League of United Latin American Citizens (LULAC), Council #4434 v. Mattox Memorandum Opinion and Order

    UNITED STATES DISTRICT COURT 
FOR THE WESTERN DISTRICT OF TEXAS 

MIDLAND-ODESSA DIVISION FI LED
LEAGUE OF UNITED LATIN AMERICANS 
CITIZENS (LULAC), COUNCIL #4434§
et al. §

Plaintiffs, §
AND §
HOUSTON LAWYERS ASSOCIATION §
et al. §

Plaintiff-Intervenors §§V. s
§JIM MATTOX, et al.' §

State Defendants §
AND - • §
JUDGE SHAROLYN WOOD AND §
JUDGE F. HAROLD ENTZ §

NOV 08 lS8$. ,,
U. S. DISTRICT COURT. ' ClfFttCS OFFICE

BYi^VW^TT.. DEPUTY

MO-88-CA-154

MEMORANDUM OPINION AND ORDER
The above-captioned cause came on for trial before the Court 

on September 18,.1989. This suit was brought by named individual 
Plaintiffs and members of the League of United Latin American 
Citizens ("LULAC"), Council #4434, LULAC Council #4451 and LULAC 
Statewide. Plaintiffs are Mexican-American and Black citizens of 
the State of: Texas, Plaintiffs seek (1) a Declaratory Judgment

~  • : \  ; ;  •; r - . r :  '  c-. • • . . . . .  ^ v

that the existing at large scheme of electing State District Judges 
in nine (9)- .target counties of the State of Texas violates 
Plaintiffs' civil rights by unconstitutionally diluting the voting 
strength of Hexican-American and Black electors in violation of

1



«

Section 2 of the Voting Rights Act of 1965, as amended, 42 U.S.C. 
S 1973 (West Supp. 1989) ("Voting Rights Act")1; (2) a permanent 
injunction prohibiting the calling, holding, supervising or 
certifying any future elections for District Judges under the 
present at-large scheme in the target areas; (3) formation of . a 
judicial districting^scheme by which District Judges__in the target 

elected from districts which include single member 
districts; and (4) costs and attorneys' -fees. : a*; * rt f -rri ? i:c

This case really had its beginning in 1965, when Congress

Section 2 provides in pertinent part:
"(a) No voting qualification or prerequisite to 
voting or.standard, practice, or procedure shall be 
imposed or applied by any State or political 
subdivision in a manner which results in a denial or 
abridgement of the right of any citizen of the United 
States to vote 'on account of race or color .... ;
"(b) A violation of subsection (a) of this section 
is established if, based upon the totality of 

tcircumstances, itr> is - shown that the political u 
processes leading to nomination or election in the 
State" or political subdivision are not.equally open r.. s ; 
to participation by members of a class of citizens 
protected by subsection (a) of this section in that 
its members have less opportunity to participate in 
the political process and elect representatives of 
their choice. The extent to which members of a 
protected class have been elected to office in the ' 
State or political subdivision is one circumstance 
which may be considered: Provided. That nothing in 
this section establishes a right to have members of 
a protected class elected in numbers equal to their-’ r...." 
proportion in the population." \

(Emphasis in the original.) r r::r>-, ...c,, - ..

2



passed the Voting Rights Act and it was signed by President 
Johnson. This Act, as everyone knows, had as its purpose -"to -rid
the country of racial discriminating in voting." • --

The next chapter in the saga was the holding in Chisom v. 
Edwards, 839 F.2d 1056 (5th Cir. 1988), cert, denied, sub nom, 
Chisom v. Edwards, 109 S .Ct. 310(1989]-(Chisom I). In Chisom I 
Judge Johnson held: "Minorities may not be prevented' from using 
Section 2 [of the Voting Rights Act] in their efforts to combat 
racial discrimination in the election of state judges; a contrary 
result would prohibit minorities from achieving an effective voice 
in choosing those individuals society elects to administer -and 
interpret the law." ,• • ; -. -- - ---

- Having concluded, as will later be pointed out in formal
Findings of Fact and Conclusions-of Law, that there "is racial 
discrimination in the election of state judges in some counties of 
the State of Texas, and the law plainly being that uuch 
discrimination is prohibited by the Voting Rights Act, this opinion 
should not come as any surprise to the attorneys or judges of this 
State. ^

Mr. Justice Holmes, in Southern Pacific Co, v. Jensen. 244 
U.S. 205, 221, in dissenting, said:

I recognize without hesitation that judges do and - J! 
must legislate, but they can do so only intersti— 
tially; they are confined from molar to molecular

3



motions.
This dissent has been on the books for 82 years and, while 

this Court recognizes that some judges may legislate, this Court 
is extremely reluctant to do^so. Legislation should be done by 
legislators. This Court has determined that our current system, 
as it applies to some counties, violates Section 2 of the Voters, 
Rights Act. Some fixing has to be done, because the current system 
is broken. n o x ni rr„- r n1- . ,

In writing this opinion, I am cognizant of the fact that our 
Texas Constitution will_need to be amended. Legislators should 
seriously consider nonpartisan elections for District Judges. As 
Chief Judge Tom Phillips, pointed out in his testimony, it really 
makes no sense that judges are selected because of their political' 

filiation. A judge should decide matters before him without , 
regard to partisan p o l i t i c s I t  speaks well of our current 
judiciary that our sitting judges have been able to make decisions f 
without regard to whether the judge is Republican or Democrat.

As long as judges, ; however, are selected on a partisan 
ballot, there will be some rancor and enmity between the successful 

the unsuccessful candidate. The loser is going to have regrets 
hy virtue of the fact that she or he did not secure enough votes 
in an election. It makes no sense to believe that a judge is

4



selected because the top of the ticket is either weak or strong.—  

This Court felt the animosity between certain judges in the u.
courtroom. _There is no need for this. Certainly-judicial reform ---
will not make all candidates live by the Golden Rule, but it is a 
step in the right direction, x . - •_ .. v. ,;

It was brought to the Court's attention that perhaps a 
majority of the voters in a General Election, and for that matter, 
in Primary Elections, have no idea of the qualification of a judge 
for whom they vote. Their vote is cast because a straight ticket i_La~
is being cast,' 'and a 'straight ticket includes judicial nominees __
from a particular political party.

If the Constitution is to be changed, would it not make 
sense to have judges elected when members of school boards or city 
councils are elected? These races are traditionally nonpartisan, - 1 7 ?
and people going to the polls to vote for school trustees or mayors L.i: c 
have for nthe most part some idea of the qualifications of the 
candidates. Judges could be selected at the same time in order to - ̂ .. 
make sure that one was not getting votes simply - because one is "  n: 
Democrat or Republicans Minority voters could go to the polls - 
with their heads held high and with some realization that their 
preferred candidate either would be or could be elected.

Certainly, it is not Court's intention to tell the

5



legislature how its job is to be accomplished. Single member 
districts may or may not be the answer if we are to continue to 
have partisan elections. There may be easier and better solutions 
that can evolve through the legislative process.

These are troubled- waters. n One liesitates to plunge into 
such waters, because our system of selecting judges has, for the. 
most part, served us well for many many years. Our Congress, 
however, in -1964, made changes. LOur: Courts ’have construed thosei 
changes, and it is ' now necessary to move forwardo so that 
minorities can realize the rights legally bestowed upon'them, and 
which have, in the past, been denied.-

THE PRESENT AT-LARGE SYSTEM , il. „ ^
This litigation ’challenges trhe system of electing 172 

District Court Judges at—large from areas composed of entire 
counties.2 ■ jl ... . .... :__. v

The present system of electing District Court Judges in 
Texas requires that each judge be elected from a District no 
smaller than a county. Tex. Const. Art. 5 § 7a(i) (Vernon Supp.

The counties at issue are: Harris, Dallas, Tarrant, Bexar, 
Travis, Jefferson, Lubbock, Ector and Midland.

6



1989) .3 - _Each Judge serves a term of four (4) -years.~ Tex. Const. 
Art. 5 § 7 (Vernon Supp. 1989). Candidates for District Judge must 
be citizens of the United States and the State of Texas, licensed 
to practice law in this State and a practicing lawyer or Judge of 
a Court in this State, or both combined for four years. Id. 
Candidates must have been a resident of that election district for 
at least-two |2) years and reside in that district during his or 
her term of election. Id. District Court Judges must be nominated 
in a primary election by a majority of the votes cast. Texv 
Election Code § 172 .DG3=tVernon 1986) . - Each candidate's political - 
Party is indicated on the election ballot.Judicial candidates are 
usually listed far riown on an election ballot. 'They run for t 
specifically numbered courts and must secure a plurality of the 
vote in the general election to win a judicial seat. -.ho

s . i -  I W T .-n i -  • ■ ; i-* r r > r ... C  C  O i C ; '  "■.; i r  1.0 o f -  v  . i c  -. • . ■ •; : ‘  i ' '  xrc

r- • METHODOLOGY, DATA AND ELECTIONS ANALYZED V : ,, . i&J fa- 
Statistical analysis is the common methodology employed and t 

accepted to prove the existence of political cohesiveness and

-This system is "at-large" because judges are elected from i 
the entire county rather than from geographic subdistricts within -V’ the county. ■-

7



racial bloc voting necessary to establish a voter dilution case.4 
Ecological regression analysis5 and extreme case analysis6 were the 
types of statistical analysis used by Plaintiffs' experts in the 
present case.7

In Thornburg v. Ginales. 478 U.S. 30, 106 S.Ct. 2752, 92 
L.Ed.2d 25 (1986), Justice Brennan held thatracial bloc^voting can 
be established by a type of abstract statistical inquiry called 
"bivariate regression analysis." This analysis correlates the race 
of the voters and the level of support given to the candidate. 
Id. at-61-» - - If a candidate is supported by a large proportion of 
the minority group yet does not win, the vote is declared to be 
racially polarized in a legally significant sense and racial bloc 
voting is taken to be established.

All-variables beside race of the voters and support given 
the candidates that might also explain voters' , choices are 
expressly excluded from consideration. In Justice Brennan's view, 
"[i ]t is the difference between the choices made by [minorities] 
and whites - not the reasons for that difference - that results in 
[minorities] having less opportunity than whites to elect their 
preferred representatives." Id. at 63.

5c Ecological regression analysis shows the relationship 
between the ethnic composition of voting precincts and voting 
behavior, i.e.^ which candidate receives how many-votes from each 
race/ethnic group. This type of analysis incorporates the 4ise of 
a coefficient of correlation or Pearson r, accompanied by an 
estimate of the statistical significance of r, the coefficient-Of 
determination and the regression line. See Overton v. City of 
Austin. 871-F.2d 529, 539 (5th Cir.i 1989) . i.he ' =;-ni :cans ,

6 Extreme case or homogenous precinct, analysisalooks to 
homogenous precincts in which almost all of the people of voting 
age belong to one ethnic group. If race/ethnicity reflects voting 
behavior, then election results in predominately minority precincts 
should differ from results in predominately Anglo precincts. .

•j ^The majority which agreed with Justice Brennan that voter 
dilution was demonstrated by the'impact or results of the Zimmer 
factors and the Gingles threshold analysis deserted him when he 
came to the proof of the second and third Gingles factors. -'..

8



The data used by Plaintiffs to support their statistical 
analysis varied according to the type of information available to 
them since the 1980 Census. Plaintiffs used voting age population 
data by census tract to establish the Ginales 1 factor of size and 
geographic compactness. Plaintiffs used a variety of data sets to 
establish the Singles 2 cohesiveness and Ginales 3 white-bloc 
voting factors depending on information available in the County in 
question. __

In Counties where Plaintiffs presented a case on behalf of 
Hispanics ’• only, they 'relied on ~the percentage of Hispanic

• Justice White maintained that under Justice-Brennan's test 
there is racially polarized voting whenever a majority of whites 
vote differently from a majority of blacks, regardless of the race 
of the candidates. Ginqles. supra, at 83. To illustrate his 
disagreement, Justice White posited the hypothetical which assumed 
an eight-member multimember district that was 60% white and 40% 
black, the blacks being geographically located so that two safe 
black single-member districts could be drawn. Justice White 
further assumed that there were six white and two black Democrats 
running against six white and two black Republicans. Justice White 
wrote, "[u ]nder Justice Brennan's test, there would be polarized 
voting and a likely § 2 violation if all the Republicans, including 
the two;: blacks, are elected, and 80% of the blacks in -the 
predominately black areas vote Democratic." Id. at 83. Justice 
White concluded that such analysis was "interest—group politics 
rather that a rule hedging against racial discrimination." Id. at 83.

Justice O'Connor and the three other Justices for whom she 
wrote did not reject bivariate regression analysis solely to 
establish political cohesiveness and assess the minority groups 
prospects for electoral success. , Id. at 100. However, Justice 
0 Connor did reject Justice Brennan's position that evidence that 
explains divergent racial voting patterns is irrelevant.

9



registered voters in voting precincts in any given year. These 
figures were based on Spanish surname counts done by the Secretary 
of State of Texas. In other instances, Plaintiffs used counts of 
Black and Hispanic total or voting age population in each.precinct 
of a particular county.- When counts were not available, Plaintiffs 
based their analysis on 1980 census information. In some counties, 
precincts retained the same boundaries reported in the 1980 census. 
1980 census data, from precincts with unchanged boundaries were used 
in those counties. In several counties, Plaintiffs reconfigured 
precinct lines8 and used demographic data from these newly created- 
precincts. When relying on census data, Plaintiffs calculated the 
number-of non^minorities ̂ within precincts by subtracting the number 
of Hispanics and Blacks from the total number of persons within the 
precinct. -

Plaintiffs' experts only reviewed elections where a minority 
candidate-opposed an Anglo.£ They preferred to analyze general 
elections, however primary elections were analyzed when no minority 
candidate made it past that stage of the electoral process.

The Supreme Court in Thornburg v. Ginoles. supra, requires

This process requires comparing new precinct maps with 
their new lines and census block maps that show racial composition 
of the blocks. This process is frequently used to update precinct data.

10



the analysis of several elections to determine if there is a 
pattern of voting related to race/ethnicity. In the present case, 
when there were District Court elections in a county in question 
in which a minority opposed an Anglo, Plaintiffs relied solely on- 
analysis of District Court elections. In some Counties this 
included both general and primary elections. Where there were not 
enough such District Court elections other elections were analyzed. 
First, County Court elections in which minorities opposed Anglos 
were selected. Next, Plaintiffs turned to Justice of the Peace 
elections where the election district was at least as large as a 
city within the county at -issue. Finally, if no relevant local 
judicial races occurred, Plaintiffs analyzed statewide judicial 
elections. See Testimony of Dr. Robert Brischetto.

All jurisdictional:;_ prerequisites necessary to the 
maintenance of the claims'of the parties have been fulfilled.
After reviewing the testimony'and exhibits introduced at rtrial, as- 
well as the arguments and authorities of counsel, the Court hereby 
enters the following Findings of Fact and Conclusions of Law 
pursuant to Federal Rule of Civil Procedure 52. ....

11



FINDINGS OF FACT
INDIVIDUAL PLAINTIFFS

1. The names and counties of residence-of the ten (10)
named individual Plaintiffs are as follows: (a) Christina Moreno - 
Midland; (b) Aquilla Watson - Midland; (cj Joan Ervin - Lubbock; 
(d) Matthew W. Plummer, Sr. - Harris; (e) Jim Conley - Bexar; (f) 
Volma Overton - Travis; (g) Gene Collins - Ector; (h) Al Price - 
Jefferson; - (.i) Mary -Ellen Hicks - Tarrant; and (j) Rev. James 
Thomas - Galveston. Each named Plaintiff is a citizen of the 
United States registered end -qualified to vote in District Court 
elections in Texas.-Except for Christina Moreno, who is Hispanic, 
each named Plaintiff Black. j • . 't,.: ''

ORGANIZATIONAL PLAINTIFFS
2. Plaintiffs LULAC #4434 and LULAC #4451 are local 

chapters of the larger Statewide LULAC organization. J Members of 
the LULAC Statewide organization reside in all of the counties 
challenged in this suit. Depo. of John Garcia. The organization 
is composed of both Mexican-American and Black residents of the 
State of Texas. The members of LULAC #4434 reside in Midland 
County. The members of LULAC #4451 reside in Ector County.

3. - i - Plaintif f-Intervenor the Houston Lawyers Association

12



("HLA"), is an association of Black lawyers in Harris County. The 
participation of Plaintiff-Intervenor the Texas Black Legislative 
Caucus ("TBLC") is limited to the remedy stage of this litigation.

DEFENDANTS & DEFENDANT-INTERVENQRS
4. Defendants are sued in their official capacities only* 

Defendant' Jim Mattox is the Attorney General of the 
State of Texas and charged with the responsibility of enforcing the 
laws of the State. /

Defendant George Bayoud is.-Secretary of-State of Texas. 
As such he functions- as chief elections officer charged with 
administering the election Jaws of the State. Secretary Bayoud'is 
substituted as a party in this litigation for former Secretary of 
State Jack Rains. oC 2 . ' . . ) ^  l ... :

Defendants Thomas R. Phillips, Michael J. McCormick, Ron 
Chapman, Thomas J. Stovall, James Fr Clawson, Jr., Joe E. Kelly,: 
Robert M. Blackmon, Sam M. Paxson, Weldon Kirk, Jeff Walker, Ray 
D. Anderson, Leonard Davis and Joe Spurlock, II are members of the 
Judicial Districts Board9 created by Art. V Section 7a of the Texas

Several members of the Judicial Districts Board- were 
replaced by new members during the interim of this litigation. 
Michael J. McCormick replaced John F. Onion, Robert M. Blackman^ 
replaced Joe B. Evans and Jeff Walker replaced Charles Murray.

13



Constitution and Art. 24.911 et seq. of the Texas Government Code.
The Judicial Districts Board is charged with reapportioning 
districts from which District Court Judges are elected. . .. .

----- Sitting- District Court Judge Sharolyn Wood, 127th
District Court, Harris County and Judge Harold Entz, Jr., 194th =
District Court, Dallas County Intervened in their individual 
capacities as Defendants.10 ' ;

GINGLES THRESHOLD ANALYSIS 
Size and Geographic Compactness 

• "5* Harris County. - Harris County has the largest 
population among the nine target Counties in this case. Plaintiffs 1 
are proceeding only on behalf of Black voters in Harris - County. ror 
With a total population of 2,409,544,11 its Black population i s 1 01. 
473,698 (19.7%). There are 1,685,024 people of voting age,12 with .--'In 
305,986 (18.2%) voting age Black residents of Harris County.

10 Thirteen District Court Judges from Travis County initially - ■ y
intervened as Defendants. The Court struck their intervention at their request. • •.•••

11 In each County, Plaintiffs rely upon the 1980 Census for 
total population of Blacks and Hispanics within the County.

For all Counties in this case, Plaintiffs relied on a - -.1- 
computer print out of voting age populations prepared by the Data ^-ir: 
Center at Texas A&M University _ directly from 1980 U.S. Census~ m e  tapes ~ : : • ; T ; 1 h • • •

14



-- There are fifty nine (59) State District Courts in 
Harris County. Black residents are concentrated in the North 
Central, Central and South Central .sections of Harris County. H- 
04, p.> 2, Map, of Proposed Districts.13 Evidence was introduced 
that nine (9) Black single member districts of greater than fifty 
percent (50%) Black voting age population were possible. Id. at 
1; Plaintiff-Intervenor Harris. County ( "P-I.-f H") Exhibits 2, :2a.

6. Dallas County. Dallas County is the second largest 
County involved in this case. -Plaintiffs are proceeding only on 
behalf of Black voters in Dallas County. Dallas County has a total 
population of 1,556,549. Its Black population is 287,613 (18.5%).. 
There are 1,106,757 people of voting age, with 180,294 (16%) voting 
age Black-residents . Plaintif f s ' DDallSs Ccfiinty ("D",). Exhibit 017- 

There were thirty six /(36) State District Courtsmin

Plaintiffs' Harris County ("H") Exhibit 01.

13 Plaintiffs drew districts in each County of approximately 
equal size based on the number of District Courts in the County. 
Plaintiffs calculated the size and number of precincts in each 
proposed district on the basis of both total population and voting 
age population. This Court recognizes that the concept of "one man 
one vote" does not apply to the judicial elections. Chisom I . 
supra, at 1061. Accordingly, this Court's analysis rests upon 
Plaintiffs' calculation based upon voting age population. 
Plaintiffs drew each district on this basis under the assumption 
that each district should contain l/m of the voting age population 
in the County,-with n being the number of District Courts in the 
County. ‘ -Plaintiffs' Post Trial Brief at 11.

15



Dallas County at the time this case was filed. On September 1, 
1989, the Texas Legislature created a thirty-seventh State Judicial 
District-Court in Dallas County. Black residents are concentrated 
in the Central and South Central sections of Dallas County. D-04, 
p. 2, Map of Proposed Districts based on 3̂6 District Courts_ 
Evidence was introduced that seven (7) Black single member 
districts 'of greater than fifty percent (50%) Black" voting age 
population were possible. Id. at 1, 3-9;14 Plaintiff-Intervenor 
Dallas ("P-I D") Exhibits 34. Plaintiff-Intervenors' Exhibit 7 
reflects that there are approximately 36 homogeneous precincts of 
90% Black population. __  ___  . _ . _ ___i-.-t ___  .

7 Tarrant County. Plaintiffs are proceeding-only on 
behalfr of Black voters in Tarrant County. Tarrant County has a 
total ^population of 860,880>  ̂The rBlack population-of "Tarrant 
County is 101,183 (11.8%). There are 613,698 people of voting age, 
with 63,851 -(10.4%) voting age'Black residents of Tarrant County. 
Plaintiffs' Tarrant County ("Ta") Exhibit 01. .*

There are twenty three (23) State District Courts in 
Tarrant County. Black residents are concentrated in the Center of

Proposed single member districts 1 & 3 barely meet the 
Overton majority - minority voting age population requirement. 
These proposed districts contain 51.33% and 52.05% black voting age 
population respectively.

16



Evidencethe County. Ta-04, p. 2, Map of Proposed Districts, 
was introduced that two (2) Black single member districts of 
greater than fifty percent (50%) Black voting age population were 
possible. Id. at 1_., , :=;• i.'V' ' :: 1-h= -,-.vr

8. Jexar County. Plaintiffs are proceeding only on behalf 
of Hispanic voters in Bexar County. _ ;.Bexar County - has a .total 
population of 988,800. Its Hispanic population is 460,911 
(46.61%). There are 672,220 people iof vating1-age with 278,577 
(41.1%) voting age Hispanic residents of Bexar County.e- 
Plaintiffs' Bexar County~("B") Exhibit 01. --

There are nineteen (19) State District Courts in Bexar 
County. Hispanic residents are concentrated in the Central and 
South Central sections of the County comprising most of the 
population! of the City of San Antonio. B-04, p.-2, Map of 
Proposed Districts. Evidence, was introduced ithat eight (8}r
Hispanic single member districtsvof greater than .fifty percent
(50%) Black voting age population were possible. Id. at 1.

9. Travis County. Plaintiffs are proceeding only on 
behalf of Hispanic voters in Travis County. With a total 
population of 419,335, its Hispanic population is 72,271 (17.2%). 
There are 312,392 people of voting age with 44,847 (14.4%) voting 
age Hispanic residents of Travis County. Plaintiffs' Travis

17



There are thirteen (13) State District Courts in Travis. 
County. The largest concentration of Hispanic residents in one 
area, if at all, appears to be located in the Eastern portion of 
the County. Tr-04, p. .2; Tr-05, p.l, Map of Proposed Districts. 
Mr. David Richards testified that in his opinion the Hispanic 
community was pretty ~ - well ~ dispersed in “'Travis County. 
Nevertheless, evidence! was _ introduced that one (1) !combined 
minority single member district of greater than fifty’percent (50%) 
Hispanic voting age population was possible. Id. at 1. 
Plaintiffs! Exhibit Tr-04 depicts the single member -. Hispanic 
district proposed for Travis County, v The Court finds that"it is 
without moment that the proposed district appears to be minimally 
contiguous. ■ ' .• Courts i r . i

10. Jefferson County. Plaintiffs are proceeding only on 
behalf of Black voters in Jefferson County. cJefferson County has 
a total population of 250,938. Its Black population is 70,810 
(28.2%). There are 179,708 people of voting age of. which there are 
44,283 (24.6%) voting age Black residents of Jefferson County. 
Plaintiffs' Jefferson County ("J") Exhibit 01.

There are eight (8) State District Courts in Jefferson 
County-r Black residents are concentrated in the Central and South

County ("Tr") Exhibit 01.

18



Eastern portions of Jefferson County. J-04, p. 2, Map of Proposed 
Districts. ''Evidence was'^introduced that two (2) Black single 
member districts of greater than fifty percent (50%) Black voting 
age population were possible. Id. at 1 ’ - J ’- « —

11. Lubbock County. Plaintiffs are proceeding on behalf • n>.
of the combined Black and Hispanic voters in LLubbock County. .There ...
is a total population of 211,651 in Lubbock County. The Black ^  _. 
population of Lubbock County is;15,780 (7-5%), while the Hispanic 
population is 41,428 (19.6%). There are-150,714 people of voting ; 
age, with 9,590 (6.4%) voting age Black residents and 22,934 ;
(15.2%) voting age Hispanic residents. The combined minority 
votings age : population is 32,524 (21.6%). Plaintiffs' Lubbock'■ ;:al r
County ("L") Exhibit 01. : * ; ( ,

There are six (6) State District Courts in the Lubbock----.. - 
Crosby County area. The combined minority population is non - 

concentrated in thes.North Eastern, Eastern and South Eastern Lons 
sections of those Counties. L-04, p. 2, Map of Proposed Districts. 
Evidence was introduced that one (1) combined minority single 
member district of greater than fifty percent (50%) Black voting 
age population was possible. Id. at 1. This remains true when 
Plaintiffs controlled for voting age population of non-United Vs 
States citizens of Spanish origin. Plaintiffs' Exhibit L-ll. •• -V

19



12. Ector County. Plaintiffs are proceeding on behalf of
combined Black and Hispanic voters in Ector County.... The total 
population of Ector County is 115,374. Its Black population is 
5,154 (4.5%) and the Hispanic population is 24,831 (21.5%). There. 
are 79,516 people of voting age. The voting age population by 
minorities consists of 3,255 (4.1%) Black voters and 14,147 (17.8%) 
Hispanic voters for a combined minority voting age population of 
17,402 (21.9%). Plaintiffs' Ector County ("E") Exhibit 01.

■"7There are four (4) State District Courts in Ector 
County.' Minority residents are "concentrated in the Southwest 
section of the County. E-04, p. 2, Map of Proposed Districts. 
Evidence -was introduced that-one (1) combined minority single 
member district of greater than fifty percent (50%) minority voting 
age population was possible. Id. at 1. It is possible to draw a 
district of combined minority population of voting age even if non­
citizen voting age HiSpanics are eliminated -from .the calculations. 
Plaintiffs' Exhibit E-13. •_ - ,

13.: Midland County. Plaintiffs proceed on behalf of Black 
and Hispanic voters combined in Midland County. Midland County has 
a total population of 82,636. Its Black population is 7,119 (8.6%) 
and its Hispanic population is 12,323.(14.9%). There are 57,789

r hr >—  v* r f
people of.voting age,.-4,484 (7.8%) voting age Black voters and

20



6,893 (11.9%) voting age Hispanic voters. The combined voting age 
population is 11,377 (19.7%).  ̂Plaintiffs' Midland County ("M")' 
Exhibit 01. ------ —  -

There are three (3) State District Courts "in Midland 
County. Black residents ~are concentrated largely in the 
Northeastern, East Central and Southeastern sections of Midland 
County. M-04, p. 2, Map of Proposed Districts. Evidence was 
introduced^that one (1) combined minority single member district 
of greater than fifty percent (50%) combined voting age population 
was possible. Id. at 1. It'is possible to draw a district in 
which the combined minority population is in the majority even if 
non-citizen Hrspanics of 'voting age are excluded. Plaintiffs' 
Exhibit M-15.

Political Cohesion and White Bloc Voting
14. Racially polar-ized voting indicates that the group 

prefers candidates of a particular race.15 ‘Monroe v. City of' 
Woodyille, No.' 88-4433, slip op. at 5573, (5th Cir. Aug. 30, 1989).

The Supreme Court in Ginqles adopted the definition of 
racial polarization offered by Dr. Bernard Grofman, appellees' 
expert.-Dr. Grofman explained that racial polarization "'exists 
where there is a consistent relationship between [the] race of the 
voter and the way in which the voter votes' ... or to put it 
differently, where 'black voters and white voters vote 
differently.'" Ginqles. 478 U.S. at 53 n. 21.

21



Political cohesion, on the other hand, implies that the group 
generally .unites behind a single political "platform" of common 
goals and common means by which to achieve them. Id.at 5573.

The inquiry into political cohesiveness is not. to be 
made prior to and apart from a study of polarized voting. The 
Supreme Court made clear that "[t]he purpose of inquiring into the 
existence of racially polarized voting is twofold: to ascertain
whether minority group members constitute a politically cohesive 
unit and to determine whether whites vote sufficiently as a bloc 
usually to defeat the minority's preferred candidates." Gingles,:. 
478 U.S. at 56.

15. Plaintiffs presented testimony of two experts. - Dr. 
Richard Engstrom ("Dr. Engstrom") testified only about Harris and
Dallas j.Counties. Dr. Robert r.Brischetto ("Dr. Brischetto")

t. '* r.-. T O - ’ T i m o r . V  O t  D r  . F T 1- s >« r’ .V OU! j  i; . i r
testified concerning all other counties at issue in this case.:

16. Harris County.
a. Dr. Richard Engstrom testified on behalf of Plaintiffs 

and Plaintiff-Intervenors in Harris C o u n t y D r .  Engstrom used 1980 
U.S. Census counts of total Black population by precinct to analyze 
1980 election results. For .1982, 1984, 1986 and 1988, Dr. Engstrom 
used precinct voter registration estimates supplied by Dr. Richard 
Murray, a non-testifying expert. K.Plaintiffs' Exhibit P-I H-08ri-

22



Dr. Engstrom verified or "matched" the reliability of Dr. Murray's 
estimates and the 1980 Census counts by comparing Dr. Murray's 
estimates to an Hispanic precinct voter registration list compiled 
by the Secretary of State; Testimony of Dr. Richard Engstrom. 
Dr. Engstrom testified that there was "a very good: match.",

b..̂ . Dr-Engstrom analyzed,17 _ general elections in Harris 
County. He calculated "r". values16 between 0.798 and 0.880 for the 
17 elections analyzed.17 tPlaintiffs' Exhibit P-I H-01 pp. 1-2.
Dr. Engstrom's regression analysis shows a strong relationship 
between race and voting patterns in Harris County. See Appendix. 
A to this opinion ("Appendix"), Plaintiffs' Exhibit P-I H-01 pp. 
1-2. All of his correlation coefficients18 exceed .79 (79%) except;

16 The "r" value describes the relationship between the racial 
composition of a precinct and ..the number of votes a particular 
candidate receives. Testimony of Dr. Engstrom. To put it another 
way, "how consistently a vote for Black candidate changes as the 
racial composition of the precinct changes." Id.

"Crucial to the validity of regression analysis are .the 
values for 'r' and 'r[squared]' , which measure the strength of the 
correlation and linear relationship of the variables being 
examined, in this case the race of the voter and the candidate he 
supports." Overton, 871 F.2d at 539.

The "r" value is also referred to as the "correlation 
coefficient" or "Pearson r." A positive Pearson r shows that as 
the percentage of minorities in a precinct increases, so does the 
support that a minority candidate receives. A Pearson r of -1 
shows the opposite, as the percentage of minorities in a precinct 
increases, cthere is decrease in the support that a minority 
candidate receives. A Pearson r of 0 shows that there is no

23



one. Id.., see section on Bivariate Regression under the column 
heading of Correlation Coefficient. Dr. Robert Brischetto
generally testified with regard to the counties in issue other than 
Dallas and Harris County, that a ..Pearson r of 1 (100%) would show 
perfect correlation. He further testified that social scientists - 
consider anything over 0.50 (50%) as showing a strong correlation.

c. Further, each Pearson r is accompanied by an estimate 
of the likelihood that the estimate would occur by chance. This 
figure is known as the significance level. In the regression 
analyses for Harris County, as well as all the counties in issue, 
the significance level was much smaller than the generally accepted 
level of extremely high significance of . 05.19 Testimony of Dr. 
Robert Brischetto; Testimony of Dr. Richard Engstrom. Dr. Engstrom 
testified that the probability that the Harris County estimates-- 
would have occurred by chance were less than 1 out of 10,000.

H d. The lowest squared for these analyses is “
approximately .62 (62%). This describes the percentage of the
variance in voting behavior explained by race/ethnicity. Testimony

relationship between the racial/ethnic composition of precincts and 
voting behavior. 1 , • . •

19 A -significance level of -.01, for example, ;means that the 
Pearson r in question would have occurred by chance only one time 
out of .one hundred. Cvcx ■ : -  -j. " G r

24



of Dr. Robert Brischetto.20 Squaring these "r" values21 to
calculate coefficients of determination demonstrates in the present 
case that race explains at least 62% of the variance in voting in 
all 17-elections relied on by Plaintiffs and Plaintif f-Intervenors.._ 

m e .  t= The one judicial race that did not exceed the 79% 
figure actually had a negative correlation. This race involved 
Mamie Proctor, a Black candidate running on a Republican ticket 
against Henry Schuble, an Anglo, for State Family Court 245. In 
the 1986 Proctor race, the correlation coefficient was -0.836 
(approx. -84%). Id. at 1; Plaintiffs' Exhibit P-I H-10 p. 2. ..This 
reflects that, as the percentage of Blacks in voting precincts 
increases, 'Proctor*s support decreased. In other words, even 
though Ms. Proctor is Black, she did not receive the support of the 
Black community. Hence, she was not the preferred, .candidate of ; 
Black voters in Harris County. Dr. Engstrom testified on cross 
examination tJhat the "candidate of choice" .was the-candidate who

For example, if a Pearson r is .5, then 25% (5 x 5 or r 
squared) of the variance in voting behavior is explained by 
race/ethnicity.

This figure is also known as the coefficient of 
determination. It is the coefficient of correlation or Pearson r 
multiplied by itself. It shows how much or little "noise" there 
is around the line ofT correlation or, in other words, "the 
percentage of variance in the vote that is explained by the race 
of the voters." Overton, 871 F.2d at 539 n. 11.

25



received the majority of the black vote, not necessarily the Black 
candidate. - - : • - -

f. When Dr. Engstrom controlled for Hispanic votes, Dr.
Engstrom's regression analysis shows that Blacks consistently gave — 
more than 97% of their vote to their preferred candidate. Id., see r 
last two columns. - , .J

g. Dr. Engstrom's homogenous precinct analysis
corroborates the results of his regression analysis .t> See Appendix ; 
A, Plaintiffs' Exhibit P-I H-01 pp. 1-2. It shows that Black 
voters in Harris County gave more than 96% of their votes to the 
preferred candidate of Black voters in every election except 
Proctor's . : Ms . Proctor received 5% of the Black vote. ~

' - h . ~ Finally, in all counties including Harris County, 
Plaintiffs "weighted"1 precinct data in f order 'to account for 
variations in the population size of the various precincts. 
Testimony of Dr. Richard Engstrom; O v e r t o n rsupra, at 537. Dr. 1- 
Engstrom testified that on the basis of his analysis the Blackc 
community in Harris County votes cohesively in-general elections 
for State District Court Judges. i. Harris County
Defendant-Intervenor Judge Sharolyn Wood-("Judge Wood"), attacks:-^- 
Plaintiffs and Plaintiff-Intervenor's proof on the following’: 
grounds: (1) Dr. Engstrom failed to establish the reliability of

26



his data set; (2) absentee votes were not allocated to election 
returns; (3) the analysis does not reflect the effect of the influx 
of the Vietnamese population into Harris County and traditionally 
Black -precincts; and {4) the .analysis fails to reflect black 
candidate successes in primary elections or uncontested races.-- -■

j. In reference to the reliability of the data set, Judge 
Wood points to numbers on Dr. Murray's printouts that have been 
written over.,: struck out or crossed through, pencil notations and 
other marks. This Court finds the data set to be reliable.

k. - ..In response to the other concerns, Dr. Engstrom 
testified that: (1) primary elections were not examined in Harris 
County because those elections were not filtering out the candidate 
of choice of Black voters; (2) uncontested races do not assist 
researchers in their analysis; (3) the appropriate comparison in 
Voting Rights cases is Black and non-Black; (4) while :he did not 
specifically control for Asian Americans, they would be included 
in the percentage of non-Black votes; and (5) the range of absentee 
votes between 1980 and 1986 never exceeded 2.2% to 7.6%, while in 
1988 that range rose to approximately 13.6% per precinct. This 
Court finds that Dr. Engstrom's testimony adequately addresses 
these concerns. The Court further finds that the lack of control 
for absentee votes and Asian Americans does not significantly

27



affect Dr. Engstronr's analysis.
-1. The State Defendants and Defendant-Intervenors argue 

that it is a candidate's political party and the strength of 
straight ticket party voting that determines ~the result of any 
election contest and not the difference between the preferred 
candidates of whites and minorities. In support.of -this argument, 
Defendants and Defendant-Intervenors point to the.1982 and 1986 
Democratic sweep for judicial candidates in- Harris County and a 
similar Republican sweep in the years 1984 and 1988. n All 
Defendants attribute this phenomenon to top of the ticket straight 
party voting.22

m. Correlation and regression can also prove the third 
Ginqles prong by showing that a white bloc vote exis-ts. This is 
shown when the percentage of ■ votes received by .'the minority 
candidate decreases as the percentage of minority persons of voting 
age decreases . In other twords, the minority candidate^receives 
fewer votes as the percentage of non-minority persons in a precinct 
increases.; Regression results estimate the percentage of non­
minority support for minority candidates, otherwise known as the

In 1982, Senator Lloyd Bentsen was the lead Democratic 
candidate on the ballot. In 1986, Governor Mark White represented 
the top of the ticket Democratic candidate. In Presidential 
election years 1984 and 1988, President Ronald Reagan and President 
George Bush, respectively, were the top Republican candidates.

28



Anglo cross over vote. Plaintiffs' Exhibit P-I H-01 pp. 1-2, 
column 4. This is also referred to as the Y intercept. . .. ..

n. Dr. Engstrom calculated Y intercepts for the Black 
preferred candidate between 29 and 39 percent for the 17 elections 
analyzed. The highest Y intercept was 33.6%, but this percentage 
of the non-Black vote was for the non-preferred candidate Mamie 
Proctor. The highest percentage of Anglo cross over votes 
received by the preferred candidate of Black voters was 39 percent. 
See 1986 race Carl Walker, Jr., Black Democrat against George 
Godwin; Id. This is corroborated by a 40% Anglo cross over vote 
figure calculated for the same race in homogenous precincts of 90% 
or more non-Black population. Td. at column 1. Mr. Walker was 
the Black preferred candidate and won. > Plaintiffs' Exhibit P-I H 
10. Two other Black preferred-candidates-drawing opposition .inJthe 
1986 elections lost their elections even though they had identical 
Black community support; These two candidates had slightly?.less; 
Anglo cross over vote. Plaintiffs' Exhibit P-I H-01 pp* J.-2, 
column 1. Five other Black preferred candidates drawing opposition 
in what appears to be county-wide elections lost in t h e -1986 
elections. -= Plaintiffs' Exhibit P-I H-10.23 This ianalysis

j These candidates are: Bonnie Fitch, Raymond Fisher,
Francis Williams, Sheila Jackson Lee, and Cheryl Irvin-ar.

29



demonstrates that an Anglo bloc vote exists. Dr. Engstrom 
testified that the Anglo or white bloc vote in Harris County is 
sufficiently strong to generally defeat the choice of the Black 
community-.-*- This Court a g r e e s -? •- r :•1 r- ,-.

o. Plaintiff-Intervenor Sheila Jackson Lee also testified
about political cohesiveness among Black voters in .Harris County. 
Ms. Lee has lived in Harris County approximately 11 years and has 
been a candidate in several- judicial^ e l e c t i o n s P l a i n t i f f s '  
Exhibit P-I H-01 pp. 1-2; Exhibit P-I H-10 pp. 1-3. She had many 
different endorsements and campaign strategies but still lost. 
She testified that her loss was attributable to not getting enough 
white votes. f •. v 1^3 ^Urv.. ]•••••'••• : • • • - r. i.V J . \

p. This testimony was supported by the deposition
summaries of Thomas Routt,! Weldon Berry, Francis -Williams and 
Bonnie Fitch. _,ul' v r.-

q. Defendant-Intervenor Wood presented the testimony^of 
Judge Mark Davidson. As a hobby, Judge Davidson analyzes the 
results of judicial elections in Harris County. His testimony 
concerned his views on what he has termed "discretionary judicial 
voters" ("DJV").24 Judge Davidson testified that 15% of the vote

He defines DJV's as voters who vote for at least one 
judicial candidate of one party and at least one of the other 
party. DJV's are also referred to as "swing" voters.

30



in judicial elections in Harris County were DJV's. The remaining 
85% split roughly evenly between straight . Democrat party and: 
straight Republican party voting. Based upon his analysis, Judge 
Davidson believes that race and ethnicity are irrelevant_to voting., 
behavior as it relates to. ;the .judiciary in Harris County. : Hew 
opines that DJV's determine the outcome of judicial contests in 
Harris County and the DJV vote can somewhat be garnered by various 
campaign factors. While this Court finds Judge Davidson to be a 
credible witness, under controlling law, the Court finds that his 
testimony is irrelevant. _

r. The Court further finds Defendant-Intervenor Wood's 
contention that -the Black preferred candidate lost their respective 
judicial races due to their failure to win the Harris County bar 
or preference poll orito" obtain the Gay Political Caucus ("GPC"), 
endorsement to be legally incompetent.

s. The complete data set used by Dr. Engstrom was used by 
Defendant's expert, Dr. Delbert Taebel for his analysis of Harris 
County. Dr. Taebel did not weight his precinct data to.account for 
variations in population size of various precincts in Harris County 
or any other county at issue.

t. Dr. Taebel analyzed 23 District Court general elections 
where minorities opposed white candidates -in Harris C o u n t y S t a t e '

31



Defendants' Exhibit D-05 pp. 9, 13, 29, 33, 37, 41, 45, 53, 61, 81, 
85, 89, 93, 97, 101, 105, 137, 141, 145, 161, 165, 173 & 177.-
Black and white voters voted differently in all 23 District Court 
elections. Id. The Blackr preferred candidate won"only six'(6)" 
times." The Black preferred candidate won seven (7) of 11 County 
Court general elections. Id. D-05-pp. -lr 5, 17^-21, 25, 109, 113,- 
117, 121, 175 & 129. Blacks and whites voted differently in each 
of those elections . Id.a mDr. * -Taebel i also analyzed >nine r ( 9 ) h  

judicial primary elections; seven (7) for District Court posts and 
two (2)-County Court posts. Id. D-05 pp. 49, 57, 65, 73, 77, 145 
157, 169 & 181. The Black preferred candidate won six (6) of the 
nine (9) primaries. Interestingly enough, each preferred candidate 
winning the primary lost the general election. Id. D-05 pp. 61, 
69, 81, 153,: 161, & 17 3 . ~ v;-.v • v — -i-i

17. Dallas County.: - ~ >'  ̂qu ~ • m - -
a. Dr. Engstrora used the same data set for his analysis 

of Dallas County. However, the 1980 Census counts were updated in 
1982 and 1988 by the Dallas County Elections Office by
reconfiguring precincts according to the changes made in precinct 
lines. Testimony of Dr. Richard Engstrom. Dr. Engstrom accepted 
the updated census counts for 1982 and 1988 as reliable. Id. In

32



the intervening years of 1984 and 1986, Dr. Engstrom looked for 
precincts that combined or split and aggravated precinct counts 
for those precincts. Id. ' ••• ■ . .

b. Dr. Engstrom analyzed seven (7) general elections_for 
State District Court where Blacks opposed Anglos between 1980 and. 
1988 in Dallas County. The correlation coefficient or "rM values 
exceed 0.864 (86%) for six ̂ 6) of the seven (7) elections analyzed. 
See Appendix A, Plaintiffs' Exhibit D-02. Dr. Engstrom's
homogenous precinct analysis and regression analysis shows a strong 
relationship between race and voting patterns in Dallas County. 
Id., see columns 2 & 3. Dr. -Engstrom -testified that the-
significance -level was much smaller than the generally-accepted 
level of extremely khigh significance of .05 and that the 
probability bhat the Dallas County estimates would have occurred 
by chance were less than 1 out of 10,000.

c; I- The lowest’K r L squared cbor - these analyses is 
approximately .75 (75%). This figure is found from multiplying the 
r value by itself for Jesse Oliver's judicial race in 1988. This 
coefficient of determination demonstrates that race explains at 
least 75% of the variance in voting in at least six (6) of the 
seven (7) elections relied on by Plaintiffs and Plaintiff- 
Intervenors. - —  -

33



d. Plaintiffs' Exhibit D-02 further shows that in five (5) 
of the seven (7) elections as the percentage of Blacks increased 
in precincts, so did Black support for the preferred candidate of 
Black voters;■>> See Homogeneous precinct analysis, column 2;-~ rrr~-

e . r■ Bivariate regression analysis reflects a negative 
correlation for Carolyn Wright's.-, .judicial .race in .1986.. Judge 
Wright is a Black who ran on the Republican ticket. She received 
-1.5% of .r the Black vote <i.and 71-.7% gof ̂ the non-Black vote. 
Plaintiffs' Exhibit D-02, columns 4< & 5. The correlation 
coefficient was -0.872 (-87%). -Id^ column 3. This reflects 
that, as the percentage of Blacks in voting precincts increases, 
Judge Wright's support decreased::: in other words, even though Ms. 
Wright is Black, she did„not receive the support of the Black 
communityg Hence, she was not .the preferred candidate of JBlack 
voters in Dallas County. Black voters also failed to support 
Judge Baraka, a Black Republican candidate in 1984.

f. When Dr. Engstrom controlled for Hispanic votes, Dr. 
Engstrom's regression analysis shows that Blacks consistently gave 
more than.97% of their vote to their preferred candidate. Id., see 
last two columns. Dr. Engstrom's analyses shows that Blacks are 
politically cohesive in general elections for State District Court 
in Dallas County. - • • ; : -

34



g. His analysis is confirmed'by the testimony of 
Plaintiff-Intervenors' Joan Winn White, Fred Tinsely, H. Ron White 
and Jesse Oliver. The Exhibits ^reflect that each Plaintiff- 
Intervenor received 97% or better of the Black, homogenous precincts 
and at least 83% of the votes in precincts with Black population 
of 50% to 90%. "Plaintiffs Exhibit P-I D-16 - D-22a.

h. Plaintiffs calculated the percentage of votes for-the 
Black preferred candidate, Jesse Oliver, and his white opponent'. 
Brown, in each of the proposed hypothetical single member 
districts. Plaintiffs' Exhibit D-12a. They .repeated this 
procedure.-for^ the judicial races involving the Black preferred 
candidates in Plaintiffs' Exhibit D-t)2 and Nathan Brin (an Anglo 
preferred by Black voters in Dallas County). Plaintiffs' Exhibits 
D-12b,i 12c & 12d. ■ In each , instance,an the Black t. community's 
preferredl* candidate received a r majorityr of votes a in each 
predominately Black hypothetical districts i, > :■

i. Defendant-Intervenor Judge Harold Entz ("Judge Entz"), 
attacks Plaintiffs and Plaintiff-Intervenors evidence on the ground 
that: (1) the data is based on total population and not voting age 
registered voters; (2) the analysis does not reflect changes in the 
distribution of population over time as a result of growth of 
Dallas suburbs and geographic dispersal of minorities; (3) Dr.

35



Engstrom did not control for absentee or Oriental votes; (4) there 
is a stronger association between partisan affiliation and success 
then there is between race and success; and (5) the analysis shows 
what happened, but not why it happened.“ In support of fiis fourth 
attack, Judge Entz argues that five of the seven elections analyzed _ 
involved Black candidates who are the candidate of choice, while 
all seven involved Democratic candidates who were the Black 
preferred candidate of choice. Thus, Judge Entz concludes that 
political party is a; better predictor of the Black preferred 
candidate and that candidate is a victim of partisan politics not 
discriminatory vote dilution.

'^ j . Dr. Engstrom testified-that: (1) he was never given-
precinct data by race and voting age registered voters; and (2) the 
range of support for the Democrat.candidates between 1980 and 1986 
varied 10 to 17 percentage points. Thus, Dr Engstrom concluded 
that something other than just straight party voting is going on 
in judicial elections. i'-. 1

k. Dr. Dan Wiser's testimony confirms Dr. Engstrom's 
results. Dr. Wiser's data set was based on 1980 Census data, 
Dallas County election returns and Dallas County precinct data 
adjusted for changes in precincts. Precincts that split were 
reconstructed by estimating the part of the precinct that shifted

36



to another and apportioning the registered vote based on the shift 
and past history. Testimony of Dr. Dan Wiser. The adjusted data 
was checked against the 1986 Justice Department submissions, id. 
Plaintiffs' Exhibit P-I D-H. Ninety eight percent (98%) of the 
vote in homogeneous precincts of 90% Black voters went to the Black 
preferred candidate. Plaintiffs' Exhibit P-I D-ll, D-16 through 
D-23a. At least 83% of the Black community vote supported the 
Black preferred candidate in homogenous precincts of between 50% 
and 90% Black. Id. ‘ ' t c

"1. Dr. Wiser calculates that the Asian community only 
comprised approximately 2^6% of the total Dallas County population 
as of 1985. Plaintiffs',Exhibit P-I D-03. He testified that the 
best estimate of the growth of the Asian community between 1985 and 
the present is supplied by the Bureau of Census. Plaintiffs' 
Exhibit P-I D-02. He believes there has only been a growth of 
approximately 3% between 1985 and 1988 and does not agree with 
estimates of Asian leaders in Dallas County.

m.-.; Plaintiffs and Plaintiff-Intervenors established the 
third Ginqles prong by showing that a white bloc vote exists. The 
Y intercepts-calculated by Dr. Engstrom for the Black preferred 
candidate ranged between 29 and 39 percent for the seven elections 
analyzed. Plaintiffs' Exhibit D-02. The highest Y intercepts were

37



61.8% and 71.7% for Judges Baraka and Wright respectively, the non­
preferred candidates. Id. The highest percentage of Anglo Cross 
over votes received by the preferred candidate of Black voters was 
approximately 39 percent. -1 Id.-,- 1980 race -involving Joan Winn 
White. There are 197 precincts in Dallas County that are 90% or 
greater white population. Plaintiffs' Exhibit P-I D-06 & 07.

n. This is corroborated by Dr. Engstrom's homogenous
precinct analysis and Dr. Wiser's analysis. r,.-Id. at-column 1. - 
This analysis demonstrates that an Anglo bloc vote exists. The 
Court finds on the basis of the exhibits and testimony of Dr. 
Engstrom and Dr. Wiser that the Anglo or white bioc vote in Dallas 
County is sufficiently strong to generally defeat the choice of the 
Black community. ..

o. Dr. Anthony Champagne testified that judicial elections 
in Dallas County were characterized by strong partisan affiliation 
rather than racially polarized voting. Dr. Champagne analyzed 
contested District Court general elections between 1976 and 1988. 
Plaintiffs' Exhibit P—I D-06-A. Dr. Champagne bases his opinion 
on the steady increase of Republican victories in Dallas County 
over time. Plaintiffs' Exhibit P-I D-07-A pp.1-2. Only seven (7) 
of the contested general elections analyzed involved Blacks 
opposing white candidates. Plaintiffs' Exhibit P-I D-09-A p. 1.

38



No Black candidate running on the Democratic ticket won a general 
election. Two Black candidates running as Republicans■won. Id. 
at 1. - The Court" noted, supra. that it was the non-Black vote 
that gave rise to the success of these two candidates . ■~ See Finding
of Fact 17. e. - • v ' • .rl., r. ..U

p. Dr. Taebel analyzed nine judicial elections -in which 
Blacks opposed Anglos. In eight of the nine, Blacks and Anglos 
voted differently. State Defendants Exhibit D-06 pp. 1, 13, 17; 
21, 37, 69, 73, 81 & 89; See Appendix B, Plaintiffs' Re-Evaluation 
of Dr. Taebel's Reports ("Re-Evaluation") for Dallas County p.l. 
The Black... preferred candidate won only once. Id. This sole 
victory arose in the 1988 Republican primary. Id. The Black 
choice won only five (5) of the other twelve primary and general
District Court and Appellate ! 'Court races analyzed. : Id.';
Plaintiffs' Re-Evaluation p. 2. k* •:' ' r.

18. Tarrant County. r- rr ' • c'̂ --
-• a. • Dr. Robert Brischetto ("Dr. Brischetto”) testified 

concerning on behalf of Plaintiffs and Plaintiff-Intervenors in 
Tarrant County and the remaining counties at issue. He weighted 
his analysis in all remaining counties. Dr. Brischetto used Black 
population data by precinct from the 1980 Census for thirty four

39



(34) precincts in Tarrant County where precinct lines had not 
changed. He analyzed four (4) elections in which Blacks opposed 
Anglos in Tarrant County (three judicial elections and the 1988 
Democratic Primary). See Appendix A, Plaintiffs' Exhibit Ta-02.

b. In Tarrant County and other Contested counties where
there was a large representation of three ethnic/racial groups, DrV 
Brischetto used multiple regression analysis. Dr. "Brischetto 
testified that this approach shows the effect of the percentage_of 
Hispanics in precincts, for example, upon the votes received by a 
minority candidate, when accounting for the effect of the 
percentage of Hispanics. The statistical calculation that shows 
the effect is called the "Partial r." :r- •

c. Dr. Brischetto calculated "Partial r" values of -87%, 
-80% and ~90%~ respectively for the three judicial elections 
analyzed. Plaintiffs' Exhibit Ta-02. "There was "a negative 
correlation in the 1986 Salvant - Drago race and the 1986 Sturns - 
Goldsmith race. Salvant and Sturns were Black candidates running 
as Republicans. They did not receive the support of the Black 
community. Id. Approximately 93% of the Black voters in precincts 
analyzed voted for Drago, while approximately 85% of Black voters 
voted for Goldsmith. Id. The likelihood that the estimates would 
occur by chance (significance level) was much smaller than .05.

40



Testimony of Dr. Robert Brischetto. Dr. Brischetto's regression 
analysis shows a strong relationship between race and voting 
patterns in Tarrant County. The strength of the correlation is 
dependent on the size of the number not on the positive or negative 
value assigned to it. The negative correlation in the Salvant and 
Sturns races merely reflects that as the percentage of Blacks in 
voting precincts increases, the support for Salvant and Sturns 
decreased.

d. The lowest r squared for these analyses is 
approximately 64% for the 1986 race for Criminal District Court 
Place 1. Race explains at least 64% of the variance in voting in 
all elections relied on by Plaintiffs and Plaintiff-Intervenors in 
Tarrant County.

e. Plaintiffs' Exhibit Ta-02 further analyzes the Jesse 
Jackson Democratic Presidential Primary in 1988. The Partial r for 
Jesse Jackson was 98%. Although the Jackson race was not a 
judicial election, its analysis corroborates the judicial elections 
analyzed. However, Dr. Brischetto testified that he would reach 
the same conclusions without considering the Jackson contest.

f. Dr. Brischetto's homogenous precinct analysis 
corroborates the results of his regression analysis. Plaintiffs' 
Exhibit Ta-02. It shows that Black voters in Tarrant County gave

41



more than 89% of their votes to the preferred candidate of Black 
voters in every election analyzed. ; c---- = • .-•

9* - -Dr* Brischetto also recompiled and reanalyzed Dr. 
Taebel's work concerning Tarrant County. Plaintiffs' Exhibit Ta- 
iO. Plaintiffs' Exhibit Ta-10 compiles all of Dr. Taebel's 
analysis of countywide elections for judicial positions when Blacks 
opposed Anglos. Dr. Taebel also found negative correlation of 
-63% and -60% in the Salvant and Sturns elections respectively. 
Id. While these correlation figures are not as high as those 
found by Dr. Brischetto, they still reflect a strong correlation. 
See Finding ;of Fact-16.b; last sentence-— 1 —  - -  ’ -

h * --D r * Taebel used bivariate regression in his analysis. 
Dr. Brischetto is of the opinion that had Dr. Taebel used 
multivariate: analysisv'. his correlation estimates would1 have' been 
more precise,,. Further Dr. Brischetto believes that the r values 
wouid bave been higher, because the analysis -would have eliminated ' 
the effect of Hispanics. while Dr. Brischetto did not agree with 
Dr. Taebel's statistical methodology, he reviewed Dr. Taebel's work 
because Dr. Taebel's data set was more complete. r-,.

i._ This Court finds, on the basis of all' of Dr. 
Brischetto's analysis, the Black community in Tarrant County votes 
cohesively in general elections for State District Court Judges.

42



j. The Court further finds that the Anglo bloc vote in 
Tarrant County is sufficiently strong to defeat the minority 
community's preferred candidate. In the three general elections 
analyzed, the preferred candidate of Black voters lost every time. 
This is true even though each of the Black preferred candidates 
had a sizeable percent of Anglo cross over votes. Plaintiffs' 
Exhibits' Ta-02; Ta-10. The Y intercept~ reflects that Anglo 
support for-the Black preferred candidates was between 42% and 49%. 
Id. Ta-02.^ec. ,, .. . ; U; ; .... • . ... •>.

_k. The testimony of Plaintiff and sitting District Judge 
Maryellen Jiicks corroborates: this analysis. ' Judge Hicks is Black. 
She testified that the only time she ran against an Anglo in a 
countywide judicial election she lost. Plaintiffs' Exhibit Ta-10,- 
County Criminal Court Place 1.- She feels that she lost because she - 
could not convince Anglos to vote for her. She also believes that 
she could«not win if she had Anglo opposition because of the Anglo 
vote. - rv ; ■ : . ,,r,. . .;m,, ... .

1. -.Judge Hicks testified that implementation ; of single 
member districts in Tarrant County Jiad .immediate effects. .Before 
the districts went into effect, only two Blacks had been elected 
to School Trustee positions. Since single member districts were 
implemented, two Blacks and one Hispanic have consistently been

't j f r on

43



Trustees. Two Blacks and one Hispanic also took office on the Fort 
Worth City Council as a result of single member districts being 
implemented for that body. Further, after single member districts 
were established for State Representative offices, two minorities 
were elected to the Texas House of Representatives.25. - r. .

m.' In the five primary and general judicial elections- 
involving Black candidates analyzed by Dr.~Taebel, the Black choice 
won only once. State Defendants Exhibit D-39 pp. 1, 29, 33, 37 & 
57; See Appendix B, Re-Evaluation for Tarrant County p.l* It is 
clear that Blacks and Anglos voted differently in these races, id. 
In District Court general- elections that did notr have a Black 
candidate, the oandidate preferred by Black voters won three (3) 
of five (5) times. Id. D-39 pp. 13, 17, 21, 25 & 61; Re-Evaluation 
at 1-2': ~-In-.three other ̂ judicial -general elections the-candidate 
of choice of the Black community won all three times. Id. D-39 pp. 
9, 49 &-65? -Re-Evaluation at 2. Two of the three were Appellate 
Court elections, while the third involved the County Court at Law. 
Id. The candidate of choice also won all three primary judicial 
elections analyzed by Dr. Taebel. Id. D-39 pp. 5, 41 & 49.' --:

After the lines were redrawn in 1982, one minority has been 
elected; rr "-

44



19. Bexar County.
-a. Dr. Brischetto based his analysis of Bexar County on 

Spanish surnamed registered voter data by precinct from the office 
of the Secretary of State of Texas. Dr. Brischetto testified that 
this data was the closest measure of actual registration data by 
precinct. Dr. Brischetto used bivariate regression analysis in 
Bexar County because of the very small Black population in^the
County r .-i--- ------  - • - •_

- b. He analyzed six (6) general elections from 1980 to 1988 
in which Hispanics opposed Anglos. See Appendix A, Plaintiffs' 
Exhibit- B-02. He calculated "rJ1 values for Hispanic preferred 
candidates between 86% and 88%.v Id. f His regression analysis shows 
a strong relationship between race and voting patterns in Bexar 
County. In all but one race, ras the percentage :of'Hispanics 
increased?'support for the Hispanic preferred candidate increased. 
Dr. Brischetto testified that the probability that 'correlation of 
this size would happen by .chance was much smaller '-than the 
generally accepted level of .05.26 • c ' • h : . - j:

c. In the 1982 Barrera - Stohlhandski race, the Hispanic

The significance level for each election is .0000. 
Plaintiffs' Exhibit B-02. Dr. Brischetto testified that there was 
practically no [or zero] probability that these correlations would happen by chance.

45



candidate, Roy Barrera, Jr. running as a Republican, received very 
little Hispanic support. The correlation coefficient for Mr. 
Barrera was -80%. Id. As the percentage of Hispanics in voting 
precincts increased, Barrera's support decreased. Barrera received 
approximately 17% of. the Hispanic vote. Id. He was not the 
preferred candidate of Hispanic voters in Bexar. County. o :l ..

•::i■ d.;..... The lowest r : squared for - these r analyses is 
approximately 64% for Mr. Stohlhandski, an Anglo running as a 
Democrat in the 1982 Barrera - Stohlhandski race. The highest r 
squared was 77% for the 1986 Cisneros - Peeples race. This 
demonstrates inBexar County that race explains at least 64% to 77% 
of the variance in voting in all six elections.

e. Dr. Brischetto's background and homogeneous precinct 
analysis confirm the fact that iHispanics are politically cohesive 
in Bexar County. Dr. Brischetto lives in Bexar County and analyzed 
election behavior there in a Section 2 case involving the San 
Antonio River Authority.-., Plaintiffs' Exhibit B-16v- There he 
found polarized voting along racial and ethnic lines in a 
nonpartisan election involving low profile campaigns. Dr. 
Brischetto's homogeneous precinct analysis shows that Hispanic 
voters in Bexar County gave 73% to 93% of their votes to the 
preferred candidate of Hispanic voters in every election.

46



f. Dr. Brischetto controlled for absentee votes in 1988 
elections based on allocated data from the Bexar County Elections 
Administrator. He testified that the additional data did not • •
change his conclusions. .....

• g. Plaintiffs presented evidence from four hypothetical 
districts carved out of existing precincts for each of the six 
elections analyzed.- Plaintiffs' Exhibits B-12a - 12e. Almost ~' 
always, the Hispanic candidate who actually lost at-large would 
have won if he had run from a hypothetical majority Hispanic y ̂  a ., 
district. _.__In one case, the 1988 Republican primary between^ 
Arellano and White, the Hispanic^-candidate won in cnly three of the.-.wr, = , 
four hypothetical districts.~ Id. B-12e. . ■ • v--- — , — *na

h. In the 1988 Arellano — White Republican primary for the I
150th District Courts Arellano * ran...as jianii appointed Incumbent. <- ross 
White,.an Anglo, decided late in the campaign that he did not want
to run for office. .. It was too -late to withdraw, but he endorsedro get. 
his opponent Arellano. White nevertheless -won. Adam Serrata Judae 
testified in his deposition that this was a classic example of r.-iw.: 
polarized voting. Deposition Summary of Adam Serrata ("Serrata 
Depo. ") . . hss

i. Other testimony suggests the same conclusion. J u d g e .. 
Anthony Ferro testified in his deposition that he -ran for County ’.‘.n-rr.

47



Court at Law four times in Bexar County. He won two races were he 
did not have Anglo opposition. Deposition Summary of Anthony Ferro 
("Ferro Depot") at 1. Both Messrs. Serrata and Ferro testified 
that it is not possible to get elected in Bexar ^County to the 
position of District-Judge without Anglo support. Id.; Serrata v 
Depo. — . - • -u: --.

- j .- Dr/ Brischetto further concluded that the-Anglo bloc , 
vote in Bexar County is sufficiently strong to defeat^the Hispanic - 
community's preferred candidate. In the six elections analyzed, th 
the preferred candidate of Hispanic voters won only once. See 1988 
Mireles - Bowles race. The Y intercept reflects that non-Hispanic 
support for the Hispanic preferred candidates was between 18% and 
35%. It is' not surprising that the one Hispanic candidate of 
choice who won also received_the highest percent of Anglo cross 
over votes. -r‘ >.1-,

k. Judge Ferro testified that he has only been able to get w  

elected when he did not have an Anglo opponent. Ferro Depo. Judge 
Paul Canales testified that voters in Bexar County pay attention 
to the race/ethnicity of candidates in-judicial elections. -

l. The effect of fairly drawn single member districts has 
had a positive effect on minority election results in Bexar County. 
Immediately after the creation of single member districts in White - -

48



v. Reoester. Hispanics were elected to the Texas House of
Representatives. further, immediately after the City Council 
implemented single member districts, the number of minorities on 
the San Antonio City Council increased. Serrata Depo.; Ferro Depo.

• m. Whites and Hispanics voted differently in 28 of the 29 njn e t  

judicial elections involving Hispanic candidates in Bexar County. -,_j 
State Defendants Exhibit D-07 pp.. 2-5U.-18; See Appendix B , Re- - 
Evaluation for Bexar County p.1-2. In the twelve general elections 
analyzed by Dr. Taebel, the Hispanic preferred candidate won three .sLr-r 
(3) times. Id. D-07 pp. 4, ’5, 7, 15-16, 18-21 & 25-28; Re-
Evaluation at 1. Only one of those was a District Court election.
Id. D-07 at 5. The Hispanic choice won six (6) out of 18 primary cv ~ . 
elections. Id. Re-Evaluation a.t 1-2.  ̂ -- ■ r

' J ' v— —  '■ ^ • • ■ • j . « .  j  • -  - i v 4

‘•■20 .anTravis County. ' r ? • -nv/ed.
Sc.a.r.ppDr. Brischetto analyzed othree (3) 1988 countywide rved

judicial elections in Travis County: one primary election for the one 
345th District Court and two County Court, at Law general elections. t.;- 
Dr. Brischetto testified that there has only been one Hispanic - 
Anglo District Court election between 1978 and 1988. . In that race, 
the Anglo won. Plaintiffs' Exhibit Tr-11; Testimony of Jim 
Coronado. Mr. David Richards testified that the Republican party nonov

! ■ i ; ;  : n t  | - v  i'i i i i •

49



is insignificant in Travis County. Hence, Mr. Richards concluded 
that the Democratic Primary is the true testing ground for opposed 
candidates in judicial elections.

b. Dr. Brischetto used Hispanic population data by 
precinct from the 1980 Census reconfigured +to 1988 'precinct 
boundaries. He based his polarization and homogenous precinct 
analysis ton total population figures for Blacks, Hispanics and 
Anglos in approximately 178 precincts ̂ (virtually ali^of t-them) in 
the County. Amalia Rodriguez Mendoza^ the Travis County Registrar 
of Voters, provided the data.

—  c » Dr. Brischetto 's~multivariate or multiple regression 
analysis shows that the Hispanic community in Travis County is 
politically cohesive, when the effect of the Black vote is 
considered. Dr. Brischetto calculated "Partial r" valges~of -84%, 
85% and 90% respectively for the three judicial elections analyzed. 
See Appendix A, Plaintiffs' Exhibit Tr-02. The Hispanic preferred 
candidate received at least 77% of the Hispanic vote ‘ in one 
election27 , 93% in the Democratic Primary election and 95% in the 
Garcia - Phillips race. Id. The likelihood that the estimates 
would occur by chance (significance level) was much smaller than

rThe 1988 County Court at Law race between Castro Kennedy 
and Hughes. Castro is the Hispanic preferred candidate. 
Plaintiffs' Exhibit Tr-02.

50



.05. Testimony of Dr. Robert Brischetto. Dr, Brischetto's 
regression analysis shows a strong relationship between race and 
voting patterns in Travis County. : ~- • y

d. The homogenous precinct analysis for Travis County 
establishes 4a similar pattern. Plaintiffs' Exhibit Ta-02. It? 
shows ■that Hispanic voters gave more than 63% arid as high as 90% 
of their votes to the Hispanic preferred candidate.

e. Dr. Brischetto also reanalyzed the same three elections 
using bivariate regression analysis based upon voter registration 
data. See Appendix J A, -Plaintiffs' Exhibit ~'Tr-19. These 
correlation figures are very close to those -calculated using 
multivariate analysis, and clearly reflect strong correlation;- See 
Finding of Fact 16.b. last sentence. With either data set," Dr. 
Brischetto's analysis shows that as the percentage of Hispanics in 
precincts increase, so does support for the Hispanic, preferred 
candidate. The r squared figures all exceed approximately 64%.- -

1 f. The Hispanic preferred candidates took the majority of 
the votes from Plaintiffs' hypothetical districts even though they 
lost countywide. Plaintiffs' Exhibit-Tr-12.

g. The State Defendants were concerned that Plaintiff's 
did not analyze Statewide judicial or legislative elections. See 
Cross examination of Jim Coronado; Cross examination of Dr.

51



Brischetto. Dr. Brischetto testified that Plaintiffs focused on 
local elections when that data was available and these elections 
were not reached in Plaintiffs' hierarchy of priority. He further 
testified that the elections analyzed were the closest in nature 
to District Court elections. Dr. Brischetto felt that once he had 
three elections he could determine-a sufficient pattern.,- - This
Court agrees;28 - ; ...

h.- The State Defendants attack Dr. Brischetto-'s analysis ■ w *  

on the ground that he did not take “into account: (1) absentee buoPi­
voting;-and ( 2 )  the number of non-United States citizens, Blacks • ■ 
or Anglos with Spanish surnames “in Travis County.

urivr Dr. Brischetto;controlled for absentee-votes in 1988 t-i 
elections in Bexar County. He testified that Bexar County had the 
highest absentee voting than anywhere in the State.r JHe_;concluded:~c -ir. 
in his Bexar County analysis that absentee voting did not change >.\i lv 
his conclusions. See Findings of Fact 19.f. This Court: .finds ui 
that the results would not be significantly different in Travis5 panic 
County. .,. . .. . . . a

j♦ Spanish surname counts were based on persons who - 
identified themselves in Census counts as being of Spanish origin.

28 •Ginqles itself relied on only analysis of three elections 
in Senate District 22 (1978, 1980 & 1982) and House District 21 
(1978, 1980 & 1982). Ginqles. 478 U.S. at Appendix A.

52



While the Court recognizes that the Census definition of Spanish 
origin includes many parallel ethnic backgrounds, this Court finds 
that the probability of overlap of Black and/or White voters is 
very slight. _ ... .

k. Finally, the State Defendants claim that the analyses 
of the ‘ Democratic' Primary between Judge Gallardo (the Hispanic 
preferred candidate) and McCown is misleading. Witnesses for the 
State Defendants testified that Judge Gallardo lost because he was 
a bad judge. Depositions of Becky Beaver & Fernando Rodriguez; 
Testimony ! of David Richards. "While this may be true, under 
controlling law, it is the correlation between- the race of the 
voter and the selection of certain candidates that is .crucial to 
this Court's inquiry. Gingles, 478 U.S. at 63.

l. The Court further finds that the Anglo bloc vote in
Travis County is ̂ sufficiently ̂ strong to defeat the minority 
community's preferred candidate. The preferred candidate of; 
Hispanic voters lost each election analyzed. Two of the Hispanic 
preferred candidates received approximately one third Anglo cross 
over votes. Plaintiffs' Exhibits' Tr-02; Tr-19. The other
candidate received only approximately 14% Anglo cross over votes. 
Id., Tr-02.

m. In each of the hypothetical districts, the candidate

53



of choice of the Hispanic community received the most "votes; in two 
districts the candidate of choice received a majority.

n. Dr. Taebel analyzed the same three elections analyzed 
by Plaintiffs' expert.■ State Defendants Exhibit D-08; See Appendix 
B, Re-Evaluation for Travis County p. 1. .His analysis confirms 
that in these three races whites and Hispanics voted differently 
and the Hispanic preferred candidate lost each time. Id. D-08 pp. 
33, 37, 41... TheoHdspanic preferred-candidate -fared - better in
Appellate elections winning one primary ,runoff and two general 
elections. .Id. D-08 pp. 25, 29 & 45. Hispanic and white voters 
did not vot-edifferentiyrinr these three election contests but did 
so in ■ the '1984 and:: 1986 Democratic _ primary for County Court 
numbers 1 and 7. Id. D-08 pp. 33 & 41* . , .

21- Jefferson Countv l co s ,
a •' Brischetto used Black population data by precinct'

from the 1980 Census for all of his analysis in Jefferson County.. 
He testified that population had changed -very little in Jefferson 
County. Plaintiffs'_Post Trial Brief at 95. Only those precincts 
that retained unchanged boundary lines were used in his analysis.

b. He analyzed five (5) Democratic primary elections, two 
(2) Democratic primary runoffs and the 1988 Presidential Democratic

54



primary. See Appendix A, Plaintiffs' Exhibit J-02 pp. 1-2. Four 
of the five primaries analyzed involved Justice of the Peace 
contests. The fifth was :for a' County Court at Law post. ; Dr. 
Taebel did not analyze any of these elections. State Defendants' 
Exhibit D—09» Each of the Justice of the Peace election precincts 
covered at least an entire city which are the-largest urban areas 
of the County. Precinct 1 covers the City of Beaumont, Texas." 
Precinct 2 covers the City of Port Arthur, Texas. Tom Hanna 
testified in his Deposition that running for1 office from these 
precincts is equivalent to running at large from the two cities.

Brischetto testified that'there-were no primary or general- 
elections for District Court seats that pitted Black against Anglo.

c. Dr. Brischetto^used multivariate regression analysis 
in his examination of Jefferson County separating out the effect- 
of the Hispanic votei He calculated "Partial r" values between 66% 
and 97% for the judicial primaries and runoff elections analyzed. 
Id. The partial r for the Black preferred candidate in the 
Democratic Presidential Primary, Jesse Jackson, was 97%. Id. The 
livelihood that the estimates would occur by chance (significance 
level) was much smaller than .05* Id. Dr. Brischetto's regression 
analysis shows a strong relationship between race and voting 
patterns in Jefferson County. The Black preferred candidate

55



received a clear majority of Black community support in at least 
five of the seven judicial contests analyzed. Id., multivariate 
and homogeneous analysis for 1972 to 1978. In the 1982 primary for 
Justice of the Peace, Precinct * 1, Blace- 2 , the Black preferred 
candidate Cannon received approximately 51% of the Black community 
vote, while two opponents split the remaining 49%.

' ~d. In one instance, the Black preferred candidate did not
receive . a. .majority,; of the Black community vote. In the 1986- 
Democratic Primary for Justice of the Peace, Precinct 1, Place 2,
the Black preferred candidate, Wilmer Roberts, only received 47% 
of the Black"vote ^40% in homogeneous precincts). The other 53% 
(60% in-homogeneous precincts^ was split between four candidates.
John Paul Davis, a Black attorney from Jefferson Country, testified 
in his ̂ Deposition that he supported the:white candidate because she 
was the most liberal at the time he made his choice and Mr. Roberts
announced late-in the race.

' e * The r squared figures range from 44% for one race (1972- 
runoff) to 94% for three races (1978 & 1982 judicial primaries and
1988 Presidential primary). It is clear from Dr. Brischetto's 
analysis of voting patterns in Jefferson County that as the 
percentage of Blacks increase in a precinct, the percentage of 
support for the Black preferred candidate increases.

56



. f. Dr. Brischetto examined the votes cast in a 
hypothetical district for the 1978 Democratic Primary between Mr. 
Davis and an Anglo opponent. The analysis shows that Davis 
received more votes in each precinct and a majority of the vote in 
the district.. Plaintiffs' Exhibit J-09.

g. State Defendants argue fhat the three races analyzed 
in 1982-, -1986 and 1988 either show no racial polarization or a 
victory for the Black preferred candidate. This Court disagrees. 
As the Court discussed, supra. the Black preferred candidate was 
supported -by a majority of the Black community in the 1982 
Democratic Primary. See Finding of Fact 21.c. With-reference to 
the 1986 Democratic Primary, the Court finds that the State 
Defendants' evidence is -not conclusive that the Black community 
either would not have cohesively supported Wilmer Roberts had he 
announced earlier or that the Black community cohesively, supported 
some other candidate. Plaintiffs' Exhibit J-02; Finding of Fact 
21.d. rThe Court further finds that while Jesse Jackson.may have 
carried Jefferson County in the 1988 Presidential Primary, that 
fact alone is a far cry from whether the Black preferred candidate 
is successful;in Jefferson County. n.

1 h. State Defendants further point to the 1984 Democratic 
Primary between John Paul Davis and Donald Floyd, both of whom are

57



Black, to demonstrate that the Black community is not politically 
cohesive in Jefferson County. While Mr. Floyd won the primary and 
the election, Defendants did not demonstrate that the Black 
community split their vote or failed to support one candidate-over 
another. - -- rne

. i. This Court finds on the basis of. the foregoing 
discussion'^that the Black community in Jefferson County votes 
cohesively in judicial elections. ___  ____ _______________

j. In at least f ive iof \the seven elections analyzed it is
clear that blacks and whites voted differently and the preferred 
candidate of Black voters lost every time. The Black communities 
candidate of choice received 25% to 41% of the Anglo cross over': 
vote in election years 1972 and 1974. The percentage dropped 
thereafter to a low of 2% for Wilmer Roberts in 1986 and arrange 
of 7% to 10% for the other two judicial races. Plaintiffs'
Exhibit' J-02. Although the Black preferred candidate received 70% 
to 93% of the Black community vote in five of the seven elections 
analyzed they still lost countywide.

k. The Court finds that the Anglo bloc vote in Jefferson 
County is sufficiently strong to defeat the minority community's 
preferred candidate.

l. “No Black attorney has run for the position of District

58



Judge in Jefferson County. Deposition Summary of John Paul Davis. 
Mr. Davis feels that Black lawyers do not run for the-‘office 
because of the high probability of defeat. Id.

m. Implementation of single member legislative and 
Commissioner's Court districts resulted in the election of Black 
preferred candidates to those positions. Deposition Summary of 
Thomas Hanna. • • •• ■ r " .. • = • -

’ 22 s Lubbock Countv. - - : ; l. . n. :
a.' Dr. Brischetto used population data from the 1980 

Census.precinct boundaries-for his analysis in-Lubbock County. He 
initially based his review on 30 ofi 76 precincts which had not 
changed between 1980 and the relevant elections analyzed He 
analyzed additional precincts that he was able to reconfigure, by 
use of Census block maps'.29

* b. Dr. Brischetto relied on appellate judicial contests. 
He testified that no relevant local judicial contests involved a 
minority opposed by an Anglo candidate. He further testified that 
he did not analyze local Justice of the Peace races because the

He analyzed 48 of the 76 total precincts in the 1986 
primary, 44 of 7 6 in -the 1986 runoff, 48 in bhe 1986 general 
election and 47 in the 1988 general election. See Plaintiffs' Post 
Trial Brief at 109 n. 55.

59



Justice of the Peace precincts were not at least as.:large as a 
major city. He analyzed two (2) Supreme Court general elections, 
two (2) Democratic primary elections and two (2) Democratic primary 
runoffs. . See Appendix.A, Plaintiffs' Exhibit L-02, pp. 1-3.^

- lC. . Dr. Brischetto used bivariate, multivariate regression 
and homogeneous precinct analysis in his examination of Lubbock 
County. The bivariate analysis produced correlation coefficients 
in excess of 87% with a corresponding r square figure of 76%. id. 
He used multiple regression analysis to show that Blacks and 
Hispanics vote together. This analysis revealed that the two 
groups favored the same candidates in each elect ion^r^JTd..T.TTThe 
lowest-partial r calculated for Hispanic voters was 78% in the 1986 
Democratic Runoff for Supreme Court place 4. The lowest partial 
r for Black-voters was>56% in the 1986 Democratic Primary for the 
same Court prior to the runoffrelection. The likelihood that the 
estimatesMwould occur by chance (significance level)-, was- much 
smaller than .05. Id. Dr. Brischetto's regression and homogeneous 
precinct analysis ..shows a strong relationship between
race/ethnicity and voting patterns in Lubbock County. The
combined minority preferred candidate received a clear majority of 
combined minority community support in each election analyzed. Id. 

" d. It is clear from Dr. Brischetto's analysis of voting

60



patterns in Lubbock County that as the percentage of minorities 
increase in a precinct, the percentage of support for the minority 
preferred candidate increases. This Court finds that Blacks and 
Hispanics are cohesive as -a group in Lubbock" County judicial 
elections i - ' - i . ... : r ' . . ...: ; - f ;'

e. * ' Maria Luisa Mercado, a Hispanic attorney from Lubbock- 
County, testified that Blacks and Hispanics work together in the 
County on many significant issues. Deposition Summary of Maria 
Luisa Mercado ("Mercado Depoi1").

f. The State Defendants point to the 1984 race for Justice 
of the -Peace -between Sedeno, a Hispanic candidate running as a 
Democrat, - against a Black‘Republican candidate/McKinley Shephard 
to illustrate that Blacks and Hispanics do not vote cohesively as 
a group. The Black boxes voted for Mr. Shephard. Mercado Depo. 
at 2. ; Dr.'-'Brischetto testified that this race was not analyzed 
because the Justiceeof /the Peace precinct in question split the 
City in half . ’• "It did not include a large majority of the County 
or a large metropolitan area." Testimony of Dr. Robert Brischetto. 
This Court finds that the Sedeno - Shephard race does not 
illustrate that Blacks and Hispanics do not vote cohesively in at-
large judicial elections. The Court further finds that Blacks and 

Hispanics opposing each “other says less about the collective

61



cohesiveness of the two groups when either opposes an Anglo.3® --------
g. ' Minorities and whites voted differently in each

election analyzed. However,-the preferred minority candidate won 
on two .of six occasions. Plaintiffs' Exhibit L-02, p. 2, 1986 
Primary for Court of Criminal Appeals, Place 1 and 1586 Runoff;ty-?Vsr- 
In one of those two races^i the minority preferred" candidate 
received 46% of the Anglo cross over votes from homogeneous white 
precincts of 9 0 % to_100% white population.. Id., 1986 Runoff. The 
minority communities candidate of choice received 39%, 40% and 41% ;p .. . 
of the Anglo cross oveir vote, respectively, in three other 
elections:andLstill lost.. Id.,-1986 General Flection, 1988 General- - 
Election and ; 1986 Democratic Primary, ^Supreme Court, Place 4 , v ’
respectively. The Court finds that the Anglo bloc vote in Lubbock 
County is sufficiently strong to defeat the minority community*smemnei.' 
preferred candidate.

h. Defendants argue that Justice Gonzalez may possibly 
have received more Anglo votes in the 1986 Democratic Runoff with 
36% than either of his three opponents, assuming the remaining 64% 
of the Anglo votes were evenly split. Defendants conclude on that 
basis that Anglos did not vote overwhelmingly against Justice ■ :

State Defendants further point to Hispanic - Black state 
representative races in Lubbock County in 1984 and 1986.

62



Gonzalez. This Court disagrees. Assuming arguendo that Defendants 
assumption is correct, the Court finds that Anglo's did - 
overwhelmingly vote against Justice Gonzalez even if they did not 
vote overwhelmingly for a different candidate. " - ' / v y ”

Cci~-u‘ Dr. Brischetto testified concerning some countywide > ^ 
elections in Jones v. City of Lubbock. 727 F.2d 364, 383 (5th C i r . - - -  
1984). His conclusions in Jones corroborate his testimony in this • ~ •
C A S S  • j -  •— «w : c i l ia  X y  Z 6 u  LWO ( A j iSupTHUit- {.’0*1 T  »’■ ci J.

Ms. Mercado testified that - Black and Hispanic. " ;; 
candidates have not been successful in at large elections. Mercado ---- 
Depo. -She'testified that:she-carried all minority boxes and zero 
Anglo boxes in her 1978 bid for City Council. Id. Blacks and 
Hispanics have been successful running for School Board and County 
Commissionerfs positions after the implementation of single member---the 
districts Id. ■. cm.j--̂ c.iiworiv gi i— ■. ■ ■ k :_ i: -i~v-. t — .

l-n k;j - Dr/ Taebel only analyzed two uof the same Appellate ̂ panic 
Court contests analyzed by DrBrischetto.ys. State Defendants' ntior 
Exhibit D-10 pp. 17 •& 25. In both/ minorities and whites voted "in­
differently and the minority choice lost. - Similar results were - - j. -'1 
obtained in two County Court at Law General Elections analyzed by 
Dr. Taebel; Jd. D-10 pp. 5 & 9. However, in those two races there if 
was no ' minority candidate. See Appendix B, Re-Evaluation of for

63



Lubbock County.

23. Ector County. - ; —  .< “ ■ i?-
a. Dr. Brischetto used population data from the 1980 

Census precinct boundaries to analyze 24 of the 31 total precinctsr- 
in Ector County which had not changed between 1980 and the relevant 
elections analyzed. As in Lubbock County, he relied on appellate ; 
judicial contests. He analyzed two (2) Supreme Court General 
Elections and two (2) Democratic Primary Elections. See Appendix
A, Plaintiffs’ Exhibit EP-02, pp. 1-2.TT He testified that "no County 
or District Court contests involved a minority opposed by an Anglo-: _ 
candidate. - -- ----- .— --- --

b. iDr. Brischetto used the same statistical analysis used
in Lubbock County. Bivariate analysis was used to separate the 
white and aninority votes. 'Testimony of ‘Dr. Robert Brischetto. 
Multivariate analysis was used to separate the/Black and Hispanic .
vote. Id.— The bivariate analysis produced correlation . 
coefficients in excess of 78% with a corresponding r square figure
of 61%. Plaintiffs' Exhibit E-02, pp. 1-2. Multiple regression 
analysis shows that Blacks and Hispanics vote together. This - -  

analysis revealed that the two groups favored the same candidates 
in each election. Id. The lowest partial r calculated for

64



Hispanic voters was 46% in the 1986 Democratic Primary for Supreme 
Court Place 4. The lowest partial r for Black voters was 60% in 
the 1988 General Election for Supreme Court, Place 3. The 
likelihood.that the estimates would occur by chance (significance 
level)- was 1 much smaller than .05. Testimony of Dr. Robert 
Brischetto. <«<.••! _ .j- _  , w  . ;'-a ai -

 ̂ c ^ ~  rA clear majority of the combined minority community 
supported-_the . .preferred minority • candidate in each- election 
analyzed. Even in the race for Supreme Court, Place 4, ; Justice 
Gonzalezreceived 42% of the Hispanic vote and 65% of the Black 
community-vote. Id. Dr. Brischetto's regression and homogeneous 
precinct analysis shows" '• a •/n strong relationship between 
race/ethnicity and voting patterns in Ector County. The lowest 
level of combined support is reflected as .50% in the Democratic 
Primary for }Supreme Court, Place 4. Id., Homogeneous precinct 
analysis, p. 2. Dr. 'Brischetto attributes the lack, of stronger 
minority group cohesiveness in that race to the iact that one of 
the candidates in the ^Primary was > from Ector County. Id.. 
Candidate Gibson. However, in the General Elections for 1986 and 
1988, homogeneous precincts of 80% or more combined minority gave 
more than 80% of their vote to the minority preferred candidate.

d. :it is clear from Dr. Brischetto's analysis of voting

65



patterns in Ector County that as the percentage of minorities 
increase in a precinct, the percentage of support for the minority 
preferred candidate increases. This Court finds that Blacks and 
Hispanics are cohesive as a group in Ector County judicial 
elections. : • • ... . c OU L̂ VLiU-’ .

e. Minorities and whites voted differently in each 
election analyzed. Minorities supported the minority preferred 
candidates'in much, greater percentages than Anglo voters. The 
preferred minority candidate won only one race analyzed. See 
Plaintiffs^ Exhibit"^-02; ;p:? 2, 1986 Primary for Court of Criminal 
Appeals, Place 1. —  -- —  ■■ ..... ■ .

f M i n o r i t i e s  have been elected to Justice of the Peace 
and County Commissioner's positions from predominately minority 
precincts. Deposition Summary of Lawrence Leo Barber ("Barber 
Depo. ") . : ’• ’ -l-iini: i £ ’ * • .;i . . r. - . r r.ir- ■

:-! g. -T 'Dr. Taebel*s analysis of‘the same two Appellate Court 
contests confirmed Dr.?BriSchetto's analysis. State Defendants' 
Exhibit D - U  pp. 21 & 37. In both, minorities and whites voted 
differently and the minority choice lost. Dr. Taebel further 
analyzed five (5) General Election judicial contests that did not 
involve positing an Anglo against a minority. Id. pp. 5, 9 , 13, 
29 & 33. Minorities and whites voted differently and the minority

66



preferred candidate lost in three of the five. See Appendix B, Re- 
Evaluation of Ector County. .

24. Midland County.
a. Dr. Brischetto based his analysis on population data 

from the 1980 Census.= He analyzed 11 of the .36 total precincts for 
1986 and : 10 of 36 for 1988 that had boundaries that had not 
changed.. =... He was also able, • to reconfigure boundaries for 22 
precincts in 1986 and 23 in 1988. Testimony of Dr. Robert 
Brischetto. He relied on appellate races and one Justice of the 
Peace race since there have been no local - countywide-election 
contests: in which minorities opposed Anglos. The Justice of the 
Peace race encompassed the entire City of Midland. Testimony of 
Aquilla Watson. He analyzed three elections in.-“total. See 
Appendix A,zPlaintiffs' Exhibit M-02. Dr. Taebel did not analyze 
the Justice-of the. Peace contest.

b. Dr. Brischetto used bivariate regression analysis in 
Midland County. The bivariate analysis produced correlation 
coefficients in excess of 89% with a corresponding r square figure 
of 79%. Id. Better than 85% of the combined minority voted for 
the minority preferred candidate in each race... Id. The likelihood 
that the estimates would occur by chance (significance level) was

67



much smaller than .05. Testimony of Dr. Robert Brischetto. Dr. 
Brischetto's regression and homogeneous precinct analysis shows a ~ 
strong relationship between race/ethnicity and voting patterns in 
Midland County. ...

c. It is clear from Dr. Brischetto's analysis of voting
patterns in Midland County that as the percentage of minorities -. 
increase in aprecinct, the percentage of support for the minority H n c  

preferred candidate increases. This Court finds that Blacks and-- uwo 
Hispanics are cohesive as a group in Midland County judicial - •••
elections^ ---v - K- . L io n s :  1 : * - ' • ■_

d. It is further clear that minorities and whites voted
differently in each election analyzed. - Minorities supported the 
minority preferred candidates in much greater percentages than 
Anglo voters. The preferred minority candidate lost each race 
analyzed despite the large percentages--of combined minority 
support. Id. L - :

e. This analysis is supported by Dr. Brischetto' s analysis and 
and testimony in Lulac v. Midland ISP. 648 F.Supp. 596, 600 (W.D.
Tex. 1986), aff'd. 812 F.2d 1491 (5th Cir. 1987), vacated 829 F.2d 
546. Plaintiffs' Exhibit M-05. :

f* Aquilla Watson testified that she received very few 
Anglo votes. She only carried four (4) of the thirty-six (36)

68



precincts. Only one of the four included some Anglo cross over 
votes. Testimony of Aquilla Watson. 15 Minorities have been elected- 
to the School Board and County Commissioner's Court from 
predominately single member districts. Id. : . . ,

' gi: Dr. Taebel analyzed four (4) judicial contests iniwhich 
a minority candidate ran against one or more white candidates. 
State Defendants' Exhibit D-12 pp. 9y 21, 25 & 29. Minorities and 
whites^voted differently and the minority choice lost in the two 
General Elections analyzed, ,Id. pp. 25 & 29. The minority choice 
also lost in both primary elections, but there is some indication 
that minorities and some white voters “voted the same. Id;- pp;- 9 
& 21. ■ See Appendix BVrRe-Evaluation of Midland County. z .7 \ Q

i. r.V ff i (•."! f * • :n .
2.' ACCESS TO THE POLITICAL PROCESS c r  ! hr H

- er vi i.c ■ History of Discrimination -ding their doe
25. The effect of past discrimination against Blacks and 

Hispanics in areas such as education,.: employment and health in most 
of the Counties in question is either well chronicled or 
undisputed. See, e.q. , Lulac v. Midland ISP. 648 F.Supp.^ 596, 600 
(W.D. Tex. 1986), aff'd,-812 F.2d 1491 {5th Cir. 1987), vacated 829 
F • 2d 546; Campos v, City of Baytown.'840 F.2d 1240, 1243 (5th Cir.-

69



1988), reh'q denied. 849 F.2d 1240, cert, denied. _ _  U.S. ___
( 1989); Lipscomb v. Jonsson. 459 F.2d 335 (5th Cir. 19.72); Graves 
v. Barnes. 343 F.Supp. 704, 725 n. 15, 730-34 (W.D. Tex. 1972), 
rev'd in part and remanded sub nom. White v. Regester. 412 U.S. 755 
(1973), on remand, 378 F.Supp. 640, 644 (1974); Terrazas v.
Clements. 581 F.Supp. 1329, 1334 (N.D. Tex. 1984); United States 
v. Texas Ed. Aqcv.. Etc., 564 J . 2d 1£2. 163 (5th Cir. 1977), reh'q 
denied. 579 F.2d 910 (1978), cert, denied. 443 U.S. -915 (1979); ; 
Blackshear Residents Organization v. Housing ^uth. of City of 
Austin, i 347 F.Supp. 1138 (W.D. Tex. 1971); Jones v. City of
Lubbock, 727 F.2d 364, J83 . . ( 5fth . Cir. 1984); United -States v.
CRUCIAL, 722 f.2D 1183, 1185 (5th Cir. 1983). See also Plaintiffs

1 •• : • a ' r i iand Plaintif f-Intervenors Exhibits reflecting ̂ s o c i a l
stratification. • . s. • v r̂ -v \~r,-y ;;y -

26.i This history touched upon many-aspects of the Jives :gf
minorities' in the Counties in question including their access to
and participation in the.democratic system governing this State
and their socio-economic status. "The administration of -justice
in Texas was overwhelmingly dominated by Anglo males in 1968, and
the overall pattern [had] changed very little" by 1978.
Plaintiffs * General Exhibit ("Gen") 02, Texas: The State of Civil
Rights (Ten Years Later, 1968-1978^, A Report of the Texas Advisory

70



Committee “to the United States Commission on Civil Rights at 22 
(1980); City of Port Arthur. Texas v. United States, et al.. 517 
F.Supp 987, 1020 (D.D.C. 1981) (three judge court), aff'd. 459 U.S. 
159 (1982). ....... --.. — -  --

Enhancement
27. Candidates for District Court must run for a specific 

Judicial District Court seat. This is equivalent to a numbered 
post system . 31 District Judges must be nominated in the primary by 
a majority of the votes. ’ ’

'“This provision insures that essentially white-voting-- • ~ -----
majorities have a 'second shot' at [minority] 
candidates who have failed to muster a majority of •" 
the votes in the first election. Time and again, in 
election after election, minority candidates win a 
plurality in the first election, only to lose the 
runoff in highly racially polarized voting. " y J 7 u U;

Testimony of Dr. Charles ’Cbtrelif-atT-491, HearingsntBefore the
Subcommittee on Constitutional Rights; of the Committee on~ the
Judiciary, United States Senate (94th Cong. 1st Sess.) S. 407, S.
903, S. 1409, S. 1443 (1975); Plaintiffs' Exhibit Gen-03 at 491.

"A numbered-post system requires a candidate to declare for 
a particular seat on a [Governmental body. The candidate then 
runs only against other candidates who have declared for that 
position. The voters then have one vote for that seat. The system 
prevents the use of bullet, or single shot, voting. Campos. 840 
F •2d at 1242 n. 1 [citing Gingles, 478 U.S. 38-9 nn. 5 & 6 ].

71



Finally, the size of at least five of the nine target counties 
further enhance the problems that minority candidates face when 
they seek office. Plaintiffs' Exhibit Tr-15 shows that Harris, 
Dallas, Tarrant, Bexar and Travis ;Counties have very large 
populations. See also Plaintiffs' Exhibit P-T D-4.

... r Slating •. t ?, • - - .
28. Slating has been defined as the creation of a package 

or slate of candidates, before filing for office, by an 
organization"with-sufficient strength to make the election merely 
a stamp of approval::of the,pre-ordained candidate group. Overton. 
871 F.2d at 534. Dr. Wiser depicted the Republican Party in 
Dallas County as a white-dominant slating group. This.Court finds 
that such characterization is at odds; with the governing law and-; 
facts of this case. Plaintiffs and Plaintiff-Intervenprs did not 
present- evidence of slating in Harris, Tarrant, Bexar, Travis, 
Jefferson, Lubbock, Ector and Midland Counties. 7

7 Racial Appeals , _ r.. r
29. Plaintiff-Intervenor for Dallas County, Joan Winn 

White," argued that racial appeals were injected into her 1980 
judicial race against Charles Ben Howell when an advertisement he

72



ran made reference to his opponent (Ms; White) as the "affirmative 
action appointee." Plaintiffs Exhibit P-I D-30. The Court notes 
and Ms. White testified that the term "affirmative action" is used 
in reference to sex as well as race. The Court finds that there 
is nothing inherently racist about referring to '-an affirmative 1 

action judicial appointment.
30. - Plaintiff-Intervenors from Dallas County also argue 

that racial appeals were inserted into the 1986 election between
Royce West and John Vance and the 1988 Republican Primary between -...
Larry Baraka and Brook Busby. This Court agrees."' In the West -
Vance race, Mr. Vance- 7 made a racial appeal • by inserting his ----
opponent's picture in a campaign advertisement financed by Hr. 
Vance's campaign. In the Baraka — Busby race, Ms. Busby campaigned 
with literature pointing out that her opponent was a Black Muslim, ■'•'toys 
Plaintiffs and Plaintiff-Intervenors did not present evidence of 
racial appeals in the" remaining Counties at - issue *

- Electoral Success
31. Since 1980, seventeen Blacks have run for State 

District Court Judge in Harris County. Only 2 (approx: 12%) won. ' o: 
Plaintiffs' Exhibit H 07. Seven Blacks have opposed Anglos in 
District Court General Elections In Dallas County and won only two -~

73



elections (29%) . Plaintiffs'-Exhibit D-09. However, neither of 
these .candidates was the candidate of choice of the Black 
community. Only one Hispanic candidate of choice won in Bexar 
County in six Hispanic -Anglo elections. Plaintiffs' ExhibitB- 
11. The Black community's preferred choice achieved the District 
Court bench only once ...out ̂ of three elections.. when ^Blacks ran 
against Anglos in Tarrant County. Plaintiffs' Exhibit Ta-07.
Only one Hispanic candidate ever ran against an Anglo .for a 
District Judge seat in Travis County. The Hispanic candidate lost. 
Plaintiffs' Exhibit Tr-11. No minority candidate has run for the 
office, of District Court Judge in Jefferson County. John Paul 
Davis testified at his deposition that the at-large ̂ system 
discourages eligible Black attorneys from running because the 
chance of success is^ so slim. At least! three, Black attorneys 
sought appointment to dthe District Court bench. ( Deposition Summary 
of John Paul Davis ("Davis Depo."). Similar testimony was 
elicited on behalf of Plaintiffs in Lubbock County. Mercado Depo. 
No minority candidate has run for District Court Judge in Lubbock, 
Ector or Midland County.

32. State Defendants argue that the eligible pool of 
minority lawyers, rather-than eligible minority voters, is the 
appropriate reference point for evaluating the extent of electoral

74



success. State Defendants' Exhibit D-04. The Court notes that 
the two cases relied upon by the State involve Title VII issues and 
do not address the relevant statistical pool in a S 2 case." See
Richmond v, J. A. Croson Co.. 109 S.Ct. 706, 725-26 (1989); Wards 
Cove Packing Co. v. Atonio, 109 S.Ct. 2115^ 2121-22 (1989). State 
Defendants recognize that the pool of eligible lawyers is small, 
due in part, to historical discrimination. -The Court finds that 
even if there is some relationship between the low number of 
minority judges and the number of eligible minority lawyers, that 
fact does not explain why well qualified eligible minority lawyers 
lose judicial elections* .* -_: ___ _ __  ... ... 1 **

• ' O'- Responsiveness
■33 4 ! This Court cannot find anything in the record, ; to 

suggest,a lack of responsiveness on the part of?Judges in any of 
the Counties jin question to the particularized..needs of members of 
the minority community.

'• v Tenuousness
34. Several reasons were offered for the maintenance of the 

at-large system. State Defendants and Defendant—Intervenor Wood 
argued that (1 ) judges elected from smaller districts would be more

75



susceptible to undue influence by organized crime; (2 ) changes in 
the current system would result in costly administrative-changes 
for District Clerk's offices; and (3) the system of specialized 
courts in some counties would disenfranchise all voters rights to 
elect judges, with jurisdiction over some matters. Plaintiff- 
Intervenors, HLA, allege that the at large scheme .was . adopted with 
the intention to discrimiante against Black voters in violation of 
the Fourteenth Amendment to -the-United States Constitution.-

.35v Chief Justice of the Texas Supreme -Court, Thomas
Phillips, testified that the purpose of Article 5, Section 7a(i) 
of the Texas Constitution was to create the Judicial Districts 
Board whicht could equalize the dockets ofi District Judges.-, - To 
further that goal, Article 5, Section 7a(i) requires that judges 
be elected from districts no smaller than a county. -Apparently, 
the rationale for such provision is that District Judges should not 
be responsible to voters over an area smaller than an area where 
they have primary jurisdiction. • .r- . j-r : . r

136. Plaintiff-Intervenors offered the Deposition summary 
of Senator Craig Washington in support of their claim that 
discriminatory intent was .. the focus of the legislative 
deliberations surrounding the passage of Section 7a(i). j The Court 
notes that Senator Washington sat on the Conference Committee and

76



signed the Conference Committee Report recommending the adoption 
of the Senate Joint-Resolution containing the exact language of 
Section 7a(i)Tex. S. J.'Res.; 14, 69th Leg. (1985). See Defendant 
Intervenor Wood's Exhibit 59. Subsequently, Senator Washington on 
the Senator floor voted for-the adoption of S.J. Res 14.  ̂Id. The 
Court further notes that three Hispanic Senators votedin favor of 
S.J. 14: Senator Barrientos* Senator Truan and Senator Uribe.

37. Plaintiffs and Plaintiff-Intervenors have the burden 
to establish t.hat the at-large system is maintained on a tenuous 
basis as a pretext for discrimination. Overton. 871 F.2d at 535. 
While the Court does not JLLncLthat_ the present system is maintained 
on a tenuous basis as a pretext for discrimination, the Court"is 
not persuaded that the reasons offered for its continuation are 
compelling.!- v- i.<-. -j A ? . - i -1 ■ n-yo-rt o •

38 ; ^ Under a-single-member scheme or some Mother scheme 
Judges may be made responsible :to jvoters over an area no smaller 
or larger than the area where they have primary jurisdiction. This 
Court finds no reason why all Judges cannot exercise general 
jurisdiction over their geographic area of responsibility. The 
Court further finds that administrative - functions and ■ jury- 
selection could continue to be done on a countywide basis.—';"

39. Our legislative body has seen fit in the past to create

77



in some counties specialty courts. In the mind of this Court this 
is wrong. Judges of civil dockets or judges of criminal dockets 
have equal access to legislation and published opinions. They are 
not intellectually inferior to judges who hear civil, criminal and 
domestic cases. The body of law is large, but is handled capably 
and well by most judges _ia, this State who hear all. types i.of 
litigation. Lawyers specialize. Judges are capable of rendering 
fair, honest and just decisions without concentrating in one narrow 
field of law.’ i .  ;07i-' ijc j.’; ; t- j no o?

r r -  STATE DEFENDANTS ' ANALYSIS
"  , : v r v  • • ; ^  TT:->. ’ \ • -7 G e n e r a l  r; .: f-'f-.i . l:.-’.. 7-r i ,•

40. State Defendants argue that the Supreme Courts 
incorporation in G i n q l e s of the Senate Report accompanying the 
1982 Amendment to § 2, signals a .return to the Supreme Court's pre- 
Ginqles analysis in Whitcomb v. Chavis. 403 U.S 124 (1971). In 
Whitcomb, the Supreme Court rejected a racial vote dilution 
challenge to an at-large system for electing state legislators, 
essentially on the ground that partisan preference best accounted 
for electoral outcomes in Marion County, Indiana. The Court in 
Whitcomb concluded that there was no indication in the record of 
that case that Blacks were being denied access to the political

78



system.. - -----... •
41.. This Court is not convinced that the State Defendants 

are making the correct call. r!’ln any event, the Court ^inds that 
this Court's analysis of the Senate factors applicable to the 
present case point to the continual effects of—  historical 
discrimination hindering the ability of minorities to participate 
in the political process V - '

42. Next, State Defendants are of the opinion that there ;■ 
are really two questions before this Court, depending on what  ̂

electoral stage 'is being analyzed. At the primary stage the 
question is whether the minority candidate of choice in the 
Democratic Primary is prevented more often than not by a" Democratic 
white bloc vote from being the party's nominee in the General 
Election. State Defendants' Post Trial Brief at 9. At the General 1  

Election stage the question becomes whether there ''is a pattern of 
substantial' desertion from the Democratic party by white voters to 
vote for a Republican candidate, thereby denying victory to the 
minority candidate of choice. Id. at 10. This Court-finds such 
a distinction unimportant. Assuming the first two elements of the 
Ginqles test are met and the Senate factors point to vote dilution," 
it is unimportant whether a white bloc vote, which is sufficient - 
absent special circumstances - usually to defeat the minority's

79



preferred candidate, takes place at one election stage, both stages 
or by Democrats or Republicans.

43. The issue of partisan voting was before the Supreme 
Court in Gincles. The Court had no difficulty concluding that 
voting polarized along racial, not partisan, lines. Singles, 478 
U.S. at 61-62. Party affiliation, is^s.imply;. irrelevant under the 
controlling law. Further, "the addition of irrelevant variables 
[to regression or statistical analysis] distorts the equation and 
yields results that are indisputably incorrect under § 2 and the 
Senate Report." Id. at 64.

" " ' ~ ' Statistical ~ ~ * ... .
1 4 4 . : The complete data set used by Dr. Engstrora in Harris 

and Dallas Counties was used by Defendant's expert, Dr. Taebel for 
his analysis Of those Counties.. Dr♦ Taebel' s data,set for analysis, 
in the other_seven counties appears to be very similar. He did 
drop homogeneous precincts from his analysis if; there was more than 
a ten percent (1 0 %) change :in precinct boundary census data since 
the 1980 Census counts. Dr. Taebel analyzed both primary and 
General Elections in not only minority -Anglo contests, but also 
minority Republican candidates opposed to white candidates and 
white - white contests. He also analyzed elections in which the 
minority preferred Candidate ran unopposed.: This Court finds that

80



unopposed election contests and white versus white contests are not 
germane in this Circuit to this Court's analysis. Westweqo 
Citizens For Better Government v. Westweao. 872 F.2d 1201, 1208 n. 
7 (5th Cir. 1989); Campos v. City of Baytown. .840 F.2d 1240, 1245
(5th Cir. 1988), reh'q denied, 849 F.2d 1240. cert, denied. ___
U.S. _____ (1989); Citizens For a Better Gretna v. City of Gretna.
834 F.2d 496, 503 (5th Cir. 1987) . - - • ---•-

. ; . - - CONCLUSIONS OF LAW ....
“1. This Court has jurisdiction pursuant to 28 U.S.C. § 

1331, 18 U.S.C. § 1432_.and 42 .U.S.C. S-1973C. Venue is proper in 
this District pursuant to 28 U.S.C. § 1400(b). 1 -----

2. It is settled in this Circuit that § 2 of the Voting 
Rights Act applies t o t h e  judiciary. Chisom :V. Roemer. 839 F.2d 
1056 (5th Cir.. 1988), cert. denied, sub nom. Chisom v. Edwards. 109
S.Ct. 310 (1989) (Chisom I). However, it isclear that at-large 
judicial elections may not be considered per se violative of § 2 . 
Furthermore, the Court holds that § 2 applies.equally as well to 
State District Judicial elections ns it does to appellate 
elections . 32

“State Defendants argue that State District Judgeships 
cannot_be analogized to legislative or appellate posts, which by 
nature are characterized by collegial decision making. While the 
Court recognizes that State District Judges- function as sole,

81



Standard Under The Voting Rights Act 
3. In Thornburg v. Ginqles. 478 U.S. 30, 106 S.Ct. 2752,

92 L.Ed.2d 25 -(1986), the Supreme Court construed Section 2 of the-4- 
Voting Rights Act, as amended, to require a three-part threshold • 
test to demonstrate a violation of Section 2. The minority group 
must be able to demonstrate that: (1 ) it is sufficiently large and
geographically compact to constitute a majority in a single-member - r-

cz i . ! t ; r  t'.r... • i\ ri i .'.-.zt . ' Tiiz n  v  2 c. n ~ l h

district { "Ginqles 1") ; (2) it is politically cohesive (-"Ginqles - - 
2"); and (3) the white majority votes sufficiently as a bloc to 
enable it -in the absence of special circumstances - usually to
defeat the minority's preferred candidate ("Ginqles 3"). Ginqles.

i - i g i  • ' T • - tt: • ' : • • • >  - .  1 u r  r
478 U.S. at " 50-52. Failure to establish any one of the three-

- ■ ' * '• ' '■* ’ * * ’ 1 ’ ' - • • . I }•'*- v.V _T

threshold criteria is fatal to Plaintiffs' case, r Overton; 871 F.2d
i (- cem . co w h i r r  r.hrr s • ;> . <>r pu i ;_Li

at 538. us n s U . y  large : I ‘ n-a-jor
'■ o r - i rt.-qu r ■■r.v-.nr.s a .r. i •••> m g  :c shot, provis ior.- . j l ner vot
F4. However, Plaintiffs do hot achieve victory by satisfying!-' 

the three Ginqles factors alone. Monroe v. City of Woodville. No. r 
88-4433, slip op. at 5571, (5th Cir. Aug. 30, 1989). Instead,
Plaintiffs must prove under the totality of the circumstances that

independent decision makers, the Court concludes that there is no 
indication that Chisom's extension of § 2 to judicial elections was 
meant to be limited to collegial judicial bodies.

82



as a result33 of the challenged at large system Plaintiffs do not 
have an equal opportunity to participate in the political processes 
and to elect candidates of their choice. Id. at 5571; Ginqles. 478 
U.S. at 44. The Senate Report which accompanied,the 1982 amendment 
to § 2 specifies certain objective factors which typically may be 
relevant to a S 2 claim . 34 S.Rep. No. 97-417 (1982) (hereinafter

33 In White v. Reqester. 412 U.S. 755, 93 S.Ct. 2332, 37
L.Ed.2d (1973), the Supreme Court applied what has come to be„known 
as the "results test" indicating that a violation of § 2 could be 
proved by showing discriminatory effect alone. (Emphasis added.) 
Congress made clear by the 1982 amendment to § 2 that the "results 
test" is the relevant legal standard to be applied by this Court.-

34 Typical factors jLnclu.de-:— ....  - .....- r . r -
"1 . the extent of any history of official

discrimination in-the state ox political subdivision that touched 
the right of the members of the minority group to register, to 
vote, or otherwise to participate in the democratic process;

"2 . the extent to which voting in the elections of the 
state or cpoli-tical subdivision is racially polarized; ; ficc:.

"3. the extent to which the state or political 
subdivision has used unusually large election districts^ majority 
vote requirements, anti-single shot provisions, or other voting 
practices or procedures that may enhance the opportunity for 
discrimination against the minority group;

"4. if there is a .candidate slating process, whether, 
the members of the minority-group have been denied access- to that 
process;

"5. the extent to which members of the minority group 
in the state or political subdivision bear the v effects - of 
discrimination in such areas as education, employment and health, 
which hinder their ability: to participate effectively in the
political process; ..

i : u "f> • whether political campaigns have been characterized
by overt or-subtle racial appeals;

"7. the extent to which members of the minority group 
have been elected to public office in the jurisdiction.

83



S. Rep.). This list of factors is neither comprehensive nor 
exclusive . 35 "There is no requirement that any particular number 
of factors be proved, or that a majority of them point one way or 
the other'. """Singles, supra. at 45 Tempting S.Rep. at 29].

. 5.-’ Singles 1  requires proof that the minority population 
is sufficiently large and geographically compact to constitute a 
majority in a single member district. Gingles. supra, at 50. To 
satisfy-the Gingles 1 requirement, Plaintiffs must be able to draw 
a single member district in which a majority of the voting age 
population is minority. Overton. 871 F.2d at 535. Plaintiffs 
have satisf ied-“this ' requirement with regard to all of the nine_ 
target"counties at issue in this Case. The minority population 
is sufficiently large and geographically compact t o ,constitute a 
majority in at least- one single-member district; Black, Hispanic 
or combined, in each, of the nine counties at issue in this case.

t.‘ ! - * ri * : ' .l I '• »Ti ' 1 ; /- r
i- "Additional factors that in some cases have had probative 

value as part of fP]laintiffs ' evidence to establish a violation 
are: -.ar-.r. - . : r.n. si:-::is v .. i ;; ----- _

"whether there is a significant lack of responsiveness 
on the part of elected officials to the particularized needs of the 
members of the minority group.

"whether the policy underlying the state or political 
subdivision's use of such ... voting practice or procedure is 
tenuous." S.Rep. 417, 97th Cong., 2dSess. 28-29 (1982), reprinted 
in 1982 U.S.Code Cong. & Admin.News 177, 206-207.

S. Rep. No. 97-417 (1982).

84



6 . Evidence of racially polarized voting "is the linchpin 
of a ection 2 vote dilution claim," Citizens For a Better Gretna 
v. City of Gretna, 834 F.2d 496, 499 (5th Cir. 1987) and is 
relevant to establishing two of the three elements set forth in the 
Ginales decision - the minority groups political cohesiveness 
(Gingles 2 ) and the ability of the white majority usually to defeat 
the minority's preferred candidate (Gingles 3). Westwego Citizens 
For Better Government v. Westwego. 872 F.2d 1201, 1207 (5th Cir. 
1989) fciting:Gingles,-supra, at 56]. These factors are usually 
established by statistical evidence of racially polarized voting 
by the voters in the relevant political unit.~ X!ampos v. City of 
Baytown, 840 F.2d 1240, 1243 (5th Cir. 1988), reh'g d̂enied. 849 
F.2d 1240, cert. denied. ____ U.S. ____  (1989). '

7. In analyzing statistical data, the best-available data 
for estimating the voting behavior of ;various groupsJ in the 
electorate would come from exit polls conducted upon a random 
sample of voters surveyed as -they leave the polling place on 
election day, but such evidence was not introduced at trial. See 
Chisom v. Roemer. No. 86-4057, slip op. at 11 (E .D . La. Sep. 13, 
1989) (Chisom II  ̂; Defendant-Intervenor Wood's Ex. 40. The best 
available data for estimating the participation of various groups 
in the electorate is sign-in data contained in the official records

85



of registered voters. ~ Chisom II, slip op. at 12. The best
indicator of participation is obtained by dividing the number of 
persons who signed-in to vote by the number of persons in the 
voting age population. Id. at 12. .......  ... ...  ...

8 . Absent an exit poll, sign-in data and voting age
population data, . analysts employ the bivariate - ecological 
regression technique to estimate the voting behavior of various 
groups in. the electorate . 36 Id. at 12. . . .......

9. -riFor purposes of political cohesiveness and racially 
polarized voting, examining only those elections that had a 
minority-member as a candidate, is the proper method of analysis. 
Campos. 840 F.2d at 1245. In order to.^show cohesion,::t.he "proper 
standard is the same as Ginqles; whether the minority group 
together-votes in a cohesive manner for the r.minority; candidate. 
Id . 37 ,:In counties where-Plaintiffs proceed on behalf of a combined

36 Like the Court in Chisom II, this Court is not convinced 
that precise correlation between the race of voters and their 
voting preferences can be made on the basis of the statistical 
analysis presented. However, no better data is provided, and the 
Court has given the statistical data considerable weight. See 
Chisom~II7 -slip op. at~13.

37f The Court in Campos rejected the City of Baytown's argument 
that in order to show cohesion when there are .two minorities that 
make up the minority group, Plaintiffs must show first that Blacks 
are cohesive, next, that Hispanics are cohesive .and finally, that 
Blacks and Hispanics together are cohesive. Campos, 840 F.2d at 
1245. prefer:“.

86



minority, if the statistical evidence is that Blacks and Hispanics 
together vote for the Black or Hispanic candidate, then cohesion 
is shown. Of course, if one part of the group cannot be expected 
to vote with or does not vote -with the other part, the combination 
is not cohesive.- id. - . ; : l h--

10. --In evaluating the -statistics necessary for Plaintiffs 
to prove racial bloc voting, this Court is bound by recent Fifth 
Circuit^authority to consider statistical' .evidence from judicial 
elections and from exogenous elections. ? 8 o.'. lh».- C • ■.> ■ ce­

ll.. This Court is satisfied that the statistics relating to 
exogenous elections in the present case qualify as a sufficiently 
"local appraisal" to establish some degree of racial bide voting.

12. This Court concludes under the controlling law that the 
statistical' evidence >: furnished by the -expert witnesses j.for 
Plaintiffs3and Plaintiff-Intervenors to be legally competent and 
highly.probative. Ginqles, 478 U.S. at 52-54; Overton. 871 F.2d *

See Chisom II. slip op. at 40; Citizens for a Better 
Gretna, 834 “F.2d at 499. "Exogenous" elections are those which 
overlap the boundaries of the relevant unit. -"Exogenous" elections 
are contrasted"with "indigenous" elections which involve only^the 
geographic unit at issue. Westweqo. -872 F.2d at 1206 n. 10. 
County-wide elections represent the relevant geographic unit in the 
present-case.

87



at 537-540. 39

13. The final determination, h o w e v e r m u s t  be made by an 
evaluation of the "totality of the circumstancesincluding the 
factors listed in the Senate Report. Westweqo. 872 F.2d at 1206. 
The Court must determine, on the basis of a "searching, practical 
evaluation," of past and present reality whether the political 
process is _ open to minority voters. Gingles. 478 U.S. at 45
fquoting S^Rep. at 30, U.S.Code Cong. & Admin.News 1982, p. 208]. 
Such a determination is dependent on the facts of each case and 
requires "an intensely local appraisal of the design and impact of 
the contested electoral: mechanisms.,"—  Gingles . 478 U.S.^-at 79 . 
fquoting Rodgers v. Lodge. 458 U.S. 613, 621, 102 S.Ctv 3272, 73 
L.Ed.2d 1012 (1982)]. The appraisal in this case must be
conducted on a district^-by-district basis. Gingles, supra, at >59 
n. 28 (the inquiry into the existence of vote dilution is district 
specific).- . ■ .- i .•• •. 5 • ..... t - •

1 4 .  This Court recognizes that judicial elections are 
characterized by less voter interest than high profile candidates 
receive at the top of the ticket. However, under the controlling 39

39 . Unlike the statistical analysis in Overton. Plaintiff and 
Plaintiff-Intervenors' experts in the present case established 
confidence levels of statistical significance and used consistent 
measures of minority voting strength. Overton. 871 F.2d at 537- 
540.

88



law, party affiliation, straight party ticket voting and campaign 
factors do not constitute legally competent evidence in the present 
case. This Court rejects the State Defendants' argument that there 
can be no "functional view of the political process" without taking 
into account political party as the principal factor affecting such 
races.; ™The Supreme-Court .in Ginqles made clear that it is the jn 
difference between choices made by blacks and whites alone and,note I id 
the reasons why they vote dif ferently that eis the central inquiry:: a}
of § 2. Ginqles. 478 U.S. 61-62.

15. Congress and the Courts have recognized that "political 
participation by minorities tends to be depressed where minority 
group members suffer effects of ■ prior discrimination such as • ê r. 
inferior education, poor employment opportunities, and 1 o«l „ -K7d 
incomes. ud Ginqles. .478 U.S. at 69-dCongress clearly concluded 17 *r- \ 

that provisions such as majority vote requirements, designated 
posts, and prohibitions against bullet i: voting could serve uto*r ion* 
further dilute the voting strength of minorities. Id. at 56; Jones 
v. City of Lubbock. 727 F.2d at 383 (finding that majority vote 
requirement further submerges political minorities) . .'.y p . - i s c c

16 . : nThis Court concludes that under^ the relevant law the 
at-large system for election State District Judges- in the nine Li- 
target counties interacts with social and historical conditions to b-v-f

89



cause an inequality in the-opportunity enjoyed by black and white 
voters to elect their preferred candidates. Ginqles.478 U.S. at 
47.

17. Defendants' lead expert, Dr. Taebel reviewed many 
election contests which the. - Fifth--Circuit determined -are not 
germane to Voter Dilution.C a s e s D r .  dPaebel analyzed races in 
which Anglos opposed Anglos. Campos v. City of Baytown. 840 F.2d 
1240, 1245 (5th Cir. 1988). Dr. Taebel also reviewed non-judicial-, 
elections. v . ' •••; . ■••0 "

IB. Costly reorganization of the State at-large system of 
general and specialized Courts and disruption of County 
administrative duties such as jury selection are not— sufficient 
grounds for maintaining an otherwise flawed system. Westweqo, 872 
F.2d at 1211 Tin reliance on Dillard.v. Crenshaw County. 831 F..2d 
246, 250-51 (11th Cir. 1986)]. , :.vl b o d y  a!.: a - .

19. Congress did not contemplate that such considerations.'
would play a role in determining whether there has been a violation 
of section 2r Id. at 1210-11. ...

20. On the strength of the evidence of racially polarized 
voting in the context of. the.-i "totality of-the circumstances” test 
and considering the substantial evidence presented by Defendants 
to the contrary, this Court concludes that Plaintiffs -have

90



demonstrated a violation of § 2 of the Voting Rights Act in each 
of the nine counties in question. W e s t w e g o 872 F.2d at 1203 & 
1209. - ------

' "fourteenth and Fifteenth Amendment Claims .
21. — Proof of racially discriminatory-intent :©r purpose is

required to show a violation under either the Fourteenth or 
Fifteenth ̂ Amendment to the United States Constitution i Ghisom II, r 
supra, at 41 [citing: Kirksey v. City of Jackson, Miss. . 663 F. 2d
659 (5th Cir. 1981); Washington v. Davis. 426 U.S. 229, 239-41 
1976) ] .

22. Proving racial'discrimination as the motivating factor
in a state legislative body "is often a problematic undertaking." 
Hunter- v. Underwood; r.471cU.S..-. 222^ 227-2& (1985).. aProof must be 
presented that the legislative body as a whole possessed the intent 
to discriminate. Id. at 229-32. I t cis impossible to conceive
that four leading minority members of the State-Senate would vote 
to send an individiously discriminatory measure affecting the 
entire state to the voters with their own seal of approval on it.

23. Plaintiffs and Plaintiff-Intervenors failed to prove, 
as a matter of law,that the present at large system for electing 
State District -Judges in the State of Texas was instituted with the -

91



specific intent to dilute, minimize or cancel the voting strength ~ 
of Black and/or Hispanic voters.- Accordingly, the Courtds of~the 
opinion that the following Orders are appropriate: - ■ - - -..-

IT IS ORDERED that the present at-large system of electing 
State District Judges in the counties of Harris, Dallas, Tarrant, 
Bexar, Travis, Jefferson, Lubbock, Ector and -Midland ^violates L 
Plaintiffs' civil rights by unconstitionally diluting the voting 
strength of Hispanic and Black electors in violation of Section 2 
of the Voting Rights Act of 1965, as amended, 42 U.S.C. S 1973 
(West Supp. 1989).' • .

IT ISn FURTHER ORDERED that Plaintiffs 'and Plaintiff-
Intervenors request to Permanently _Enjoin the State of Texasdrom

~ -.gnest L = .• - '— -■-** thee: l o g y ,
calling, holding, supervising or-certifying anyifuture -elections

-ir ti^rr c*± ( tno dictinguishsd e.*«jwrts --
for State District-judges under dhe presentr-at large scheme inTthe

v ~ ip  i t  a i e u u i u  u i  a iG a o r s p n ie n ts  .
target areas is taken under ̂ advisment. QThe Court recognizesithe

r . • . -j - ' c ' t ! r-.  ̂ ’ . ’ * - ^ ___4- - y ~ — • - — i-'i ' -* ■•■■'•'i'v.-. wli . »Vv_ \_iO
possibility that corrective relief may be-available at ar dated date 
before future elections for State District Judges take place. 
Chisom v. Roemer. 853 F.2d 1186, 1189 (5th Cir. 1988).

The Court is hopeful that Governor Clements will include the
issue of an alternative State District Court election scheme as 
part of his call of the Special Legislative Session on November 13,

92



1989. Depending on the progress that is made in the Legislature, 
if any, prior to January 3, 1990, the Court will thereafter 
entertain a Restraining Order or Motion to Enjoin future State 
District Court elections pending the Remedy Phase of this 
litigation.

IT IS FURTHER ORDERED that the issues of Costs of Court and 
attorneys fees are expressly reserved until the conclusion of this 
litigation.

Chief Judge Charles Evans Hughes, in 1936, in an address to
the American Law Institute, said:

How amazing it is that, in the midst of controversies 
on every conceivable subject, one should expect 
unanimity of opinion upon difficult legal questions!
In -the — highest ranges of thought, in theology, 
philosophy and science, we find differences of view 
on the part of the most distinguished experts —  
theologians, philosophers, and scientists. The 
history of scholarship is a record of disagreements.
And when we deal with questions dealing with 
principles of law and their application, we do not 
suddenly rise into a stratosphere of icy certainty.

This area of the law is not a sphere of icy certainty. 
Should the Legislature fail to adopt a satisfactory Remedy in the 
Special Session (provided Governor Clements includes this matter 
in his call) this Court will consider the granting of an expedited 
appeal to the Fifth Circuit to determine whether or not the

93



Declaratory Judgment of this Courtaas properly made.
o KSIGNED AND ENTERED this day of November, 1989

jucius D. Bunton 
Judge

94



APPENDIX "A"
Plaintiffs' & Plaintiff-Intervenors' 

Statistical Analysis



■ J ;; i m  t  r ,RACIAL DIFFERENCES IN CANDIDATE PREFERENCES IN DISTRICT JUDGE 
ELECTIONS IN HARRIS COUNTY. TEXAS 

.a GENERAL ELECTIONS. 1980-1988
Prepared by Richard L. Engstrom, Ph.D

HOHOGENEOUS PRECINCTS ------ ---- 3IVARIATE REGRESSION- WITH CONTROL FOR HISPANICS
Year Black

Candidate
(Party)

X of Non-Black 
Votes

X of Black 
Votes Correlation

Coefficient
X of Non-Black 

Votes
X of

Black Votes
Partial

Correlation
X of Black 

Votes

1BB0 Bonner
(Daaocratlc)

38.6 96.9 0.822 37.8 103 0.909 98.4

1982 Janes
(Democratic)

35.6 97.5 0.799 34.8 104.2 0.B95 99.3

190 2 Routt (W) 
(Democratic)

38.5 98.1 0.798 37.7 104.6 0.896 99.9

1982 Ward
(Democratic)

34.7 97.7 0.801
)•

33.8 104.2 0.895 99.3

1984 Berry
(Democratic)

34 97.3 0.883 32.8 102.7 0.922 100

1984 Jackson
(Democratic)

30.6 97.7 0.880 29.3 103.5 0.942 100.5

19B4
(Democratic)

36.4 97.8 0.881 35.2 103.2 0.938 100.5

1986 Berry
(Democratic)

35.3 97.7 0.851

0.847

34.2 103 0.916 99.4

1986 Plummer
(Democratic)

37 97.9 36 103.1 0.912 99.6

1986 Proctor
(Republican)

52.8 4.6 -0.836 53.6 1.1 -0.899 3.9

1986 Walker (W) 
(Democratic)

40 98.2 0.847 39.1 103.3 0.882 99.9

1988 Berry
(Democratic)

32.8 97.3 0.860 31.1 103.1 0.897 99.4

1988 Fitch
(Daaocratlc)

36.9 97.7 0.849 35.4 103.4 0.836 99.7

1988 Jackson
(Democratic)

33.4 98 0.856 31.8 103.9 0.928 100.1

f PUIMTJFPS^
KWHWDWIHlT

X



1 1 1 . I i , i . ' F . l ’ i ! . r !  I L V  I 3  i  t i j ' a * ! / '
’■ V.. : I am I9f1! : !

19B6 L e a
( D t a o c r a t l c )

39.2 98.2 0.649 37.7 103.8 0.92 100.3
198ft P l u a a t r  

( D e m o c r a t i c )
34.3
1 i 1

97.4
J (1 <4

0.850
r >  1

32.6
; 1 , t 1 l . l —  , : ■)

103.9
; i ; : j i

0.924
i i 1 -ini*

100.1
12 6ft S p e n c e r

( D a a o c r a t l c )
3f t.3

9 8  ,  ■■ 0 . 8,421 , - I . ; > * . « ,  . 103.7 0.917 100.1
( 7!» i;.. .«' • n •. !■)

•V i1 1 * ' r  < l'. ‘ . j 1 r.i 1 ui 1 '..it .11 1 I I

V i*-*
i

i i :i;. .• i.\ ,'1<. ' * ( * ( . :» ,.t . , i . i  ;

-i ,i ' h ' •«. v

i y  .. i r — 'm . . 

«. • . , m r  c i  v.. l 1 i..

11

.I 1 !.’ 1 j i

»■ !!'. ..1 «) 1 .4 1 t* •;V .'a ifif . 

U, • .4 .j ..4 v.f* <:i Cl 14 .«• h o

\1 ,! 1 1 • i li,1..4» .» . j  11< i  r  i .  \ C i« . r ; i l .  i

. a; i . i ;

V |1 ’’ 1'

• k1 > ■ * .»i • ■ i ur* 1. . »i . 

it t i i ..4 «i .' i i * . r  c i J  « c ' i r  ii. c.

11 i i

U .i * t S iv .. ‘ " t - . i i  r  . : 1 •- r

1 !. j 1.: i ' • . r  « u :•.

. i i,

< . i ’ ; * : | .* r .i;-l. . o  - - 'u  . ■ ••

,n f • i i i ' i ) >* u

( r r.

J f

llli In - J » • ’ 

r.j ,

* ; l* j*  r  1 :r f r j i i  l

i ;: i ' 4 1 i ;< .,

|.:i '

• V 1 j f. !i . . . . . i . a l l  :: C* i.j . V. Lr • ’•

.1 A \ : -.(• . M r . t  » j ' ! i* ' l. ' i

:* * •' Vv« f 1 , :» l .

1 * j.*4 i4 irk  ;  • i [ _i: i

; , , ' «!•. ■) i . •: v>. .. ■ . . i ‘i i

iu | niiii ,:j i : !|

V -

'*«

V-

I

I

V .

2 -



I I I . I l l

BLACK JUDICIAL CANDIDATES IN HARRIS COUNTY, TEXAS 
YEARS I960 - 19B8 « ]

. 'ii < t Li ■ . 5 'll1' v

'  1 I a  *  i r

Y w  Election
1980 Primary Judela1 Election

1980 General Judicial Election

1982 Prleery Judicial Election John Ja

Name
t i: • ,. 1 i

Alice Bonner
. II

Janes Mill drew
: l - . i

Won
.1,: 11 1 Fred Reynolds

ill : 
.■ 1

Lost
. ; . IT t

Alice Bonner Lost
‘I.i; tJames Muldrow 1 '■ : Lost

1982 General Judicial Elect 1m

. 1  1  ' ■  • <iJanes Mu Id row
John Peavey
Thcanaa Rou t t
Clark Gable Viand
John James■i. i i;
James Mildrow |

Outcome Section!; :..i ■ ■
Unopposed Both State District Court (civil)

fo'v | i, l i l . i ' j  . V  f . l  I D i l l  j l  ( i  ' H i " ,  .i .
County Criminal Court No. 6

hji i  ,1 ;  ! - Im l i  i i  C, a . r  I I
County Criminal Court No. 10

I . ii , i: , ; i . i l l '  ‘ i . .  ' i l l  a :, l r . i '  J ■ I
Both State District Court

ill-. I| I C a. i . I . ’ -. . r > *'• il*
80th County Criminal Court No.6

|! . ' . I *!■Unopposed 262nd State District Court (criminal court)
U 4; I a t . i  .in "I i . i ' /  Cia.-rUnopposed County Criminal Court No. 6

•••. < . i ! I , .  i ■ v  r : l ..Unopposed 246th State District Court (family I s m )

Unopposed 208th State District Court
.1 . I . > .1,. r . ' I:t . l IUnopposed 281st 8tate District Court (civil court)

,, U»t 262nd Rtate pis^rlpt,(Jpurt (criminal court)
D ôst County, Criminal Court, No. 6

Note
an Incumbent

had been appointed; ran >m lranmbent

John Pqau/y ; | ] GfJnoppgeed 246th Jtat<t QV*Xfilclt, Q»Wt (family law)
11 Thomas Routt j : Mem 208th Rtate Dlslrlpt, Court (criminal court)

I Clark Gable Ward tost 281st State District Court (civil court)
1984 Primary Judicial Election Weldon berry 1 IJWioppasad 80th State District Court (civil) Incumbent

i 1 ,, Carolyn D. Hobson ’made runoff i i.r n. • -d , . i ■County Civil Court at Law # 3 . I;

■ Freddie Jackson ' made runoff 178th Stats District Court (criminal) • . 1'

Shel la J. 'toe vibn 215th State District Court (civil)
Kenneth Levi Lost 333rd Sfcgte piatplgt, Cwpt (plvll)
Jim Muldruw Lost 351th Stats District Court (crlalnal)

1984 Runoff Election ■ i ,iCarolyn D. Ilobsop Wirt [, ..... i •, ': • .j u» i. ... -.ii 1 'iCounty Civil Court # 3
».i,. 1 r tFreddie Jackson Won r: va .178th Stats District Court (criminal)

1984 General Election Walden Berry Lost 60th $tata District Court (civil)
'i

Incumbent

- 1 - i 1

had been appointed; ran as Incumbent

) ':r



I ; b n: ii l. *3 (

1 in (»)
Carolyn 0. Hobaon'■> Loat 1
Freddie Jackman ‘ Lout 1
Sballa< J. >Laa •' • (0, Loat

1M6 Primary Election Barry Malden (0) D, Unoppnaad
< Freddie Jackaon (D) Loat

Cheryl B. Irwin (R) Uhoppoeed
Raymond Flatter (D) Men
Bomla Fitch (D) 1 1 Uboppoaad
Hobaon, Carolyn (D) lluppaad
3 H la  l a  "L . J a c k a o n  t i l l Mon
John Paavy (D) Utttppoaad

i i 1! Matthew Pliramr, Sr. (D) Uhoppoeed
Mamie Proctor (R) 
Reynold*, Fred (D)'

UhqppQ aad

Themae H. Routt (D)
■t ' ! |.‘ I . ?' 1

Carl Melkar, Jr. (D)
Unoppoaad

Franc la Mllllaaa (D) Unoppoead
IMS R u n -o ff Frad Raynolda (D) loat

Carl Mtllcar, Jr. (D) Moq
1988 Judicial Flection* Carl Mallear, Jr. (D) Mon

Carolyn D. Hobaon (D) Mon
Hatthaw Plimner Sr. (D) Loat

Bomla Fitch (D) Loat

Rayaund Flahar (D) Loat
Malden Barry (D) Loat

■ v .1! I V i I l Ml . i , ■ 1 . 1 . 1

County Civil 0ourt'>al Law » 3 
178th Stata Dlatrlct'Court (criminal) 
215th Stata Dlatrlct Court (civil)
281 Civil Dlatrlct; ‘ M

bx i J.'....111 ;•

: u. .

295 Civil Dlatrlct ‘ " 
County Criminal Court * 3 ''

Loat to Frank 0. H.lte

County Crlmllial Court * 14 won ervur lllu(unlc, Angel Fraga
County Criminal'Court M 13 1 
County Civil Court # 3 ' 
Probata #-4- '• '■ '• ;

Incumbent

Stata Family Court ‘246 1,111 Incumbent
133 Stata Civil '* 111 " 1 Incumbent 11 ljl ■ * 1 " '
Stata Family * 245 1 
County Crlmliml'court No. 11

1 . ; ■ "1 m| i 1 ; ’

: , i i* . ■ J •. 1 i.. |

made It' to runoff, era of five candldatua 
running for County Crlm. Ct. No. 11.

Stata Criminal Court 208 Incumbent
Stata Criminal 85 nude It to run-off
County Criminal Court # 4 litcumlant
County Crlmlnlal Court No. 11 Loat to David Mendoza In run-off; 

Mendoza eventually won.
Stata Crlmlnlal 85 Won agalnut Sallnaa
Stata Criminal Court 185 Mm over George Godwin
County Civil Court 3 Mon ovef) /U len ltugl.ee
133 Stata Civil Incumbent

County Criminal Court 13
loat to Lamar McCorkle 
Incumbent
Loat to Mark Atklnuon

County Criminal Court 14 Loat to Jim Barkley
, 1 281 Stata Civil Loat to Uiula Moore
1

2--



Francis Williams (D)

S « 1 U  J. pea (DĴ  

Mamie Proctor (R) 
Chtryl K. Irvin 
Thomas H. Routt, (D) 
John W. Peevy (D)

lSMPrlamry Klmctlon Ban Durant
Bonnla Pitch 
Raj acnd Flahar 
Hut thaw W. Pli— tur 
Malden Barry

\ Beverly Spencer
Freddie Jackaon

1M8 General Klactlan Bonnla Fitch
Shall* J. Lm  
Matthaw W. Iliaaar 
Malden Barry 
Beverly Spencer
Fraddla Jackaon

. : ,  t < „ |

Lost County Criminal Court 4
!;• i.v :,w>r i J .c .-!• :,t ,j

Last
|
i; Probata # 4, isj> '• 1:1 li in • li i!t:».

Lost- \ ■
State Family Court 245

Lost County Criminal Court 3
Unopposed State Criminal Court 208!■' '■<!
Ur*4JfXM»ad State Family Court 246
Lost 174th Civil District Court
Unopposed 132nd Civil District Court
lost 177th Criminal District Court
Unopposed 133rd Cli/11 District Court
Unopposed 80th Civil District Court
Won 1 333rd Clyll District Court
Won 213 Civil District Court
Lost 152ml Civil District Court
Lost 295th Civil District Court
tost 133rd Civil District Court
tost 80th Civil District Court
Last 333rd Civil District Court
Lost 213 Civil District Court

Incumbent
Lout to J<fn»̂ a E. Amiarsen

lout to Bill McCulloch
1 ; r! iLout to Usury Schubla* , . M t t M ( j

Loot to Jimmie Duncan

incumbent

lncumljHnl

run against Greg Glass 

ran against Miron Lova

ran against Jack O'nail 

ran against Dan Desnay 

ran against Lamar McCorkle 

ran gainst William R. Powsll 

ran against Davie Wilson 

ran agalp^t Gena Chambers



x._; ;.g
;• ", c -

Black 
Candidate 

Vear (Party)

Homogeneous Precincts

X of Mon-Black X of Black 
Votes Votes

: d  ^I J £ £  
O

1980 Winn
(Dem)

39.7 i 98.1

1984 Baraka
(Rep)

60.6 l 0 3.5

1984 Tinsley
(Dem)

30.0 97,4
‘i 0

1984 White
(Dem)

31.9 97.5
i ■■ i)

1986 Tinsley
(Dem)

37.5 98.3;

1986 Wright
(Rep)

70.6 4.3

1988
==>,

Oliver
(Dem)

37.9 98.3

C > 5 I J ; V ' l T . x  ; C C ■,'! : i |
A V ! jij;! 4iip»:n ?D|JS , 'm ;:i ,c I ; 1

DALLAS COUNTY /oi ‘ * 1 .
i‘ J t -liJ V..L _xJ:l:±!:!;!lG.:!.. .

AWM.YSIS OF JLOICIAL ELECTIONS

Bi-variate Regression With Control for Hisponics
Correlation
Coefficient

X of Kon-Bleck 
Votes

X of Black 
Votes

Partial
Correlation

X of Black 
Votes

.865 38.6 100.5 : :-i J .912 97.2

-.894 61.8 -0.5 -.932 2.8

.902
J

28.7
(i >

103.2 .943 99.2

.902 30.6 103.1 .944 99.1

i .677 36.6 i 104.6 .923 100.6

-.872
i i . ’
71.7

1 i..1 i, i
-1.5 -.916 2.8

I
.664 36.9 104.3 .913 100.2



r

i ; i ?in /• . i <. . . . . . . . . .  ■

i
1 )0 :>■ \'-i ,!<’ 

•: u;: 131,1 i?(2
A i : o '

ESTIMATES OF EiHNIC GROUP VQTNG IN BEXAR COUNTY DISTRICT COURT ELECTIONS: 1982-1988

« . M. 1.

ft*
'■tiM;
ii n r  | O  r:

* *1

Bivariate 
Pearson r 

Sin.

Regression Analysis 
Estimates for: 

HispanicslNon-Hisoanics

Homogenec 
90-1 00%
Hispanic

us Precinct Est. 
9 0 -1 0 0 % 

Non-Hispanic

Are ethnic 
groups 

polarized?

Does 
Hispanic 

choice win?1982 General E lection 7 1:: 4 'J
D i s t r i c t  C o u r t  # 1 4 4

0CO1 . J •’ 1 t'j
YES KD

Barrera (Hispanic) . 0000 1 7 77 24 74
; Slohlhandski 83 23 76 26

Total :i 100 100 100 1 00 ,
D i s t r i c t  C o u r t  it2 9 0 .87 YES hD

Delgado (Hispanic) .0000 103 1 8 92 2 1
Berchelman - 3 82 8 79
Total 100 1 0 0  i 1 00 " 100

1984 General E lection
D i s t r i c t  C o u r t  i t3 7 .87 I c !

YES ND
Davila (Hispanic) . 0000 1 04 26 73 35

Cornyn - 4 74 27 65
Total 100 100 1 00 1 00

1986. General E lection I
D i s t r i c t  C o u r t  i t2 8 5 .88 YES ND

Cisneros (Hispanic) .0000 95 1 2 88 22
Peeples 5 88 1 2 78

Total 100 100 100 100
1983 General E lection
D i s t r i c t  C o u r t  # 7 3 .87 YES YES

Mireles (Hispanic) .0000 106 35 93 37
Bowles - 6 65 7 63

Total 100 100 j 1 00 1 00
D i s t r i c t  C o u r t  # 2 2 5 .86 I. YES ND

Serrata (Hispanic) .0000 1 03 28 9 1 33
Specia - 3 72 9 67

i otal 100 1 00 100 100

i



V V-

E S TIM A TE S  O F E TH N IC  (3R O U P Y O T I NG IN TA RRANT C O U N 'fY E L E C T IO N S  1 9 8 6 - 1988
: J Partial r Repress ion Estimates Homogene ous Estimates Are ethnic groups polwized? Does Black

i Black Anglo Black Anglo B/A choice win?
1986 G en era l E le c tio n
Critn. f irs t. C rt. F f  4 .87
Selvent (Black) 7 46 6 41 NO NO
Drago 93 54 94 59

TotaJ 100 100 • 100 100
Crisi. f irs t. C rt. P i  f -.80
Stums (Black) 15 49 11 44 NO NO
Goldsmith 85 51 89 56

Total 100 100 100 100
1988 Dec* P rim ary (
P res id en t .93 < ~J> 1 1 .) Is "i1 VJ YES YES
Jackson (Black) 99 14 93 16
Gore+Simon+LaRouche+Hart+Dukakis 1 86 7 84

TotaJ 100 100 100 100
1988 G en era l E lec tio n < ro \

C rrn. f irs t. C rt. F i  2  i .90 - T*" r r j , \ YES NO
Davis (Black) 100 42 98 50
Dsuphinot • 0 58 2 50 '

TotaJ 100 100 100 100



ESTIMATES OF ETHNIC GROUP VOTING IN TARRANT CO. ELECTIONS 1982-1988

Bivariate Regression Analysis Are ethnic Does ‘
Pearson r Estlmai:es for: groups polarized? Black

Slq. Blacks Whites B/A choice win?
1982 Democratic Primary 
Co. C rim in a l Crt. P i i .82 YES NO
Hicks (Black) .0000 87 38
Coffee 13 62

Total 100 100 -
1986 Democratic Primary 
Co. C rim in a l Crt. P I 1 .76 YES NO
Ross .0000 57 1 1 -
Golflfeather+Ross+Pounds+Clark 43 89

Total
1986 General E lection 
Crim. D ist. Crt. P I 4 -6 3

100 100

. . .

salvant (B lack)- R .0000 3 ! 55 YES YES
Drago 97 45

Total — 100 100 ------ • .... ••• - •
Crim. O lst. Crt. P ! 1 -.6 0 " r

Sturns (Black) -  R .0000 9 57 YES NO
Goldsmith -  D 91 43

Total
1988 General E lection 
C rim in a l D ist. Crt. P I 2 .62

100 100

YES NO
C. Davis (Black) -  D .0000 103 40
Dauphlnot -  R -3 60

Total 100 100

Source: Numbers ore from  2:12 lyse s Conducted by Delbert Taebel, Deportment o f Urban Studies, 
Univ. o f Texas at Arlington

1 .1; i I 1.
* \ 1 it —

• I 'IvV
r

vL1______



r  \ \

: i; ' ; . .i j • i1 ■ ■ p’Mii :j; 'i 'c ••
ESTNV.ATES C*E ETriri c  g : ;o u ? v o iT s G  t r a y  i s  c o u : ; t /  e l e c t i o : :s :  ic e s

P&j^i-Jf
;

fv'tlliple Repress 
Hispwiics 1

on Est. 
Anglo

Homogeneous 
His ponies |

Precincts 
Anglo ,

Are ethnic groups 
polarized?

Does Hispanic 
choice win?

1SC0 D e n  P r .w y ^
P /s tr fc t  Court' F J < J .* C ..\l s/po NO
G ollerdo (Hispsnicj .0000 03 3-1 00 r-"?•-» i
fvfcCown / -1 >. 7i 66 14 03

lOtcd 100  1 100 100 10 0  '•
C o u n ty  C ourt-s.(-l.o t? , .4.OH I : wee1 LJ NO

Gsrci? (Hisps-nio) .0000 05 33 00 J •' f ! ' ! 1
fh.llipS ;■ r*.i 07 10 63

ToteJ 100 10 0 100 100
C o u n ty  Cou.’T - o t - t o u ■ .00 Y ES NO

C:-stro (Hispc-nicj .0000 { f M 03 10  '
Kennedy (EtkidtJ+Hughes 23 v6 ',-jl HIv* 1 , i

ToieJ 100 10 0 10 0 100

fUUO ItTMAî j)
“TD

^ £  >  n
' 55ij
SI “< --2 .’’i
1 "I
■^r- J



I

ic. ; { *. i
E S T I M A T E S  O F  E T H N I C  G R O U P  V O T N G  I N  T R A Y I S  C O U N

B i v a r i a t e  R e g r e s s i o n  A n a l y s i s
P e a r  S o n  r  

S i d .
E s t i m a t e s ;  f o r :

Y  D I S T R I C T  C O U R T  E L E C T IO N S - 108  8
Homogeneous Precinct Est.
9 0 -"100 % 90- I 00S

1988  D e m  P r i m a r y
D is tr ic t  Court * 3 4 5 .36

-------- ------------- 1— >— —n r
• ! 1 •: j

G a l l a r d o  ( H i s p a n i c ) .0000 101 36 r i  86 3 7
M c C o v n  i - 1  i 64 14 63T o t a l
County C o u r t -o t -L a v ,8 b

100 1 00 1 00 1 00

G a r c i a  ( H i s p a n i c ) .0000 100 36 90 ■ j
P h i l l i p s 0 64 10 63T o t a l
County C o u r t -o t -L e v * 7

100
!

100 1 00 1 00

C a s t r o  ( H i s p a n i c ) .99 76 1 1 4 , :;63 1 6
K e n n e d y  ( B l a c k )  +  H u g h e s .0000 24 86

J  i
37 84T o t a l  r

• 1 

• J

100

i

100 100

■ i i j i ;

1 00:

Are ethnic 
groups 

polarizer)?

YES

YES

YES

N

Does
Hispanic 

choice v i  n?

N O

NO

NO

1 t 
! i

r i

•j •
Urn

j :  i  E X H I B I T
■ f .



PLA
IN

TIFF'S
EX

H
IB

IT
JT-

osl

Jj
r. /

c: i .orj :
• •» -I. : . 4 /. i.. i'i* : ' .n,iu> ,t'j I- »u:i ;JM,cn;..| s| M • .

E S llh tA T E S  OF ETHNIC G n O U P  VO TIN G  IK JE F F E R S O N  COU.TTY E L E C V lb r:^  197 2 -19 80I Portiof r f E t r . . _ _ II..------—=  ------—
S

1972 D em o cra tic  Prim ary  
J. * . F ., Pet. 1r F/. 2  
Freeman (Black) 
MttrelkTra/rier.+Leibold+PaJmus 

Total
1972 D em o cra tic  Runoff 
J. * . F., Pet. / ,  FA 2  
Freeman (Black)
Trwnen

Tola!
1974 D em o cra tic  Prim ary  
-A. c. P ., Pet. 2, FI. 2  
Freeman (Black)
Kwr+Knowles

Total
, 1974 D em o cra tic  Runoff 
2. o. P., Pet. 2, FA 2  
Freeman (Black)
Knowles

Total
1978 D em o cra tic  Prim ary  
Ceunly Coart t  Cmtr, 2 /2

Davis' (Black)
St3es'-+fv1anes

Total
1982 D em o c ra tic  Prim ary  
•A *>. P., Pet. /, FA 2  
Cannon (Black)
McCassell+McCall

Total
1986 D em o c ra tic  Prim ary  
•A. v. F., Pet. I, ri. 2
Roberts (Black)
Robinson-tMcGinriis+Davis+Mller 

-ic^v Total
.» j P1988 D em o cra tic  Prim ary  

FresiJent 
jJackson (Black)
'-'Sore+Simon+laRc-uche+Hart-tOuki 

Total

^a/tial r ; ! Regression Estimates i li 
Black Anglo

Homogeneous Estime 
Black Anolo

.70
70

i !i
25 75 26

30 75 25 74
100 100 f 100 100

.66
85

i’.
: f' 

38 92 40
15 62 8 60

100 100 100 100

.75
83 26 89 25
17 74 11 75

100 100 100 100

.72
93 41 95 42
7 59 5 58

100 100 100 100

.97
10 93 13

16 90 7 87
100 100 100 100

.97
53 7 51 6
47 93 49 94
100 100 100 100

.93
47 2 40 3
53 98 60 97
100 100 100 100

.97
101 6 96 7

is -1 94 4 93
100 100 100 100

Are ethnic groups polanred? 
B/A

Does Black 
choice v/in?

ack I 
vin? |

YES

YES

YES

YES

YES

YES

NO

YES

NO

NO

NO

NO

NO

NO

NO

YES

-1-



TO
-L

i.' i ;  .0 , i 11! . I.J..

f *;'f,'• ,t: Ur'itJ:
J

i . -’i*:; ' l l : IWt ;
us E >iui<l i’£:.r i j  ! i -uu '•

ESTIMATES OF; ETHNIC GROUP' VOTING IN JEFFERSON 'COUNTY ^ L ^ E C T ld ^  'lsVz-lSR n 1,1 •; p.il i

'i960 Democratic Primary 
P res id en t \ .97
Jackson (Black) ,5 .0000 101
Gore+Simon+LaRouche+Hart+Dukakis1: -1

Total i i 100

Multiple Regression Analysis Homogeneous Precincts Are ethnic Does
Partial r Estimates for: 90-100% 90-100% 4 ( groups polarized? Black

Sin. Black | Anglo Black Anglo , B/A choice win?

6
94

100

I I II
YES YES

if96 5 4 
100

j

7
: 93
H 00 ; ;

.*vt € p. ■»i fin it, v?ir re ■ 111

i

i sr }i! II

iI



1
H;i 1 :!‘l* • i ■’ '•

; for 
Comb.Mr^*

99 
1

100

90
10

100
>0 (

35
65
100

Gonzalez (Hispanic) 
Howell+Scholz (White) 

ToteJ

94
6

100

90
10
100

37
63
100

Homogeneous Estimates 
Anglo C om b.M n*

E S T IM A T E S  OF ETH N IC  G R O U P  YO TH G  IN LU BB O C K  COUNTY E L E C T IO N S : 1986 -198S  
IF ertiiJ rl Regression E^tim*j|fcs
|_____ Hispanics Black Anglo

1986 G e n e ra l E lec tio n  \  ~  ^  ^*r*^
Supreme Ct. PL 4  .95/79  
Gonzalez (Hispanic)
Bates (White)

Total
1988 G e n e ra l E lec tio n  
Supreme Ct. PL S  .93/ 88

■I"tin’; !»!;.< I
c i

39'
61
100

40
60
100

1 0
91
9

100

1 & u 89
11
100

Are ethnic groups polarized
Comb. Mn/Anglo

VES

YES

Numbers in these columns were derived from bivariate analyses, all others from multivariate analyses.

Are Hisp. 8. Black Does Comb. Mn.
cohesive choice win?

YES

YES

NO

NO

i

i



)Ei; i K v.ii-Y eLi.c].i«<i'iu ij: :■V' -t.;isi!i I Hon t ?■:.'j : p.
' )

I
ESTIMATES OF ETHNIC GROUP VQTNG IN 

Bivariate Regression Analysis 
Estimates for 

Anglo |Comb Min

LUBBOCK COUNTY GENERAL ELECTIONS: ' 1986 -1908

I .1 1; Hi

Pearson r
Sig.

Election1906 General 
Supreme Cl. PI. 4 
Gonzalez (Hispanic)
Bates

Total
1988 General Election 
Supreme Cl. PI. 3 
Gonzalez (Hispanic) 
Howell+Scholz 

total

.96
.0000

.94
.0000

Multiple Regression Analysis 
Partial r- H/8 Estimates for 
) ?ig.-H/B Anglo| Hispgnic | Black

rJ
.9S/.79 | i

II T-.) M

35 97 .oopo/.pooo 35 99 90 39 91
65 3 JUIH. 65 ! t 10 61 9

100 100 100 1Q0 100 100 100

,93/.88
I 1 Ov.V

■ i i OH

37 93 .0000 /0000 37 94 90 40 89
1163 7 . :)0C; 1 : 31 63 § 10 00

100 100 }00 .100 100 1Q0 too

x- j- * ‘I 7
I A\i. 0 1 ') ,

i : ■oc t:. : '.;.j i, -ilj:
, t.

s'

Homogeneous Precincts 
90-100% I 80-100%  

Anglo | Comb. Min

Are ethnic 
groups polarized? 
Comb. Min/Anglo

,l ,

Are Hisp. 
& Black 

cohesive?

Does
Comb, Min. 
choice win?

YES

YES

YES

YES

NO

NO

i .



' i i s
! ,l,v

I . C ' f . i  «N li'l;:.
M . I.

• I 1 ‘ ‘ I i . \ *i 'J-:? 1 Ov1
ESTIMATES OF ETHNIC GROUP VQTNG IN LUBBOCK COUn W  PRIMARY ELECTIONS

1986 Dem Primary 
Ct. Crim. App. PL 1 
Martinez (Hispanic) 
Dial+Duncan+Reagan 

Total
Supreme CL PI. 4 
Gonzalez (Hispanic) 
Ivy+Gibson-t-Humphreys 

Total
1986 Dem Runolf 
Supreme CL PI. 4 
Gonzalez (Hispanic) 
Gibson 

Total
Ct. Crlm . App. PI. 1 
Martinez (Hispanic) 
Duncan 

Total

Bivariate Regression Analysis 
Pearson r Estimates lor 

Siq. Anglo |Comb. Min

.97
.0000

.93
.0000

.87
.0000

.93
.0000

16
84
100

36
64
100

36
64
100

24
76
100

98
2

100

97
3

100

97
3

100

103
-3

100

Multiple Regression Analysis 
Partial r-H/B Estimates lor 

Stg.^H/B^ Anglo| Hispanic | Black

.98/.80
.0000/.0000

.95/.56 
.0000/.0015

.78/.66 
.0000/.0002

.88/.77
.0000/.0000

15
85

100

35
65

100

36
64

100

24
76

100

108
-8

100'.1

106
-6

100

97
3

100

105
-5

100

61
39
100

86
14

100

96
4

100

98
2

100

1986
Homogeneous Precinct! 
90-100% I 80-100% 

Anglo | Comb, Min

22
78
100

41
59
100

46
54
100

32
68
100

1 1
79
21
100

89
11

100

94 
6

100

95 
5

100

Are ethnic 
groups polarized? 
Comb. Min/Anqlo

Are Hisp. 
& Black 

cohesive?

Does
Comb. Min. 
choice win?

YES

YES

YES

YES

YES

YES

YES

YES

YES

NO

YES

NO

3-



-t
o 
-■

'1 . ■ «;■! •' : :&L.} f Olvl.i. 1: I.I.
a i ■ /:: n •'•. ,;:!y ; \‘ Tfieojl Pfpci id t.:',!: . : ! • i’. . I S

; ! lal ::
.....; JJ ' :1 ! 1

' i.i •. 
• T.f

■ • ..9i 1 . C i i '
'•0 111) Ivl ; ; ■:(

. . Ii i\1,i i 
: ; '-.mi

ESTIMATES OF ETHNIC GROUP VOTNG IN ECTOR COUNTY PRIMARY ELECTIONS: 1986
Bivariate Regression Analysis 
Pearson r| Estimates lor 

Sig. | Anglo |Comb. Min

Multi 
Partial r- H/B 

Sig.-H/B

ole Regression Analysis 
Estimates lor 

Anglo| Hispanic | Black

Homogenec
90-100%

Anglo

>us Precincts 
80-100% 

Comb. Min.

Are ethnic 
groups polarized? 
Comb. Min/Anglo

Are Hisp. 
& Black 

cohesive?

Does
Comb. Min. 
choice win?

1986 Dem. Primary
Supreme Ct. PI. A .80 .46/.71 YES NO NO
Gonzalez (Hispanic) .0000 11 53 .0381/.0002 13 42 65 14 50
Ivy+Gibson+Humphreys 89 47 87 58 35 86 50

Total 100 100 100 100 100 100 100
Ct. Crlm. App. PI. 1 .78 .50/.62 YES YES YES
Martinez (Hispanic) .0000 15 74 .0178/.0019 15 68 81 19 68
Dial+Duncan+Reagan 85 26 ' ' 85 32 ° 19 ' 81 32

Total 100 100 100 100 100 100 100

i ■' 9 l
"i v :.
1 ( 1  I vO ■ ( i.

m



O' ] i ' ! ill i;: 11

ESTIMATES OF ETHNIC GROUP VOTNG IN MIDLAND COUNTY ELECTIONS: 1586
Bivariate 
Pearson r 

Siq.

Regression Analysis 
•Estimates for;

Ahqlo | 1 Comb Min.

Homogenec
9 0 - 1 0 0 %

Anqlo

jus Precinct Est 
90-1 00%
- Comb Min.

' Are ethnic 
groups 

polarized?

Does
Comb Min. 

choice win?
1986 General Election J . J M , f *;|i. ! " I; , >; I . n.
S u p re m e  Ct. PI. A .96 YES ND
Gonzalez (Hispanic) .0000 25 90 32 89 I: .

Bales 75 1 0 68 1 1
Total 100 1 00 100 1 00

1986 General E lection
J P  P I  1 .96 YES ND
Watson (Black) .0000 1 9 9 1 26 90
Jobe. 81 —  ,9 74 . 1 On. ,,

Total 100 100 1 00 1 00
1988 General Election
S u p re m e  Ct. PI. 3 .89 YES ND
Gonzalez .0006 34 85 37 9 1
Howell+Scholz 66 1 5 63 9

Total 100 100 1 00 100

1*1 -O'-L



ESTIM A TES OF ETHNIC  G RO UP YO TKG  IK M ID LA N D  CO U K TY E L E C T IO N S : 1906  -19 88
Sivarisdte f  
Pearsonr 

Sig.

tegression Analysis 
Estimates for 

Anglo IComb.Mn

NMlip 
Partial r- Hi 6 

Sig.-H/B

le Regtession Analysis 
Estimates for 

Anglo | Hispanic | Black

Homogeneo 
90-1 cost; 

Anglo

us Precincts 
80-100/8 

Comb. Ivin.

Are ethnic 
groups polarized? 
Comb Ivln/Anglo

Are Hisp. 
& Black 

cohesive?

Does
Comb. Ivin, 
choice win?

198b G enera l E lec tion
S*rpne*ie C t. P i. -f .96 .86.'.99 YES YES NO
Gonzalez (Hispanic) .0000 24 90 .oooo; oooo 24 105 78 28 85
Bales 76 10 76 -5 22 72 15

ToiaJ 100 100 100 100 100 100 100
1986 G en era l E lec tion
J F F /  f .96 .87/81 YES YES NO
Watson (Black) .0000 17 91 .oooo; oooo 17 106 79 21 85
Jobe 83 9 83 -6 21 79 15

ToiaJ 100 100 100 100 100 100 100
1988 G en era l E lection
Supreme Ct. Pi. 3 .96 84/.82 YES YES NO
Gonzalez .0000 34 91 .0000/.0000 34 99 86 37 86
Howell+Scholz 66 9 66 1 14 63 14

.Total 100 100 100 100 100 100 100



APPENDIX "B"
Plaintiffs' Re-Evaluation of 

Defendants' Statistical Analysis



PLAINTIFFS' RE-EVALUATION OF DR. TAEBEL'S REPORTS
DALLAS COUNTY

Page # of TaebelExhibit - Year Race
Did Whites & BlacksVote Differently?

Did BlackChoiceWin?

Judicial Elections With Black Candidates:
General Elections:
District Court:

1 1980 191st Dist Ct Yes No
21 1984 Cr. Dist. 2i Yes No
37 - 1984 301st Dist Ct Yes No
69 - 1986 256th Dist Ct Yes No
73 1984 195th Dist ct. Yes No
89 1988 95th Dist Ct Yes No

County court at Lawr-.-s- •

17 1982 ;rCo- Cr- 6 Yes No
Justice .of the Peace Court r; .None
Appellate. Court> None .j.-c Cou

Primary Elections:
District Court: _ _ ___ _

81 . 1988 ,Cr. Dist 2 [RP] No —  - Yes
County Court at Law: __ .... . .

13 1982 Co. Cr. 6 [RP] Yes No
Justice of the Peace Court: None 
Appellate Court: None

1



Judicial Elections Without Black Candidates:

General Elections: 
District Court:

5 1980 95th Dist Ct . Yes No
9 1982 191st Dist Ct Yes No
25 1984 Cr. Dist 3 Yes No
33 1984 162nd Dist Ct Yes No
77 1986 298tg Dist Ct Yes .. - -- No

County Court at Law:
Justice'of the-.:Peace'Court: None
Appellate -■Court ••

65 i-- i ci v.K1986,-_...S-Ct. 4 Yes - - No
85 1988 S Ct 3 Yes Yes

Primary Elections: --- 
District Court: None

■ r '\ 1 Q r-‘ \ * i (; J ”'
County -Court .at Law: -None 
Justice of the Peace Court: None 
Appellate Court:

29 1984 Ct Cr App [DP] Yes No
41 1986 S. Ct. 4 [DP] Yes Yes
45 1986 Ct Cr App [DP] Yes Yes
49 1986 S Ct. 4 [DP-RO] No Yes
53 1986 Ct Cr Ap[DP-RO] Yes Yes

Non-Judicial Elections With Black Candidates: None

2



Non-Judicial Elections Without Black Candidates:
57 1986 Lt Gov Yes Yes
61 1986 Atty Gen Yes No

Judicial Elections with Black Candidates
JudicialElections without Black Candidates
Non-Judicial Elections with Black Candidates

SCORECARD
Whites/Blacks BlackVote Differently Choice Win

8 of 9 1 of 9

" ll"'of 12 5 Of 12-

0- of OL i 0 of O' -- •
Non-Judicial Elections without Black Candidates • • T~ 2 of 2 1 Of 2 -



PLAINTIFFS' RE-EVALUATION OF DR. TAEBEL'S REPORTS
BEXAR COUNTY

Page # of Did Whites Did HispTaebel & Hisps. ChoiceExhibit Year Race _ .Vote Differently? Win?

Judicial Elections With Hisp. Candidates: 
General Elections:
District Court: -«■■ ■ ~ ■ -- - . . • u:- :

5 1980 187st Dist Ct Yes Yes
15 1982 144th Dist Ct Yes No
16 1982 290th Dist Ct Yes No
18 1984 „ 37th Dist^Ct Yes No
19 1986 .,285th Dist Ct. Yes n v.. No
25 1988 73rd Dist Ct Yes No
26 1988 225th Dist Ct Yes No

County Court at Law:
20 1986 Co. Ct. 4 '' ' Yes No
27 1988 Co. Ct. "2Ap( Li:?y!£ Yes

Justice of the Peace Court: None
Appellate Court:

4 1980 Ct App Yes Yes
28 1988 Ct App Yes No

Primary Elections:
District Court:

2 __ 1980 131 Dist-Ct[DP] Yes No
3 1980 187 Dist Ct[DP] Yes No

1



7 1982 285 Dist Ct[DP] Yes No
9 1982 285 Dist Ct[DP] Yes Yes
10 1982 288 Dist Ct[DP] Yes Yes
11 1982 289 Dist Ct[DP] Yes No
12 1982 290 Dist Ct[DP] Yes Yes
17 1984 37 Dist Ct [DP] Yes Yes
22 1986 150 Dist Ct[RP] Yes No

1980 187 Dist Ct[DP] No No
1980 131 Dist Ct[DP] Yes No

County Court at LaWJ l , . n, ; nn J. ; hh n,
13 1982 Co. Cr. 3 [DP] Yes NO
14 1982 Co. Ct. 4 [DP] Yes No
23 1988 Co. Ct. 2 [DP] Yes Yes

Justice of the Peace Court: None 
Appellate Court:

1 . 1980 Ct App [DP] Yes‘_- j ,• =r.-f; Yes
6 juaicA?®2 Ct App [DP-RO] Yes No
8 h 1982 .Ct- App [DP] Yes . ... .. No
24 ... 1988 Ct App [DP] Yes No

Judicial Elections Without Hisp. Candidates:

General Elections:
District Court:
County Court at Law:
Justice of the Peace Court:
Appellate Court:

2



District Court:
County Court at Law:

21 1986 Co. Ct. 5[DP] Yes Yes
Justice of the Peace Court:
Appellate Court:

Primary Elections;

Non-Judicial Elections With Hisp. Candidates: None 

Non-Judicial Elections Without Hisp. Candidates: None

i n.iii «c

SCORECARD
i • Whites/Hisps. Hisp.Vote Differently Choice Win

Judicial
Elections with co. ci . --Hisp. Candidates 28 of 29 9 of 29
JudicialElections without - - --Hisp. Candidates 1 of 1 1 of 1
Non-Judicial Elections with
Hisp. Candidates . w ;; - 0 of 0 -- 0 of 0
Non-Judicial Elections without
Hisp. Candidates 0 of 0 0 of 0

3



PLAINTIFFS' RE-EVALUATION OF DR. TAEBEL'S REPORTS
TARRANT COUNTY

Page # of TaebelExhibit Year Race
Did Whites & BlacksVote Differently?

Did BlackChoiceWin?

Judicial Elections With Black Candidates: 
General Elections:- --- w . ------- -—-- . - -
District Court:

29 1986 Cr. Dist. 1 Yes No
33 1986 Cr. Dist. 4

l • *. Yes Yes
57 1988 Cr. Dist. 2 Yes No

County Court at Law: None 
Justice of the Peace Court: None 
Appellate Court: None

Primary Elections:
District Court: None 
County Court: at Law:r :

1 1982 -"Co. Cr. 1 [DP] Yes J v No
37 1986 ~ Co. Cr. 6 [DP] Yes ■=- No

Justice of the Peace Court: None 
Appellate Court: None

Judicial Elections Without Black Candidates:

General Elections:_____
District Court:

13 1982 233rd Dist Ct Yes Yes

1



17 1982 297th Dist Ct Yes Yes
21 1986 233rd Dist Ct Yes No
25 1986 325th Dist Ct Yes Yes
61 1988 17th Dist Ct Yes No

County Court at Law:
9 1982 Co. Cr. 4 Yes Yes

Justice of the Peace Court: None
— - •

Appellate Court:
49 1986 S. Ct. 4 Yes Yes
65 1988 S. Ct. 3 Yes Yes

Primary Elections: 
District Court: None 
County Court at Law:

5 1982 Co. Cr. 4 [DP] Yes Yes
Justice of 
Appellate

the
Court

Peace Court: None
•
•

41 1986 Ct.Cr.App. [DP] Yes Yes
49 1986 S. Ct. 4 [DP] Yes Yes

Non-Judicial Elections With Black Candidates: None

Non-Judicial Elections Without Black Candidates:
45 1986 Atty Gen Yes No

2



Judicial Elections with Black Candidates

SCORECARD
Whites/Blacks Black Vote Differently Choice Win

5 of 5 1 of 5
JudicialElections withoutBlack Candidates ...- • 11 of 11 8 of 11
Non-Judicial Elections with Black Candidates 0 of 0 0 of 0
Non-Judicial : - ! —  * - -
Elections withoutBlack Candidates 1 of 1 0 of 1

C -V.tl' ''' _1 _ •? ’ f
3 *• > m ij p

3



PLAINTIFFS1 RE-EVALUATION OF DR. TAEBEL'S REPORTS
TRAVIS COUNTY

Page # of Did Whites Did HispTaebel __ & Hisps. ChoiceExhibit Year Race Vote Differently? Win?

Judicial Elections With Hisp. Candidates;
General Elections:
District Court:
County Court at Law:
Justice of the Peace Court:
Appellate Court:

29 1986 S Ct 4 No Yes
45 1988 S Ct 3 No Yes

Primary Elections: 
District Court:

-----

37 _ 1988 345 Dist Ct[DP] Yes No

County ■1. i. y c 7 e nCourt at Law: N'

33 1988 Co. Ct. 1 [DP]" Yes No
41 1988 Co. Ct. 7 [DP] Yes No

Justice of the Peace Court:
Appellate Court:

1 1984 Ct Cr A [DP] Yes No
9 1986 Ct Cr A [DP] Yes No
21 1986 S Ct 4 [DP] No Yes
25 1986 S Ct 4 [DP-RO] No Yes

1



Judicial Elections Without Hisp. Candidates:

General Elections:
District Court:
County Court at Law:
Justice of the Peace Court:
Appellate Court:

49 ~ 1988 rS Ct 4 No Yes

Primary Elections:
District Court: - —
County Court at Law:
Justice of the Peace Court:
Appellate Courts________

Non-Judicial Elections With Hisp. Candidates:
5 1984 St Sen 14 No Yes
13 1986 Atty Gen No Yes

1984 St Sen [DP-RO] Yes Yes
1984 St Sen [DP] Yes Yes

Non-Judicial Elections Without Hisp. Candidates:

17 1986 Lt Gov No Yes

2



SCORECARD
Hisp.Choice WinWhites/Hisps. Vote Differently

Judicial Elections withHisp. Candidates 5 of 9
JudicialElections without Hisp. Candidates
Non-Judiciai Elections with Hisp. Candidates
Non-Judicial Elections without Hisp. Candidates 0 of 1

0 of 1 

2 of 4

4 of 9 

1 of 1 

4 Of 4 

1 Of 1

3



PLAINTIFFS' RE-EVALUATION OF DR. TAEBEL'S REPORTS
JEFFERSON COUNTY

Page # of TaebelExhibit Year Race
Did Whites & BlacksVote Differently?

Did BlackChoiceWin?

Judicial Elections With Black Candidates:
General Elections: - 
District Court: None 
County Court at Law: None 
Justice of the Peace Court: None 
Appellate Court: None

Primary Elections:
District Court: None
County Court at Law: None ____  ... ___
Justice of the Peace Court: None 
Appellate Court: None

Judicial Elections Without Black Candidates:

General Elections:
District Court: None 
County Court at Law: None 
Justice of the Peace Court: None 
Appellate Court:

10 1986 S. Ct. 1 No - - Yes
17 , .-.--1986 S. Ct. # 4 No Yes

1



Primary Elections: 
District Court: None
County Court at Law: None 
Justice of the Peace Court: None 
Appellate Court:

7 1986 Ct.Cr.App. [DP] Yes No
13 1986 S. Ct. 4 [DP] Yes Yes

Non-Judicial Elections With Black Candidates:
1 -1982- St.Rep 22 Yes Yes
4 ■ L j-3.984t -OStfRep 2 2  ' [ ‘ Yes Yes

Non-Judicial Elections Without Black Candidates:
19 1986 Gov.
22 1986 Atty Gen

Yes
No

Yes
Yes

__■ _  __ *; 7 . _

i r- LIT. c*1 . ................nr
SCORECARD

Whites/Blacks Black
• ; ! • -> ' .  C  ' ’ * 1 i •' !/'. * Vote'differently Choice Win

Judicial- ’ ; ! 
Elections with 
Black Candidates 0 of 0 0 of 0
Judicial
Elections without 
Black Candidates 2 Of 4 3 of 4
Non-Judicial 
Elections with 
Black Candidates 2 Of 2 2 of 2
Non-Judicial 
Elections without 
Black Candidates - 1 Of 2 — 2 of 2

2



PLAINTIFFS' RE-EVALUATION OF DR. TAEBEL'S REPORTS
LUBBOCK COUNTY

Page # of 
Taebel
Exhibit Year Race

Did Whites 
& Minorities 
Vote Differently?

Did Minority
Choice
Win?

Judicial Elections With Minority Candidates:
General Elections:
District Court: None 
County Court at Law: None
Justice of the Peace Court: None ____ ___  ^___
Appellate Court:

17 1986 S. Ct. # 4 Yes No
25 1988 s. c£. '# '3 Yes " No

Primary Elections: -- =—  ----- — ---  - • -— - - ---

District Court: None
j  s \ '  /"* /  DCounty Court at Law: None —

Justice of the Peace Court: None Wh.it- 3/Hj
JJ '  r> ■> f- f-

Appellate Court: None

Judicial Elections Without Minority Candidates: 

General-Elections:
District Court:- None  ̂ - --
County Court at Law:

1 1982 O o Ct. 1 Yes No
9 “T1986 Co. Ct. 2 Yes No

h - '  . •-> r~ \ * -  1 h  i (

1



Justice of the Peace Court: None 
Appellate Court:

21 1988 Ct. Cr. App. Yes No

Primary Elections: ’
District Court: None 
County Court at Law:

1 1982 Co. Ct. 1 [DP] No Yes
Justice of the Peace Court: None 
Appellate Court: None

Non-Judlcial-Elections With Minority Candidates:
13 <• 1986 : :'Atty Gen Yes V;-‘- No

Non-Judicial Elections Without Minority Candidates: None

SCORECARD

< ’'yC-LlULf ii ‘ • Whites/Minorities 
Vote Differently

Minority 
Choice Win

Judicial r— - r.-.- / - - - 
Elections with 
Minority Candidates 2 of 2 0 of 2
Judicial
Elections without 
Minority Candidates 3 of 4 1 of 4
Non-Judicial 
Elections with 
Minority Candidates 1 of 1 0 of 1
Non-Judicial 
Elections without 
Minority Candidates 0 of 0 0 of 0

2



PLAINTIFFS' RE-EVALUATION OF DR. TAEBEL'S REPORTS
ECTOR COUNTY

Page # of Did Whites Did Minority
Taebel & Minorities Choice
Exhibit Year Race Vote Differently? Win?

Judicial Elections With Minority Candidates:
General Elections:
District Court:
County Court at Law:
Justice of the Peace Court:
Appellate Court:

21 1986 S Ct
n r

4 Yes No
37 1988 s ct 3 Yes No

Primary Elections:
District Court:
County Court at Law:
Justice of-the Peace Court:---------- — ~
Appellate Court:

Judicial Elections Without Minority Candidates:

General Elections: 
District Court:

5 1980 161 Dist Ct Yes No
County Court at Law:

9 1982 Co Jud No Yes
13 1982 Co Ct Law No Yes

1



Justice of the Peace Court:
Appellate Court:

29 1988 S Ct 4 Yes No
33 1988 Ct App Yes No

Primary Elections:
District Court:
County Court at Law:
Justice of- the ^eace <tourt:
Appellate Court:

Non-Judicial Elections With Minority Candidates:
17 1986 Atty Gen Yes No

Non-Judicial Elections Without Minority Candidates:
1 1980 RR Com Yes No
25 1986 Lt Gov Yes No

2



V

SCORECARD
Whites/Minorities 
Vote Differently

Minority 
Choice Win

Judicial 
Elections with 
Minority Candidates . 2 . of. 2 0 of ,2
Judicial
Elections without 
Minority Candidates 3 of 5 2 of 5
Non-Judicial- 
Elections with 
Minority Candidates 1 of 1 0 of 1
Non-Judicial
Elections Without —  - 
Minority Candidates 2 of 2 0 of 2

2 5 -.-be 5 Ct
* »-r r~.

N o

"nr • • ------ ; i V- I

c- v . :

3



► 2

MIDLAND COUNTY
PLAINTIFFS1 RE-EVALUATION OF DR. TAEBEL'S REPORTS

Page # of Did Whites Did MinorityTaebel & Minorities ChoiceExhibit Year Race Vote Differently? Win?

Judicial Elections With Minority Candidates:
General Elections:
District Court:
County Court at Law:
Justice of the Peace Court: 
Appellate Court:

25 1986 S Ct 4 Yes NOl

29 1988 S Ct 3 Yes No

Primary Elections:
District Court:
County Court-at Law:
Justice of the Peace Court: 
Appellate Court:

9 1986 Ct Cr App [DP] "No-;; 7 " No
21 1986 S Ct 4 [DP] No "No

Judicial Elections Without Minority Candidates:

General Elections:
District Court:

1 1980 142 Dist Ct No Yes
County Court at Law: "■*

1



Justice of the Peace Court:
Appellate Court:

Primary Elections:
District Court:
County Court at Law:
Justice of the Peace Court:
Appellate Court:

Non-Judicial Elections With Minority Candidates:
5 1984 Co Atty [RP] Yes No
13 1986 Atty Gen [DP] No No
17 198 6 Atty Gen Yes No

Non-Judicial Elections Without Minority Candidates:

SCORECARD
Whites/Minorities Vote Differently Minority Choice Win

Judicial Elections with 
Minority Candidates 2 of 4 0 of 4
JudicialElections without Minority Candidates 0 of 1 1 of l
Non-Judicial Elections with Minority Candidates 2 of 3 0 of 3
Non-Judicial Elections without Minority Candidates 0 of 0 0 of 0

2