O
O
O
 

W
L
I
o
y
 

W
w
 

~ 
~ 

~J
 

Ww
 

41,99 

W
w
 

a
 
w
o
o
 

D
O
J
 

0
0
O
 
N
M
S
 

o
y
 

W
w
 

Ul
 

Information Systems Division. 
All information is based on that in the General Assembly’s 
apportionment system. 

14 

Provided by Dan Frey, GIS Analyst. 

117 WI1J34S 98 ON 

 



  

TABLE 4 

Proportion OI Total Population 1n Split Municipalities and Census 

Designated places by City or Town for 13997 Plan (HB 586--Plan A) [abridged] 

City/Town 

Charlotte 

Charlotte 

Cornelius 

Cornelius 

Davidson* 

Davidson” 

Greensbor¢ 

Greensboro 

High Point 
High Point 

High Point 

Lexington 

Lexington 

Mooresville 

Mooresville 

Salisbury 
Salisbury 

Spencer 

Spencer 

Statesville 

Statesville 

Thomasville 

Thomasville 

Troutman 

Troutman 

Winston-Salem 

Winston-Salem 

co 

9 
32 

9 
12 

Tota) 

213,815 
182,419 

308 
2+,213 

0 
4,046 

88,441 
25,080 

6 
37,200 
32,290 

2,885 
13,696 

0 
8,818 

3,211 

12,324 
58,243 

74 
1,418 

89,215 
54,270 

White 

196,172 
312,935 

304 
1,752 

0 
3,407 

78,981 
42,236 

6 
32,833 
15,677 

2,322 
S,142 

0 
&, 687 

4,442 
10,521 

© 
2,488 

9,857 
1,290 

6,249 
5,246 

74 
1,024 

74,885 
12,272 

3 

100. 
72. 

83. 
22. 

*City or town is ‘split across a county boundary. 

Source: North Carolina General Assembly, 

African- 
Amer. 

17,343 
108,484 

14,330 
41,998 

3 

w
r
 

0
 

OO
 

o
p
 

Information Svstems Division. 
All information is based on that in the General Assembly’s 
apportionment system. 

9 IMC 

13 

$9L99T2616: Xe 

Provided by Dan Frey, GIS Analyst. 

117 BIJ34S 98 ON 

 



  

23. A similar pattern of splitting cities or towns is shown 

for District 12. Nine of 13 cities or towns were split along 

racial lines to create Congressional district 12. Five of the 

Cities or towns split between district 12 and another district 

involve placing a majority of the African-American population 

into District 12 as displayed in Table 3. Also the four largest 

cities assigned to District 12 are split along racial lines. 

24. The above analysis is further confirmed by the listing 

of split cities and towns in Table 4. iii tant, Takle 4.also 

highlights the racial splits of the cities or towns assigned to 

District 12. Again the difference in the figures for Charlotte, 

Greensboro, High Point, Statesville, and Winston-Salem are large. 

On the other hand, many of the cities and towns split between 

other districts do not display large racial differences 

(exceptions are Fayetteville and Raleigh). 

pS RS 

26. The discussion of not splitting precincts by 

defendants’ experts misunderstands how racially homogeneous 

precincts are today in North Carolina and other parts of the 

nation. . .... i. louisiana'did not split a single precinct in the 

creation of the two plans invalldated by the Hays court in the 

16 

21d 03:00 00, 92 170 $94991.616: <e4 117 WIJ34S 98 ON 

 



Western District of Louisiana. Thus, given the homogeneous 

racial character of precincts in North Carolina, it i$ guite 

possible to draw districts in which race predominates using whole 

precincts. 

27. Next I present Table 5 which details at the precinct 

level by county the assignment of precincts to Congressional 

districts for the 22 counties that were split in creating the 

1997 Plan (HB 586). For the counties which were split 

predominantly on the basis of race I expect that the precincts 

with the highest percentage of African-American persons will be 

placed in either District 1 or 12 and the precincts with the 

highest percentage of non-African-American persons will be placed 

in an adjacent district. For the counties which were not split 

predominantly on the basis of race, I expect to find a less 

systematic assignment ¢f precincts to the two districts into 

which the population is divided. To assist the reader in 

following the tables, I have ranked the precincts from high to 

low by the percentage of African-American population in each 

precinct (see Table 5). The table also reports a breakdown of 

the voting age population by race as well as the total population 

by race. Finally, the district to which the precinct is assigned 

in the 1397 Plan is noted in the final column of the table. 

[Eran: insert Table 5 here] 

II WHIE4S 95  



  

39. . . . . The patterns revealed in Tables 1-4 are 

reinforced by the precinct level data in Table 5 for the six 

Split counties of District 12. Two groupings of counties are 

most apparent: those where the difference between African- 

American percentage of the total population assigned to District 

12 is greater than 20 percentage points higher than the African- 

American percentage assigned to the adjacent district and those 

where the difference in the African-American percentage is less 

than 20 percentage points. Forsyth, Guilford, Mecklenburg, and 

Rowan counties fall into the former category, while Davidson, and 

Iredell counties are in the latter category. 

40. I first examine the precinct assignment pattern for 

Forsyth County which includes the city of Winston-Salem and where 

the most severe of racial disparities exist between Districts 5 

and 12. All 15 African-American majority precincts in the county 

are assigned to District 12, while all but four of the white 

majority precincts in Forsyth County are assigned to District 5. 

No precinct in which more than 20.39 percent of the total 

population is African-American is allocated to District.5. - The 

assignment of Forsyth County precincts between Districts 5 and 12 

18 

61d 01:01 00. 92 Nf £9.991616: xk 4 117 WHI134% 90 JN 

 



  

1s the clearest pattern of the predominant use of race in the 

creation of District 12 in the 1997 Plan in the Picdmont region 

of: North Carolina. 

41. I next look at the precinct assignment pattern for 

Mecklenburg County which includes the city of Charlotte and where 

very severe racial disparities exist between Districts 9 and 12. 

All 28 African-American majority precincts in the countv are 

assigned to District 12, while all the precincts where around 85 

percent or more of the population is white in Mecklenburg County 

are assigned to District 9. Some white majority precincts are 

allocated to District 12, but for the most part these assignments 

are made to provide contiguous paths from the African-American 

majority precincts north to Iredell County so as to connect 

African-American populations in Charlotte with those in Winston- 

Salem, High Point, and Greensboro. The assignment of Mecklenburg 

County precincts between Districts 9 and 12 shows a very clear 

pattern of the predominant use of race in the creation of 

District 12 in the 1997 Plan in the Piedmont region of North 

Carolina. 

42. I next examine the precinct assignment pattern for 

Guilford County which includes the cities of Greensboro and High 

Point and where severe racial disparities exist between Districts 

6 and 12. All 17 African-American majority precincts in the 

1° 

0c °d 01:01 O00, SST 9.991.676: xB 117 WII348 St TN 

 



  

county (all of which are in either Greensboro or High Point) are 

assigned To District 12, while almost all of the precincts where 

around 80 percent or more of the population is white in Guilford 

County are assigned to District 6. Some white majority precincts 

are allocated to District 12, but for the most part these 

assignments are made to provide contiguous paths from the 

African-American majority precincts in Greensboro and High Point 

southwest to Davidson County so as to connect African-American 

populations in Greensboro and High Point with those in Winston- 

Salem and Charlotte. The assignment of Guilford County precincts 

between Districts 6 and 12 is another clear pattern of the 

predominant use of race in the creation of District 12 in the 

1297 Plan in the Piedmont region of North Carolina. 

43. I next examine the precinct assignment pattern for 

Rowan County dncluding the city of Salisbury which is split 

between Districts 6 and 12. Three of the four African-American 

majority precincts in the county (the one African-American 

majority precinct assigned to District 6 is a small non- 

contiguous part of a large white majority precinct) are assigned 

to District 12, while almost all of the precincts where around 80 

percent or more of the population is white in Rowan County are 

assigned to District 6. Some white majority precincts are 

allocated to District 12 to provide contiguous paths from Iredell 

County to Davidson County so as to connect African-American 

20 

18d 01:01. 00. 3%. 17C $9.991.616: <e4 117 BIZ34S 98 ON 

 



  

population concentrations in Charlotte with those in Winston- 

Salem, High Point, and Greensboro. The assignment of Rowan 

County precincts between Districts 6 and 12 is another clear 

pattern of the predominant use of race in the creation of 

District 12 in the 1997 Plan in the Piedmont region of North 

Carolina. 

44. Turning next to the two counties split between 

Congressional District 12 and an adjacent district where the 

difference in the African-American percentage of the total 

population is less than 20 percentage points, I examine first 

Iredell County. Only nine precincts in this county are assigned 

tC Congressional District 12, including two African-American 

majority precincts in the city of Statesville. The white 

majority precincts allocated to District 12 are there to provide 

contiguous paths from Mecklenburg County to Rowan County so as to 

connect African-American population concentrations in Charlotte 

with those in Winston-Salem, High Point, and Greensboro. The 

assignment of Iredell County precincts between Districts 10 and 

12 provides further evidence of a pattern of the predominant use 

Cf race in the creation of District 12 in the 1997 Plan in the 

Piedmont region of North Carolina. 

45. I next examine the assignment of precincts between 

Congressional Districts 6 and 12 in Davidson County. A total of 

21 

A 11:01 = Cl
i Je RTI 9.991616: <k 1175001 34S id TI 

 



20 precincts in this county are assigned to Congressional 

District 12, including two African-American majority precincts in 

the city of Lexington and cone in the city of Thomasville. The 

white majority precincts allocated to District 12 are there to 

provide contiguous paths from Rowan County to Forsyth and 

Guilford counties so as to connect African-American population 

concentrations in Charlotte with those in Winston-Salem, High 

Point, and Greensboro. The assignment of Davidson County 

precincts between Districts 6 and 12 provides further evidence of 

a pattern of the predominant use of race in the creation of 

District 12 in the 19397 Plan in the Piedmont region of North 

Carolina. 

47. veil xi "The state Of North Carolina subordinated the 

splitting of county, city, and town boundaries to a desire to 

allocate persons by race to a greater extent than necessary to 

comply with the one-person, one-vote standard of population 

equality. "This was particularly true in the drawing of Districts 

l and 12 as well as adjacent districts. A total of 22 counties 

and 41 cities and towns were split in the drawing of the 1997 

North Carolina Congressional districting plan, for a total of 63 

split political subdivisions. Only six counties and 15 citles 

and towns were split to create the other districts (several of 

22 

117 WBI345 SON  



  

  

the cities and towns were split because they are on a county 

boundary and diffezent countics were assigned to different 

districts). In the recent past no counties had been split to 

create the 11 districts of the 1960s and 1970s, while just four 

had been split to construct the 11 districts of the 1980s. The 

maximum number Of counties needed to be split to fashion a 12 

district plan is 11, allowing for one county to be split between 

each two districts. 

48. A report by David C. Huckabee, "Congressional 

Districts: Objectively Evaluating Shapes," CRS Report for 

congress (Washington: Congressional Research Service, The Library 

of Congress), May 24, 1994, also contains information from other 

states to compare with the experience of North Carclina in 

Congressional districting. 

49. Huckabee also provides information on the percentage of 

split counties allocated to the Congressional districts of the 

1990s. There are a large number of Congressicnal districts from 

around the nation which have 100 percent of the counties split 

which are allocated to a district. Most of these = plans 

involving splits of large counties in the metropolitan areas of 

the country and these plans are not comparable to the North 

Carclina setting. No single district in the country is like 

North Carolina 12 in splitting as many as six counties and sub- 

23 

ved 30:01 009 In $9.9912616: Xe 117 WIJ34S 98 ON 

 



dividing 100 percent of them. 

  

£ 50. Huckabee also provides information on the number of 

places having populations of 10,000 or more and indicates how 

many 0f these are split by district lines. In the 1997 plan, 

District 12 has eight cities or towns 10,000 or more in 

population, and all eight of them are split between districts. 

Ne oe ide 

53. A second way to assess the compactness of a 

Congressional district is to use a variety of compactness 

measures now standard in political science. These measures are 

reported upon in two works~-1) Richard H. Pildes and Richard G. 

Niemi, "Expressive Harms, 'Bizarre Districts,' and Voting Rights: 

Evaluating Election-District Appearances after Shaw," Michigan   

  Law Review, Vol. 92 (December 1993), pp. 101-205, and 2) David C. 

Huckabee, "Congressional Districts: Objectively Evaluating 

Shapes," CRS Report for Congress (Washington: Congressional 
  

Research Service, The Library of Congress), May 24, 1994. The 

actual computations of the compactness scores reported in these 

two works were conducted by Kimball Brace and Douglas Chapin of 

Election Data Services, Inc., Washington, D.C. 

54. Three measures of compactness are reported in the 

Huckabee report for all 435 Congressional districts in the U.S. 

adopted following the release of the 1990 Census of Population. 

24 

Sl d $1200. 200,92 {70 99914616: <& 4 117 WIJ34S 98 ON 

 



  

Huckabee adopts two geographic measures--a dispersion measure and 

a perimeter mcasure--—and one population measure. 

55. . .... North Carolina's Congressional District 12 is 

8till the least compact district in North Carolina on both 

indicators of ‘geographic compactness . . . .. In Appendix Ei of 

the Huckabee report, he reports a table containing the two 

geographic compactness scores for the bottom ten percent of 

Congressional districts in the nation. Using the criterion of 

having at least one compactness score in the bottom ten percent, 

North Carolina 12 would clearly continue in that compilation 

I have also taken the perimeter and dispersion compactness 

scores from the Huckabee report and revised the ranking order for 

the worst districts on each measure . . . . North Carolina 12 

ranks either 430 or 431 out of 435 in compactness using the 

dispersion measure . . . . North Carolina 12 ranks either 432 or 

433 of 435 in compactness using the perimeter measure. Thus, 

North Carolina 12 continues to be the least compact district in 

North Carolina and among the worst in the nation in terms of 

geographical compactness. 

56. Pildes and Niemi report geographic compactness scores 

for the Congressional districts of the 1980s using the dispersion 

and perimeter measures. These scores for the old 11 districts in 

North Carolina are used to compare to the scores for the current 

25 

$d TROT 0, STI £999T2616: Xe 117 WIJ3dS 99 IN 

 



12 districts in the 1997 plan.! The range on the dispersion 

  

measure in North Caxoclina xuns from a low of .26 to a Righ.of i .57 

and averages .36 across the 11 districts in the 1980s. 

The range on the perimeter measure in North Carolina runs from a 

low of .22 to a high of .46 and averages .30 across the 11 

districts in the 1980s. . 0 This comparison reveals that 

the 1397 12 district plan is less compact overall than the 11 

district plan of the 1980s, using the two standard measures of 

geographic compactness. 

411. EVALUATION OF THE POLITICAL DEFENSE OF DISTRICT 12 
  

* Uk 

£4. 0.0 ymin, I have constructed a set of tables which 

cross-tabulate the racial makeup of the population of each 

precinct by county with the Democratic registration percentage, 

the percentage of support for Democrat Harvey Gantt in the 1990 

U.S. Senate contest, and the percentages of support for the 

Democratic candidates for Lieutenant Governor and Court of 

Appeals Judge in the 1988 general election. This cross- 

tabulation will reveal that all African-American majority 

population precincts are assigned to District 12 from the six 

    

! Richard H. Pildes and Richard G. Niemi, "Expressive 
Harms, 'Blzarre Districts,' and Voting Rights: Evaluating 
Election-District Appearances after Shaw," Michigan Law Review, . 
Vol. 92 (December 1993), Table 6, pp. 189-91. 

26 

TATE LE RL £9L99TLETE: XE L117 WI234S 98 IN 

 



  

countles and that many of the precincts carried by Democratic 

candidate= in the two contests of 1988 and the one contcsoct of 

1990 were not necessarily assigned to District 12. 

65. These results are displayed in Table 6 for each of the 

ix counties and for Congressional District 12 as a whole. All 

four Davidson County precincts more than 30 percent African- 

American are in Congressional District 12. The weakest support 

for a racial explanation comes from the 1930 Gantt table as four 

of the five precincts in which he polled a majority were placed 

in Congressional District 12. Charlotte’s former mayor ran 

poorly in Davidson County, being held below 30% of the vote in 23 

of the 41 precincts. Table 6 for Davidson County also 

demonstrates that most of the precincts less than 30 percent 

African-American in population but in which most voters are 

registered Democrats were excluded from Congressional District 

12. Analyses of the contests for lieutenant governor and Court 

of Appeals show that no precinct in which the population was less 

than 30 percent African-American but which cast more than 60 

percent of their ballots for the Democratic candidate are in 

congressional District 12. 

117 LES Si ON  



  

TABLE 6 

  

DAVIDSON COUNTY 

[omitted for printing) 

FORSYTH COUNTY 

Precincts included in Congressional District 12 when classified by measures of Democratic 
strength and African-American Population 

% Democratic Registration 

% Black Population 50-599 60 - 69.9 > 70% 

<30% 1/32 0/5 0 

30-399 0/1 Va 0/1 

40-499 0 0/1 2/2 

50 -59.9 0 0 2/2 

60 - 69.9 0 0 0 

> 70% 0 0 13/13 

% Support for Harvey Gantt, 1990 

% Black Population 50-59.9 60-699  >70% 

<30% 0/7 0 0/1 

30-399 02 1/1 0/1 

40-499 22 0 0/1 

50-399 0 0 2/2 

60-699 0 0 0 ; 

28 

&C d 21:01 OU, 98 17[ $9.991616: Xe 4 117 WIJ34S 98 ON 

 



> 70% 0 0 13/13 

  

% Support for Democratic Lt. Governor Candidate, 1988 

% Black Population 50-599 60 - 69.9 > 70% 

< 30% 0/8 0/1 0 

30-399 0/2 1/1 0/1 

40 - 49.9 0 2/3 0 

50-599 0 2/2 0 

60 - 69.9 0 0 0 

> 70% 0 0 13/13 

% Support for Democratic Court of Appeals Candidate, 1988 

% Black Population 50 -59.9 60 - 69.9 > 70% 

< 30% 0/6 0/1 0 

30-39.9 0/2 Va 0 

40 - 49.9 0 2/3 0 

50-599 0 1/1 1/1 

60 - 69.9 0 0 0 

> 70% 0 0 13/13 

GUILFORD COUNTY 
[omitted for printing] 

IREDELL COUNTY 
[omitted for printing] 

MECKLENBURG COUNTY 
[omitted for printing] . 

29 

0g "d 21:07 00. 92 INC $9.991.616: Xe 4 LIT Wii3dS SN 

 



  

ROWAN COUNTY 

[omitted for printing] 

SUMMARY FIGURES FOR ALL SIX COUNTIES IN CONGRESSIONAL DISTRICT 12 

Precincts included in Congressional District 12 when classified by measures of Democratic 
strength and African-American Population 

% Black Population 

<30% 

30-39.9 

40-499 

50 - 59.9 

60 - 69.9 

> 70 

% Black Population 

<30% 

30- 39.9 

40 - 49.9 

50-59.9 

60 - 69.9 

> 70 

% Democratic Registration 

50 - 59.9 

30/139 

2 

60 - 69.9 

18/45 

8/9 

3/6 

> 70% 

1/1 

4/6 

5/5 

3/8 

9/9 

48/48 

% Support for Harvey Gantt, 1990 

50-599 

13/56 

6/8 

4/4 

1/1 

0 

0 

30 

60 - 69.9 

312 

5/5 

0/1 

> 70% 

3/6 

0/1 

4/5 

5/5 

9/9 

48/48 

1 Xe 117. 010348 2 ON 

 



    

% Support for Democratic Lt. Governor Candidate, 1988 

  

% Black Population 50 - 59.9 60 - 69.9 > 70% 

< 30% 18/45 2/12 0 

30-399 7/9 4/5 0/1 

40 - 49.9 2/3 4/5 2/2 

50-599 3/3 6/6 1/1 

60 - 69.9 0 1/1 8/8 

> 70 0 1/1 47/47 

% Support for Democratic Court of Appeals Candidate, 1988 

% Black Population 50-59.9 60 - 69.9 > 70% 

<30% 13/32 2/6 0 

30-399 7/9 3/5 0 

40 - 49.9 2/2 6/6 0 

50 - 59.9 4/4 3/3 3/3 

60 - 69.9 0 1/1 8/8 

> 70 0 2/2 46/46 

31 

Jg¢'d 21:07. 00. $9.991L616: <&4 LIT Wli3aS 2 IN 

 



66. . . . . Table €& for Forsyth County shows that all 

precincte at least 50 pexcent African-American in total 

population are included in Congressional District 12 for each of 

the measures of Democratic strength considered. Once the 

African-American population percentage drops below 50 percent, 

then regardless of the Democratic strength registered in the 

precinct, there is no guarantee that the precinct will be 

assigned to Congressional District 12. As further evidence that 

race 1s more important than party when sorting precincts between 

Congressional District’s 5 and 12 in Forsyth County, if the 

population is less than 40 percent African-American, then 

regardless of how concentrated Democratic strength may be, it is 

highly unlikely that the precinct will be included in 

Cengressional District 12. 

67. .". . . All 17 majority African-American Guilford County 

precincts are in Congressional District 12. However a number of 

precincts that are majority Democratic on the various measures 

are outside Congressional District 12. The contrast is 

pronounced in terms of party registration where three-fourths of 

the precincts 50 - 59.99 percent Democratic by registration are 

in Congressional District 6 rather than Congressional District 

12. Even in the 60 - 69.99 percent registration range, most 

precincts are excluded from Congressional District 12. Similarly 

with the Gantt vote in 1580, 18 of 22 precincts in which Gantt 

32 

117 W1J33dS oad  



  

got 50 ~- 69.99 percent of the vote are outside Congressional 

Dimexict 12. Th= pattern repeats with the other Two sets of 

election data. All of the heavily black population precincts are 

in Congressional District 12 while only some of the heavily 

Democratic precincts are included. 

68. . ... . All three Iredell County precincts more than 30 

percent African-American in population are in Congressional 

District 12 while 15 of 21 precincts less than 30 percent 

African-American are in Congressional District 10. The strongest 

evidence of a racial explanation comes from the party 

registration analysis where 13 of the 19 precincts less than 30 

percent African-American black but majority Democratic in 

registration are excluded from Congressional District 12. The 

party explanation is best supported by the 1990 Gantt analysis in 

which the only four precincts carried by the African-American 

candidate are in Congressional District 12. Despite all but two 

precincts having Democratic registration majorities, no more than 

four precincts ever cast a majority of their votes for Democrats. 

If I lower the threshold to 40 percent support for a Democrat, 

results for the Gantt and the Court of Appeals contests would 

support a racial explanation. 

68. . . . . All 34 Mecklenburg County precincts more than 30 

percent African-American are in Congressional District 12! In 

33 

72d cunt 00, snl 9991616: <k 117 W345 S90 JN 

 



contrast, of the 99 precincts less than 30 percent African- 

  

Amaexican, 82 are in Congreseional District 9. This alone 

suggests that race played an very important role in selecting 

precincts for Congressional Districts 9 and 12. The tables for 

Democratic registration and support for Gantt in 1990 fit best 

with an interpretation that race was more important than party 

since in both of these presentations, the bulk of the 30 - 39.9% 

black precincts in which Democratic strength is in the 50 - 59.9 

percent range are excluded from Congressional District 12. 

Focusing on the Gantt vote in 1990, the bulk of the 30 - 39.9 

percent African-American precincts in which he polled 60 - 69.9 

percent of the vote are outside Congressional District 12. The 

1988 Court of Appeals vote provides the strongest support for a 

party explanation for districting. IF, however, 1 include 

precincts in which Democratic strength is 40 - 49.9 percent, then 

most of the precincts are in (Congressional District 8, for all 

four measures of partisanship. 

70. . . . . Rowan County has only three majority African- 

American population precincts (all more than 70 African- 

American) and these are assigned to Congressional District 12. 

The majority-white precincts in Congressicnal District 12 tend to 

be the ones with the strongest support for Democrats. Every 

precinct that gave Gantt in 1990 a majority is in Congressional 

District 12 and all but one precinct that voted Democratic for 

34 

Std S10 00, 9 $9.991.616: ~& LIT WI 545 ed IN 

 



  

  

lieutenant governor or court of appeals in 1988 is in 

Congressional District 12. The strongest evidence for the racial 

explanation comes from the registration data where most of the 

precincts 50 - 89.9 percent Democratic in registration are not in 

Congressional District 12. If I. drop the threshold down to 40 

percent Democratic on each measure, then in the 40 - 49.9 percent 

Democratic category, most precincts are excluded from 

Congressional District 12 on each of the four measures. 

Ti. * Finally, "2 rurn to the Table 8 portion. for all six 

counties combined in Congressional District 12. District-wide 

all African-American population majority precincts are assigned 

to District 12. On the other hand, a large number of precincts 

which demonstrated their willingness to support Democratic 

candidates in the 1988 and 1990 general elections irrespective of 

the race of the Democratic candidate are assigned to adjacent 

districts and not Congressional District 12. For example, 60 of 

the 98 white majority precincts won by Harvey Gantt in 1990 were 

assigned to another district. If the motivation was to create a 

Democratic majority Congressional District 12 which was 

hospitable to an African-American Democratic candidate in the 

1990 U.S. Senate contest, a large proportion of favorable turf is 

not included in the district. A partisan district would have 

attempted to include a majority ¢f those precincts and paid less 

attention to race in the construction of the district. 

35 

Ld £1:01° QO. 2 inf 9991616: x6 117 WIJ34S 98 ON 

 



  

Y. VOTER PARTICIPATION RATES WITHIN THE GEOGRAPHY QF   

CONGRESSIONAL DISTRICTS 1 AND 12   

76. In order to address the question of whether African- 

American persons within the boundaries of Congressional Districts 

1 and 12 of 1997 have less opportunity than white persons to 

participate in the political process, I have estimated 

participation rates for selected Democratic primary, Democratic 

runoff, and general elections between 1990 and 1998 to determine 

the patterns of participation among the two. 

77. I present a summarization of these results in Table 7 

for Democratic primary and runoff elections of the 1990s (see 

Exhibits C and D for summaries of the regression and extreme case 

analyses for the elections examined). 

78. I present a summarization of these results in Table 8 

for general elections of the 1990s. 

20d gh Ser $9LYYTLETE: XB L117 WI34S 99 IN 

 



TABLE 7 

Participation Rates by Race in Selected North Carolina 
Congressional Districts of the 1997 and 1998 Plans in Democratic 

Primary and Runoff Elections of the 1990s 
(Participation as % of Registration) 

[abridged] 

White 

Office PArtic..   

U.S. Senate 

U.S. Senate 

{P) a3 

( 
State Aud. (P 

( 
P 

P 
RY i. .097 
) .098 
Py .0B8S 
) . 025 

U.S. Senate 

J.8. cong. 

117 WIESE Sr ON  



TABLE 8 

  

Participation Rates by Race in Selected North Carolina 
Congressicnal Districts of the 1997 and 1998 Plans in 

General Elections of the 1990s 
[= 

(Participation as % of Registration) 
(abridged] 

White Afr. Am. 

Year CD Office Partic. Partic. 

1990 12 U.S. Senate . 807 . 572 

1992 12 State Aud. «5533 .825 

1996 12 U.S. Senate .9547 «473 

1996 12 State Aud. “503 . 453 

1998 12 U.S. Senate e411 .345 

1998 12 P.5.4 Cong. .408 .34%6 

38 

bed 2107 00, 97 $9.991.2676: xe 117 WIJ3daS Sd IN 

 



  

V. WHITE CROSS-OVER VOTING IN NORTH CAROLINA   
  

79. I examine whether white voters in the northeast and 

Piedmont regions of the state of North Carolina in particular 

cross-over and give support to candidates of choice of African- 

American voters in recent general elections. If white cross-over 

voting 1s common in the parts of the state where Congressional 

Districts 1 and 12 were created in the 1997 plan, this 

information might have been taken into account by the architects 

of the North Carolina congressional districting plan under 

challenge. In order to address this question, I will first 

report upon estimated white crossover rates using bivariate 

ecological regression and extreme case (homogeneous precinct) 

analysis for the selected statewide general election contests of 

1990, 1992, 1996, and 1998 held within the boundaries of 

challenged Congressional Districts 1 and 12 adopted in 1997. 

Second, I will report upon estimated white cross-over rates for 

the 1998 Congressional elections held within the boundaries of 

the districts ordered by the courts in 1998. 

80. In Table 9, I report my estimates of white cross-over 

voting in the 1990 contest for U.S. Senate involving the 

candidacy of African-American Harvey Gantt, in the 1990 contest 

33 

0rd S100 00 90 Tv $YLYYTLETE: XE 4 L171 BIJ3dS 98 IN 

 



for State Auditor involving the candidacy of African-American 

  

Bill Campbell, in the 1996 U.S. Senate and State Auditor contests 

involving candidates Gantt and Campbell, and in the 1998 contest 

for the U.S. Senate where John Edwards was the candidate of 

choice of African-American voters. I conducted bivariate 

ecological regression and extreme case analyses of these 

elections. In both Congressional Districts of the 1997 plan, 

African-American candidates obtained appreciable levels of white 

cross-over support. 

40 

7d 21:01 C00, 97 £9.9912616: ~e 117 WIZ3dS Sg IN 

 



  

TABLE 9 

  

Estimated White Cross-Cwver Ratez in North Carolina 

Congressional Districts of 1997 Plan A in Statewide General 
Elections for 1990, 1592, 1996, ang 1998 

(3 of Group Crossing-Over) 

Candidate Weighted Regression Extreme Case 

1990 i2 Gantt (B) 3737 32.6 
1992 12 Campbell (B) 40.7 35.8 
1996 12 Gantt (B) 35.3 82.1 
199¢ 32 Campbell (B) 35.8 35.5 
1998 2 Edwards 41.8 40.3 

Note: The cross-over rate of white voters is estimated through 
the use of weighted bivariate ecological regression and extreme 
case (homogenous) analysis for each of the elections. 

41 

CV °d £1:01 0 00. 98 IM $9.991.610: <X€4 117 1334S 9b ON 

 



dl. In Table 10, I report my estimates of white cross-over 

voting in the 1998 Congressicnal elections contests in Districts 

land J2,. v0. In both ‘Congressional Districts of the 139% 

plan, African-American candidates obtained appreciable levels of 

white cross-over support. 

VII. ELECTORAL SAFENESS QF CONGRESSIONAL DISTRICTS 1 AND 12 
  

82. To assess the electoral safety of Congressional 

Districts 1 and 12 in the North Carolina Congressicnal 

districting plan of 1997, I use electoral history included in the 

1998 reports of plaintiffs’ expert Lee Mortimer and defendants’ 

expert David W. Peterson as well as reconstituted election 

results for a number of recent statewide elections. In ny 

scholarly work on state legislative elections, I consider any 

election in which one candidate gets 60 percent or more of the 

total vote among two candidates as being a non-competitive 

election. 

B83. . 7. I Find that within the boundaries of the district 

that Peterson estimates that Democratic candidates won over 60 

percent of the vote in two 1988 elections and the 1990 U.S. 

Senate election between Jesse Helms and Harvey Gantt. 

w* ok Kk 

£1:0T 00, 9C INC 909912616: X 117 WIJ3dS 99 MN  



  

TABLE 10 

Estimated White Cross-Over Rates in North Carolina 
Congressional Districts of 1998 Plan A in Congressional 

General Elections for 1998 
(¥* of Group Crossing-Over) 

Candidate Weighted Regression Extreme Case 
ear CD f Choice White % Whi 3 

1998 1 Clayton (B) 30.4 33.3 

1988 12 Watt (B) 32.6 31.0 

Note: The cross-over rate of white voters is estimated through 
the use of weighted bivariate ecological regression and extreme 
case (homogenous) analysis for each of the elections. 

43 

Fv d 21:01 OC. es Ing $9.997L616: Xk 117 WiJi3aS 90 IN - 

 



  

  

85. I have also reconstituted the precinct election returns 

from nine racent statewide elections: the 1990 U.S. Senate 

Democratic primary, runoff, and general elections; the 1992 State 

Auditor Democratic primary and general elections; the 1996 U.S. 

Senate and State Auditor general elections; and the 1998 U.S. 

Senate Democratic primary and general elections. By allocating 

the precincts within Districts 1 and 12 to both the 1997 and 1998 

plans, I can determine how the various statewide candidates would 

have performed within the two sets of districts. These results 

are summarized in Exhibit E. Candidates of choice of African- 

American voters are winning the Democratic primaries or runoffs 

within both districts of both plans. And candidates of choice of 

African-American voters are usually winning more than 60 percent 

of the vote in the general elections within both districts under 

both plans. These analyses suggest that neither the 1997 nor the 

1998 plan is narrowly tailored. 

CONCLUSION 

86. On the basis of my above analysis, I conclude: 

1) that race was the predominant factor used by the 

state of North Carolina to draw the boundaries of the 

1997 U.S. Congressional districts; 

(2) that the state of North Carolina in creating the 

1997 U.S. Congressional districting plan subordinated 

44 

SF °d 21:00 000, 9 INL £9.991L616: X64 117 WIJ34S 98 ON 

 



traditional race-~-neutral districting principles, such 

  

38 compactneasaz, contiguity, respect for political 

subdivisions or communities defined by actual shared 

interests, to racial considerations; 

(3) that the political explanation for the Plan A 

Congressional districts adopted in 1997 offered by 

state defendants and their expert Professor Peterson is 

flawed . ... ; 

(9) that African-American voters residing in Districts 

1 and 12 in the Act 586 plan of 1997 do not participate 

at lower rates than white voters in recent state-wide 

Democratic primary and runoff elections, indicating any 

evidence that a history of official discrimination has 

not led to politically significant differences in 

political participation in Democratic primary elections 

in the districts as drawn in 1997 (there are 

participation differences between African-American and 

white voters in general elections held within the two 

districts of the 1997 plan, indicating that candidates 

of choice of African-American voters will need some 

white cross-over support to win within the two 

districts; 

(6) that more than sufficient levels of white crossover 

voting exists in the northeast and Piedmont regions of 

45 

9v 'd 1:00 00. 9% ITC $9.9914616: Xe 4 117 WI1J3d4S ad IN 

 



North Carolina such that fair U.S. Congressional 

  

districts can be drawn that do not need to be majority 

African-American in voting age population or voter 

registration in order to allow African-American voters 

a reasonable opportunity to elect candidates of choice 

in U.S. Congressional elections; and 

(7) that U.S. Congressional Districts 1 and 12 in the 

1997 North Carolina plan are overly safe from the 

standpoint of giving a candidate of choice of African- 

American voters an opportunity to be elected, thus 

questioning whether the plan was narrowly tailored to 

gatisfy a compelling state interest. 

I declare under the penalty of perjury that the foregoing 

Declaration is true and correct. 

Executed this tenth day of September, 1999. 

/s/ Ronald E. Weber, Ph.D 

46 

Vd STO. 00. nl $9.991.676: XE 4 1171. W0II34E 0d Ji 

 



TABLE 5 

NORTH CAROLINA HB 586--PLAN A CONGRESSIONAL DISTRICTS OF 1997 

Racial Composition of Total Population and Voting Age Population based on 1990 Census Data 
of Precincts in Split Counties by Congressional District 

[abridged] 

COUNTY PRCTNAME TOTALPOP WHITEPOP WHTPPCT BLACKPOP BLKPPCT DISTRICT 

  

Davidson Thomasville No, 2910 852 

Ward No. 1 * 2731 BOS 

Ward No. 5 * 2448 807 

Thomasville No. = 2709 1830 

Thomasville No. 3417 2506 

Ward No. 4 * 2403 1906 

Cotton * 4415 3807 

Ward No. 2 * 3114 2612 

Ward No. 6 * 2896 2502 

Thomasville No. 3523 3149 

Thomasville No. 3377 3034 
W * = * * 

Thomasville No. 9 4841 4784 

Thomasville No. 3811 3753 

Liberty * 3363 3345 

Emmons * 2467 2450 

Silver Valley * 2579 2562 

Alleghany * 506 506 

Denton * 1292 1292 

Healing Springs 1644 1642 
Jackson Hill * 790 789 

Lexington No. 3 2950 940 ; 12 

126677 113296 . ’ 96357 

 



  

(8) NORTH CAROLINA HB 586--PLAN A CONGRESSIONAL DISTRICTS OF 1997 

TABLE 5 (Ctd.) 

Racial Composition of Total Population and Voting Age Population based on 1990 Census Data 

of Precincts in Split Counties by Congressional District 

PRCTNAME 

abridged) 

TOTALPOP WHITEPOP WHTPPCT BLACKPOP BLKPFPCT 

  

Ju
l 

LA 

M. L. King Recreatio 
14th Street Recreati 

Mt. Sinai Church * 
Ashley Middle School 

Happy Hill Recreatio 

Carver High School * 

Kennedy Middle Schoo 

East Winston Library 

Lowrance Middle Scho 

Memorial Coliseum * 

Winston lake Family 

Forest Pk. Elementar 

Forest Hill Fire Sta 

St. Andrews United M 

Easton Elementary Sc 

Brown /Douglas Recrea 
Mineral Springs F. S 

Hill Middle School * 

Bishop McGuinness * 
New Hope United Meth 

Hanes Community Cent 

Trinity Moravian Chu 

Old Town Presbyteria 

Middlefork #2 * 
Brunson Elementary S 

Broadbay #2 * 

Middlefork ¥3 * 
Latham Elementary Sc 

Broadbay Kl * 

South Fork Elem Scho 

3134 
2344 
1711 
2045 
3386 
4317 
3165 
2895 
3102 
2746 
1662 
2969 
3022 
4101 
2606 
5643 
3743 
2621 
2775 
4398 
6044 
2702 
2860 
3449 
2303 
4842 
5497 
2855 
3128 
3995 

19 

12 
n 
4 

14 

67 

104 

189 

197 

318 

583 

816 

745 

804 

1610 

1094 

2896 

2030 

1457 

1571 

2758 

4029 

1787 

1966 

2426 

1630 

3671 

4344 

2249 

2497 

3254 

C
O
N
 

W
N
 

(
S
L
I
N
G
 

I 
& 
oJ

 
=~
 

5 
n
s
 

H
E
F
 

O
C
C
 
O
I
N
-
 

wn
 

oS
 

70 

79 

BY 
51 
a 

. 68 

. 98 
41 
«37 
. 80 
25 
+23 
.28 

«10 

. 60 
26 
«38 
“32 
«23 
.59 
«61 

62. 
66. 

66. 

68. 

. 34 
a0, 

15. 

71 

66 

14 

74 

18 

82 

.02 

18. 

18. 

81. 

77 

83 

45 

3113 
2328 
1679 
2002 
3310 
4209 
2970 
2686 
2781 
2133 

2803 
2200 
2192 
2451 
1496 
2658 
1653 
1134 
1104 
1584 
1938 

858 

856 

1004 
641 

1107 

1121 
566 
618 

682 

o
e
 

J 
eo 

BY
 

a
 

a 
2
 

~
 

.02 

  
L
O
N
 

O
O
N



  

2 
(- 

Ju
l 

  

Parkland High School 2761 
* vr f 3 

Lewisville #3 * 2694 

Ardmore Baptist Chur 1673 
Lewisville §2 * 4079 

Abbotts Creek #2 * 4037 

Bolton Swimming Cent 2889 

Abbotts Creek #3 * 3690 

Lewisville #1 * 5005 

Clemmonsville #2 * 3404 

Reynolds High School 2095 

Kernersville #2 * 5693 

Miller Park Recreati 2091 

Abbotts Creek #1 * 4655 

Trinity United Metho 2547 

Clemmonsville #1 * 2359 

Polo Park Recreation 1689 

Jefferson Elementary 2434 

Sherwood Forest Elem 972 

Messiah Moravian Chu 1536 

Summit School * 1775 

Vihitaker Elementary 2021 

Total 265878 

Guilford [omitted for printing) 

Iredell [omitted for printing] 

Mecklenburg [omitted for printing) 

Rowan [omitted for printing) 

2479 
2296 
1643 
235) 
958 

1506 
1767 
2010 

196918 

44 

a
.
 

[e
h 

24. 

N
N
N
 

W
W
W
W
W
 

D
D
O
 

Pe
st
 
LP
S 

[A
SI
 

4 
B
e
)
 

E
E
N
 

| 

a
 

a a 
@
 

N-
S 

~~
 

QD
 
A
D
I
N
 

MN
 

W
o
y
 

No
 

C
Q
O
 
W
N
N
Y
 
W
a
 

C
N
G
 

o
o
 

w
 

aN
 

0
 

r
o
 

on
 

NM
 

.10 

83
] 

A
L
O
 

n
e
n
 

L
O
L
A
 

e
o
 

O,
m