IMPLEMENTATION AND RESULTS OF THE TEXAS AUTOMATIC ADMISSIONS LAW

IMPLEMENTATION AND RESULTS OF THE TEXAS AUTOMATIC ADMISSIONS LAW
(HB 588)
at
THE UNIVERSITY OF TEXAS AT AUSTIN
REPORT NUMBER 4

DEMOGRAPHIC ANALYSIS
FALL 2001

ACADEMIC PERFORMANCE AND PERSISTENCE OF TOP 10% AND NON-TOP 10% STUDENTS
ACADEMIC YEARS 1996-2000

Prepared by:

Gary M. Lavergne, Director of Admissions Research
Office of Admissions
The University of Texas at Austin

Dr. Bruce Walker, Associate Vice President of Student Affairs and Director of Admissions
The University of Texas at Austin

Note: The authors wish to acknowledge and thank Cindy Hargett, Senior Systems Analyst, Student Information Systems, of the University of Texas at Austin for her valuable contributions to this project.

Printer Friendly PDF Version

Introduction

This is the fourth in a series of reports on the demographic makeup of top 10% students entering The University of Texas at Austin. These reports were developed to provide easy access to data for the general public, policy analysts, political decision-makers, and fellow academicians. It documents the university’s progress towards achieving diversity after the Cheryl J. Hopwood, et.al. v State of Texas decision in 1996.

This report also documents the persistence and performance of students admitted through a process fashioned by the passage of HB 588, a legislative response to Hopwood which requires the automatic admission of any student graduating in the top 10% of his/her high school graduating class, and compares them to fall 1996, the last year in which students were admitted under a classic admissions model using affirmative action to assure the accomplishment of the University’s diversity goals. Performance data focus on GPAs of the entering class of 2000 and disaggregated by college, gender, racial/ethnic groups and college admission test score intervals, and compares them to previous classes. Persistence and graduation data include five-year percentages from the entering class of 1996 to one-year persistence percentages for 2000. The report also includes demographic data for fall 2001 entering freshmen.

As stated in earlier reports, this report is but another chapter of a larger social and educational story. It is not an evaluation of past affirmative action policies. Nor is it a statement of the success or failure of policies using or not using affirmative action. Neither is it a proposed blueprint for other universities or public institutions seeking diversity. It is merely another chapter about what has happened at The University of Texas at Austin immediately before and since the end of the use of affirmative action in admissions.

Many contemporary assaults on the use of affirmative action to achieve diversity goals took the form of referenda or changes in administrative and legislative codes. These processes generally included public input and debate. Hopwood, on the other hand, was a judicial fiat. In 1996, the US Fifth Circuit Court of Appeals removed affirmative action as a tool available to the admissions offices of the states of Texas, Mississippi, and Louisiana. For UT Austin officials there was neither choice nor debate. If the university family was to bear any resemblance to the state’s general population, a wholesale re-evaluation and re-structuring of student affairs processes (especially admissions) had to take place.

The Texas Legislature responded with the passage of HB 588, more commonly known as the “Top 10% Law.” The law is often misinterpreted as an assault on affirmative action. In reality, it was legislation generated by the advocates of diversity, led by African American and Latino legislators, to increase the chances of enrolling a diverse class, especially at the flagship universities—The University of Texas at Austin and Texas A&M University at College Station. Governor George W. Bush signed the bill into law in the spring of 1997.

During the years immediately preceding Hopwood, The University of Texas used what has been described as a “Classic Model” for making admissions decisions. The model included factors that allowed for the easy and efficient processing of very large numbers of applications. Decisions were made after the distribution of students based on a predicted freshman grade point average. The formula used, a multiple regression equation, relied on the combined SAT score (or a concorded ACT score) and high school class rank. For a variety of reasons, strict adherence to such a policy could not, by itself, produce diverse classes. The ACT/SAT, like virtually all standardized tests, has a well-documented history of differential performance gaps among socio-economic, gender, and racial-ethnic groups. To meet its long-cherished diversity goals, and in order to enroll a freshman class bearing some resemblance to the state itself, the University used affirmative action. This policy was in place until Hopwood forced its removal. The last freshman class admitted under such a classic model was during the summer/fall of 1996.

With the entering class of 1997, the University’s admission policy expanded to include the following factors:

The Academic Index (AI)
High School Record:

Class rank
Completion of UT required high school curriculum
Extent to which students exceed the UT required units
SAT/ACT score

The Personal Achievement Index (PAI)

Scores on two essays
Leadership
Extracurricular Activities
Awards/honors
Work experience
Service to school or community
Special circumstances:

Socio-economic status of family
Single parent home
Language spoken at home
Family responsibilities
Socio-economic status of school attended
Average SAT/ACT of school attended in relation to student's own SAT/ACT

For purposes of comparison, however, the entering class of 1997 does not provide a good contrast with 1996 (the last year of affirmative action). 1997 was a transition year not affected by HB 588. For that reason, this report focuses on 1996 as a base year for illustrating the use of affirmative action, and compares it to 1998 through 2001.

(Note: This report and those which precede it do not presume 1996 to be an ideal or representative year for illustrating or evaluating affirmative action policies. 1996 is used because of a fair assumption that the latest data is preferred.)

Executive Summary -- Year 2001 Highlights

The percentage of White students, the traditional majority, making up UT Austin’s entering freshman class is at an all-time low (61%).
The fastest growing racial/ethnic group is Asian American, increasing from 17% to 19% in one year. The percentage of African Americans fell from 4% to 3% while Hispanics increased from 13% to 14%.
Since 1996, the difference in mean college admissions test scores between top 10% students and non-top 10% students has narrowed to an almost imperceptible difference—only 9 points on the SAT score scale.
There is no difference in the percentages of top 10% and non-top 10% entering freshmen exempting or passing the Texas Academic Skills Program (TASP), or requiring remedial activities on the UT Austin campus.
At every ACT/SAT score interval, top 10% students outperform non-top 10% students. Top 10% students perform about as well as their non-top 10% classmates with SAT scores about 200-300 points higher. This is true in all colleges and for all racial/ethnic groups.
In those colleges with a rigorous reputation (Business, Engineering, and Natural Sciences) mean test scores of top 10% students continue to decline while mean freshman GPAs remain stable.
In terms of freshman GPA, at every ACT/SAT score interval, both males and females in the entering class of 2000 matched or outperformed the 1996 cohort.

While much more investigation is needed, it appears that the legislative intent to de-emphasize the use of SAT scores in the admissions process to achieve pre-Hopwood diversity levels without sacrificing persistence and performance has been achieved – as least when compared to the last year of affirmative action (1996).

But as stated above, this is a small part of a larger social and educational story.

HB 588’S IMPACT OF THE DIVERSITY OF THE ENTERING FRESHMEN CLASS OF 2000

As was the case with previous reports, and unless otherwise specified, populations were selected from all summer and fall enrolled students. Since HB 588 was written to benefit the graduates of Texas high schools, the populations in this report are of Texas high school graduates. The Texas Higher Education Coordinating Board, the rule-writing authority for the implementation of HB 588, defined a top 10% student as someone who the high school ranked out of a specific class size. These are referred to as “explicit top 10%” students who were automatically admitted to The University of Texas as a result of HB 588. In this report all references to “top 10%” are “explicit top 10%.”

Table 1 below illustrates the growth of top 10% students admitted to UT since 1997.

TABLE 1
TOP 10% STUDENTS ADMITTED TO UT-AUSTIN BY RACIAL/ETHNIC BACKGROUND

ETHNICITY	1997	1998	1999	2000	2001
White	2262	2561	2753	3182	3213
African American	118	143	268	291	245
Asian American	803	863	998	1034	1081
Hispanic	613	734	911	1020	1012

For the first time since the implementation of HB 588, there were fewer top 10% admits in a racial/ethnic group—African Americans, which declined from 291 in 2000 to 245 in 2001, and Hispanics, which declined slightly from 1020 to 1012. The largest increase was among Asian Americans (+47).

Table II below shows that the diversity level of the entering class of 1996 (pre-Hopwood) was restored by the fall of 1999. Since 1999, the diversity levels of the entering classes have remained stable. Before interpreting changes in n-counts, the fact that the entering class of 2001 is smaller than 2000 must be considered. The reduction in class size (-349) was more than absorbed by Whites (-354), and African Americans (-54). The percentage of White students continues to decline. The percentage of African Americans declined slightly as well (from 4% to 3%) from last year’s five-year high. The Asian American and Hispanic populations increased and are at all-time highs at 19% and 14% respectively. Since 1996, the largest changes in demographics came from the White (-4%) and the Asian American (+4%) populations.

TABLE II
FRESHMAN ENROLLMENT
1996-2000

Ethnicity	1996		1998		1999		2000		2001
	N	%	N	%	N	%	N	%	N	%
White	4159	65	4399	65	4447	63	4801	63	4447	61
Native American	34	<1	37	<1	28	<1	32	<1	34	<1
African American	266	4	199	3	286	4	296	4	242	3
Asian American	942	15	1133	17	1221	17	1325	17	1413	19
Hispanic	932	14	891	13	976	14	1011	13	1024	14
Foreign	97	2	83	1	82	1	217	3	139	2
Unknown	0	0	2	<1	0	0	4	<1	38	<1
Total	6430		6744		7040		7686		7337

ACADEMIC PREPARATION OF TOP 10% STUDENTS

For the past few decades, institutions have commonly measured the academic preparedness of its entering freshman class through the use of aggregate college admissions test scores. (It is the only nationally validated and standardized way to do it.) The University of Texas at Austin requires its applicants to submit at least one set of scores from two available college admissions tests: the ACT Assessment or the SAT I: Reasoning Test. The vast majority of UT’s applicants submit SAT scores. For purposes of this study, ACT scores were concorded to the SAT scale using a concordance table developed by the Measurement and Evaluation Center at UT Austin. (The relationship between ACT and SAT scores is well-documented. A national concordance study by ACT/ETS/College Board showed that the correlation between SAT and ACT scores for over 105,000 students was .92. A new concordance study conducted recently by the Office of Admissions Research and ACT, Inc., using a dataset of over 12,600 UT Austin applicants over a two year period, showed a correlation of .87.) From 1996 through 2001, UT admissions policy was to use the best combined score (a concorded ACT Composite or the SAT Combined score) for a single test date for each student.

Table III below shows test score differences between top 10% students and non-top 10% students. Since 1996 the mean scores of the top 10% students have fallen from 1242 to 1211. After an initial decline from 1996 to 1997, the non-top 10% group’s scores have risen to a mean only 9 points (on the SAT scale) below the top 10% group. This is almost certainly the result of the passage of HB 588 and UT Austin’s response. Scores play no role in the admission of top 10% students, and for non-top 10% decisions, test scores are combined with many other factors.

TABLE III
MEAN COLLEGE ADMISSION TEST SCORES OF ENROLLED TOP 10% AND NON-TOP 10% STUDENTS

Note: The numbers below, and throughout this report, represent SAT combined scores and concorded ACT scores. Students submitting more than one set of scores were given the benefit of the best performance from a single test date.

	FALL 1996	FALL 1997	FALL 1998	FALL 1999	FALL 2000	FALL 2001
Top 10%	1242	1237	1230	1218	1212	1211
Non-Top 10%	1186	1168	1180	1180	1193	1202

Table IV illustrates a distribution of students by SAT point intervals and approximate ACT values. The decline in mean scores for top 10% students since 1996, from 1242 in 1996 to 1211 in 2001 (-31), comes from increases in the percentages of students in the lower score intervals. For example, in 1996 the percentage of top 10% students with scores below 1000 was 4.4; in 2000 it more than doubled to 10.0% and increased slightly to 10.1% in 2001.

TABLE IV
SAT SCORE RANGES FOR All ENROLLED FRESHMEN TOP 10% STUDENTS
SAT SCORE RANGES FOR All FRESHMEN TOP 10% STUDENTS
FALL, 1996-2000

SAT Ranges (ACT Values)	1996		1998		1999		2000		2001
SAT Ranges (ACT Values)	N	%	N	%	N	%	N	%	N	%
<900 (<18)	26	1.1	42	1.7	92	3.2	126	3.8	113	3.3
900-990 (18-21)	81	3.3	130	5.2	160	5.5	208	6.2	233	6.8
1000-1090 (21-23)	320	13.2	325	12.9	391	13.4	467	14.0	455	13.3
1100-1190 (23-26)	512	21.1	511	20.3	576	19.7	699	20.9	743	21.7
1200-1290 (26-28)	585	24.1	615	24.5	722	24.7	753	22.5	761	22.2
1300-1390 (29-31)	510	21.0	508	20.3	579	19.8	616	18.4	634	18.5
1400-1490 (31-33)	296	12.2	297	11.8	318	10.9	368	11.0	378	11.0
1500+ (34-36)	98	4.0	85	3.4	87	3.0	109	3.3	106	3.1
Total	2428	100	2513	100	2925	100	3346	100	3423	100
SAT Mean	1242		1230		1218		1212		1211

THE TEXAS ACADEMIC SKILLS PROGRAM (TASP) RESULTS

The State of Texas mandates that its colleges and universities measure student readiness for college-level study through the use of the Texas Academic Skills Program (TASP) test. TASP consists of three achievement tests: reading, mathematics, and writing. Students can be exempt from TASP by an acceptable performance on either the SAT, ACT or the Texas Assessment of Academic Skills (TAAS—Texas’ state high school exit test). Table V outlines TASP results for top 10% and non-top 10% students entering the University during the summer/fall 1998 through 2001. Students are classified as "exempt" (had high enough TAAS or ACT/SAT scores), "passed" (not exempt but made passing scores on TASP), or "remediation" (scored too low on TASP and participated in required remedial activities). Because of its readiness mission, TASP focuses on minimal skills, as evidenced by the very low percentages of UT students needing remediation even among non-top 10% students. For the past three years only 1% of both top 10% and non-top 10% UT entering freshmen have required remediation, and on any given year 79% to 91% of freshman groups have exempted, and thus, never took TASP.

TABLE V
TASP RESULTS FOR TOP 10% AND NON-TOP 10% STUDENTS
1998-2001

	1998		1999		2000		2001
	Top 10%	Non-Top 10%	Top 10%	Non-Top 10%	Top 10%	Non-Top 10%	Top 10%	Non-Top 10%
Exempt	89%	82%	90%	85%	90%	88%	91%	91%
Passed	10%	15%	9%	14%	8%	10%	8%	8%
Remediation	1%	3%	1%	1%	1%	1%	1%	1%
N-count	2513	3597	2925	3596	3346	3713	3423	3255

FRESHMAN YEAR PERFORMANCE OF TOP 10% STUDENTS

In the second of this series of HB 588 reports, Dr. Bruce Walker documented the relationship between higher class rank, higher ACT/SAT scores, and lower incidences of required remediation for Top 10% students than their lower-ranking peers. With the exception of remediation rates, that trend continued with the entering class of 2000. So did superior performance by top 10% students after one year of schoolwork at UT. Table VI below shows one-year performance as measured by GPA. The data are for entering freshmen.

At every score interval, i.e., among students with comparable ACT/SAT scores, top 10% students outperformed non-top 10% students. There is a strong linear relationship between scores and GPAs, but the relationship is less evident when observations are made over time. For example, in 1996, top 10% students earned a GPA of 3.21; in 2000, top 10% students had a mean ACT/SAT that was 30 points lower, but they earned a 3.26.

TABLE VI
FRESHMAN YEAR PERFORMANCE BY SAT SCORE RANGE
1996 VS. 1998 VS. 2000
ALL STUDENTS

1996

1998

2000

SAT Ranges
(ACT Value)

TOP 10%

NON-TOP 10%

TOP 10%

NON-TOP 10%

TOP 10%

NON-TOP 10%

GPA

<900
(<18)

2.45

2.27

2.48

2.32

126

2.67

2.43

900-990
(18-21)

2.66

164

2.44

130

2.70

220

2.39

208

2.84

157

2.61

1000-1090
(21-23)

320

2.85

546

2.52

325

2.91

608

2.62

467

2.95

597

2.71

1100-1190
(23-26)

512

3.05

982

2.62

511

3.01

1,004

2.68

699

3.12

1037

2.77

1200-1290
(26-28)

585

3.24

965

2.67

615

3.27

940

2.78

753

3.31

1041

2.89

1300-1390
(29-31)

510

3.40

457

2.81

508

3.47

538

2.86

616

3.52

564

3.07

1400-1490
(31-33)

296

3.58

161

3.06

297

3.67

169

3.16

368

3.67

190

3.27

1500+
(34-36)

3.65

3.05

3.74

3.19

109

3.77

3.12

Total/Mean

2,428

3.21

3,375

2.65

2,513

3.23

3,597

2.72

3,346

3.26

3,713

2.86

SAT Mean

1242

1186

1230

1180

1212

1193

As Dr. Walker pointed out in the second report, it is not surprising that the combination of SAT scores and class rank produces superior predictions than the use of either variable alone. While certainly not definitive, the data in this and previous reports strongly suggest that the admission of large numbers of students with lower ACT/SAT scores has not had a measurable negative effect on student performance at UT-Austin.

Tables VI (a) through VI (d) show identical breakdowns as Table VI but for each major racial/ethnic group. Overall, the patterns are similar, but great care should be taken not to over-interpret some cells of data driven by very small n-counts. The performance of top 10% students is observably better. For purposes of clarity, the data are presented in two-year snapshots.

TABLE VI (a)
FRESHMAN YEAR PERFORMANCE
BY SAT SCORE RANGE
1996 VS. 1998 VS. 2000
WHITE STUDENTS

1996

1998

2000

SAT Ranges
(ACT Value)

TOP 10%

NON-TOP 10%

TOP 10%

NON-TOP 10%

TOP 10%

NON-TOP 10%

GPA

<900
(<18)

0.00

2.18

2.56

2.35

2.74

2.36

900-990
(18-21)

2.87

2.37

2.63

119

2.48

2.76

2.58

1000-1090
(21-23)

145

2.82

305

2.54

145

2.89

380

2.60

211

2.93

384

2.70

1100-1190
(23-26)

315

3.05

646

2.65

308

3.00

720

2.69

411

3.12

733

2.77

1200-1290
(26-28)

363

3.20

705

2.66

405

3.27

666

2.78

495

3.32

728

2.89

1300-1390
(29-31)

367

3.41

321

2.81

339

3.46

401

2.81

409

3.49

403

3.06

1400-1490
(31-33)

211

3.53

116

3.07

192

3.64

117

3.17

249

3.65

126

3.24

1500+
(34-36)

3.65

3.10

3.72

3.37

3.78

3.26

Total/Mean

1497

3.24

2215

2.68

1497

3.26

2457

2.73

1921

3.31

2529

2.86

SAT Mean

1268

1202

1254

1192

1247

1199

TABLE VI (b)
FRESHMAN YEAR PERFORMANCE
BY SAT SCORE RANGE
1996 VS. 1998 VS. 2000
AFRICAN AMERICAN

	1996				1998				2000
SAT Ranges (ACT Value)	TOP 10%		NON-TOP 10%		TOP 10%		NON-TOP 10%		TOP 10%		NON-TOP 10%
SAT Ranges (ACT Value)	N	GPA	N	GPA	N	GPA	N	GPA	N	GPA	N	GPA
<900 (<18)	3	1.89	19	2.28	4	2.56	23	2.28	29	2.53	15	2.28
900-990 (18-21)	6	2.46	13	2.48	11	2.57	16	2.41	30	2.77	13	2.88
1000-1090 (21-23)	28	2.65	52	2.56	22	2.77	28	2.80	36	3.07	30	2.67
1100-1190 (23-26)	30	2.86	45	2.62	19	3.18	23	2.66	26	2.82	33	2.80
1200-1290 (26-28)	20	3.27	28	2.29	7	3.04	18	2.89	23	3.19	24	2.57
1300-1390 (29-31)	4	3.06	5	2.20	3	3.56	8	3.10	11	3.36	11	3.02
1400-1490 (31-33)	-	-	2	2.97	3	3.93	3	3.16	1	4.00	2	3.84
1500+ (34-36)	-	-	-	-	-	-	-	-	-	-	1	2.29
Total/Mean	91	2.84	164	2.48	69	2.95	119	2.66	156	2.91	129	2.70
SAT Mean	1126		1086		1092		1065		1053		1107

TABLE VI (c)
FRESHMAN YEAR PERFORMANCE
BY SAT SCORE RANGE
1996 VS. 1998 VS. 2000
ASIAN AMERICAN

1996

1998

2000

SAT Ranges
(ACT Value)

TOP 10%

NON-TOP 10%

TOP 10%

NON-TOP 10%

TOP 10%

NON-TOP 10%

GPA

<900
(<18)

2.36

2.37

2.88

2.09

2.63

2.45

900-990
(18-21)

2.93

2.62

2.84

2.29

3.07

2.82

1000-1090
(21-23)

3.09

2.66

3.01

2.71

2.98

2.75

1100-1190
(23-26)

3.12

129

2.59

3.13

142

2.75

118

3.25

143

2.75

1200-1290
(26-28)

116

3.40

128

2.72

131

3.34

153

2.86

136

3.37

184

2.96

1300-1390
(29-31)

3.47

2.88

123

3.50

3.00

146

3.59

106

3.19

1400-1490
(31-33)

3.72

3.24

3.72

2.98

102

3.72

3.26

1500+
(34-36)

3.68

3.69

3.78

2.68

3.78

3.04

Total/Mean

430

3.38

461

2.74

519

3.38

542

2.79

653

3.40

606

2.94

SAT Mean

1269

1209

1271

1194

1250

1219

TABLE VI (d)
FRESHMAN YEAR PERFORMANCE
BY SAT SCORE RANGE
1996 VS. 1998 VS. 2000
HISPANIC

1996

1998

2000

SAT Ranges
(ACT Value)

TOP 10%

NON-TOP 10%

TOP 10%

NON-TOP 10%

TOP 10%

NON-TOP 10%

GPA

<900
(<18)

2.67

2.34

2.43

2.40

2.72

2.66

900-990
(18-21)

2.48

2.49

2.73

2.25

2.85

2.40

1000-1090
(21-23)

100

2.84

123

2.38

106

2.94

104

2.57

140

2.95

2.70

1100-1190
(23-26)

3.01

155

2.50

2.86

107

2.59

140

3.04

113

2.80

1200-1290
(26-28)

3.23

2.81

3.11

2.62

3.24

2.84

1300-1390
(29-31)

3.27

2.63

3.46

3.02

3.54

2.91

1400-1490
(31-33)

3.60

2.31

3.60

3.46

3.60

3.37

1500+
(34-36)

3.50

1.34

3.71

2.02

3.27

3.46

Total/Mean

396

3.00

506

2.52

414

2.96

441

2.59

591

3.04

401

2.78

SAT Mean

1142

1123

1113

1123

1093

1143

The second report in this series also addressed concerns that HB 588 might result in the admission of significant numbers of students unable to perform as well as UT freshmen did when the University used a classic admissions model—admitting students based on a predicted freshman GPA. Of special concern were those UT schools with an earned reputation for being both selective and rigorous. Table VII below examines three such schools: Business, Engineering, and Natural Science. Entering freshmen from 1996, 1998 and 2000 were compared to determine whether there were significant differences in mean GPA. None were observed during a time when each of the mean ACT/SATs declined noticeably. In summary, for three rigorous UT schools, from 1996, a year in which affirmative action was allowed, to 1998 and 2000, the second and fourth year of the implementation of HB 588, earned freshman GPA remained stable and in some cases increased slightly. As noted earlier, 1998 and 2000 had classes nearly identically as diverse as 1996.

TABLE VII
SAT MEAN AND FRESHMAN GPA FOR TOP 10% STUDENTS
BY SELECTED UNDERGRADUATE COLLEGES AND ETHNICITY

Note: N-counts are based on the number of students with ACT/SAT scores.

1996

	BUSINESS			ENGINEERING			NATURAL SCIENCE
	GPA	SAT	N	GPA	SAT	N	GPA	SAT	N
White	3.36	1266	318	3.35	1326	208	3.11	1262	333
African American	3.02	1164	20	2.50	1114	13	2.87	1128	19
Asian American	3.43	1273	97	3.44	1344	83	3.34	1237	157
Hispanic	3.23	1167	58	2.98	1214	68	2.96	1123	99
All	3.34	1252	495	3.28	1302	374	3.14	1229	613

1998

	BUSINESS			ENGINEERING			NATURAL SCIENCE
	GPA	SAT	N	GPA	SAT	N	GPA	SAT	N
White	3.28	1253	382	3.31	1322	281	3.17	1235	314
African American	3.09	1096	21	2.94	1054	5	3.04	1131	16
Asian American	3.49	1276	141	3.36	1315	126	3.29	1240	167
Hispanic	2.90	1091	78	2.98	1160	84	2.99	1133	118
All	3.27	1232	625	3.27	1290	497	3.17	1214	621

2000

	BUSINESS			ENGINEERING			NATURAL SCIENCE
	GPA	SAT	N	GPA	SAT	N	GPA	SAT	N
White	3.29	1238	517	3.28	1290	291	3.26	1251	402
African American	2.78	1040	39	3.27	1150	22	2.81	1056	39
Asian American	3.48	1250	207	3.36	1289	153	3.36	1233	193
Hispanic	3.05	1111	132	2.94	1126	129	3.09	1082	133
All	3.27	1214	901	3.23	1250	600	3.23	1208	776

These data present a “broad brush” analysis and do not suggest an absence of academic problems. In some colleges a 2.75 GPA is required to enter upper-level classes. Even though GPAs held steady, and in some cases improved, some African American and Hispanic average GPAs are very near the 2.75 level, and are unacceptably low, suggesting strongly that large numbers of students from these groups are at-risk of not moving to upper division courses. This requires additional study.

Table VII (a) below illustrates the first year academic performance of top 10% and non-top 10% students by all colleges at UT Austin for the entering class of 2000. For top 10% students the pattern of lower ACT/SAT scores but steady or slightly better academic performance continues.

TABLE VII (a)
FIRST YEAR ACADEMIC PERFORMANCE OF UT-AUSTIN ENTERING FRESHMEN
2000

	Top 10%			Non-Top 10%
	N	SAT	GPA	N	SAT	GPA
Architecture	24	1324	3.61	16	1381	3.17
Business	901	1214	3.27	87	1225	3.03
Communication	290	1172	3.32	85	1256	3.24
Education	69	1074	3.18	126	1121	2.86
Engineering	600	1250	3.23	351	1286	2.91
Fine Arts	87	1191	3.32	159	1162	2.80
Liberal Arts	554	1221	3.29	1968	1164	2.86
Natural Sciences	776	1208	3.23	879	1225	2.78
Nursing	41	1063	3.06	26	1150	2.60
Social Work	4	1055	2.91	16	1183	2.87
TOTAL	3346	1212	3.26	3713	1193	2.86

PERFORMANCE BY GENDER

Using 1996 as a base year, Table VIII shows that top 10% male students’ GPA increased slightly from 1996 to 1998 (from 3.19 to 3.24) and held steady through 2000 (3.22). During the same time period female mean GPAs ranged from 3.22 to 3.29. Both males and females showed declines in mean ACT/SAT scores. Interestingly, in all but one ACT/SAT score interval both males and females in 2000 matched or out-performed their 1996 cohorts.

TABLE VIII
FRESHMAN YEAR PERFORMANCE OF TOP 10% STUDENTS
BY SAT SCORE RANGE AND GENDER
1996 VS. 1998 VS. 2000

SAT RANGE (ACT Value)	1996						1998						2000
SAT RANGE (ACT Value)	MALE			FEMALE			MALE			FEMALE			MALE			FEMALE
	N	%	GPA	N	%	GPA	N	%	GPA	N	%	GPA	N	%	GPA	N	%	GPA
<900 (<18)	9	1	2.18	17	1	2.59	14	1	2.67	28	2	2.38	38	3	2.69	88	5	2.66
900-990 (18-21)	22	2	2.61	59	4	2.68	30	3	2.55	100	7	2.75	56	4	2.69	152	8	2.90
1000-1090 (21-23)	97	9	2.77	223	17	2.89	95	9	2.82	230	16	2.94	147	10	2.83	320	17	3.01
1100-1190 (23-26)	167	15	2.90	345	26	3.11	184	17	2.91	327	23	3.06	268	19	2.96	431	23	3.22
1200-1290 (26-28)	263	24	3.14	322	24	3.33	251	24	3.16	364	25	3.34	337	24	3.20	416	22	3.40
1300-1390 (29-31)	276	25	3.33	234	17	3.49	264	25	3.45	244	17	3.49	308	21	3.45	308	16	3.59
1400-1490 (31-33)	178	16	3.50	118	9	3.70	168	16	3.63	129	9	3.71	204	14	3.61	164	9	3.75
1500+ (34-36)	71	7	3.58	27	2	3.84	61	6	3.69	24	2	3.86	75	5	3.73	34	2	3.85
Total/ Mean GPA	1083		3.19	1345		3.24	1067		3.24	1446		3.22	1433		3.22	1913		3.29
Mean Test Score	1281			1210			1270			1200			1248			1185

Table VIII (a) shows similar patterns for non-top 10% students, but the ACT/SAT means are much more consistent.

TABLE VIII (a)
FRESHMAN YEAR PERFORMANCE OF “NON-TOP 10% “ STUDENTS
BY SAT SCORE RANGE AND GENDER
1996 VS. 1998 VS. 2000

SAT RANGE	1996						1998						2000
SAT RANGE	MALE			FEMALE			MALE			FEMALE			MALE			FEMALE
	N	%	GPA	N	%	GPA	N	%	GPA	N	%	GPA	N	%	GPA	N	%	GPA
<900 (<18)	25	1	2.12	46	3	2.35	39	2	2.30	52	3	2.34	36	2	2.38	38	2	2.48
900-990 (18-21)	67	4	2.36	97	6	2.50	64	3	2.33	156	9	2.42	59	3	2.68	98	6	2.57
1000-1090 (21-23)	212	11	2.36	334	22	2.62	242	13	2.53	366	21	2.69	228	11	2.61	369	21	2.77
1100-1190 (23-26)	491	27	2.49	491	32	2.75	459	24	2.53	545	32	2.82	483	24	2.64	554	32	2.88
1200-1290 (26-28)	591	32	2.55	374	25	2.88	548	29	2.64	392	23	2.97	616	31	2.77	425	25	3.06
1300-1390 (29-31)	319	17	2.76	138	9	2.91	380	20	2.75	158	9	3.11	387	20	2.98	177	10	3.28
1400-1490 (31-33)	130	7	3.00	31	2	3.28	126	7	3.10	43	3	3.33	131	7	3.21	59	3	3.40
1500+ (34-36)	23	1	2.96	6	<1	3.39	20	<1	2.97	7	<1	3.82	43	2	3.04	10	<1	3.47
Total/ Mean GPA	1858		2.57	1517		2.75	1878		2.64	1719		2.82	1983		2.79	1730		2.94
Mean Test Score	1213			1152			1212			1145			1219			1163

On the issue of rigor and selectivity (illustrated racially and ethnically in Table VII above), Table IX presents the same data by gender. Table IX shows very stable GPAs from 1996 to 1998 to 2000. Again, this was during a period of declining ACT/SAT scores.

TABLE IX
SAT MEAN AND GPA FOR TOP 10% STUDENTS
BY SELECTED UNDERGRADUATE COLLEGES AND GENDER
1996

	Male			Female
	N	GPA	SAT	N	GPA	SAT
Business	229	3.27	1276	266	3.41	1231
Engineering	277	3.29	1312	97	3.24	1275
Natural Science	287	3.10	1261	326	3.17	1200

1998

	Male			Female
	N	GPA	SAT	N	GPA	SAT
Business	258	3.26	1259	367	3.28	1213
Engineering	347	3.26	1298	150	3.27	1271
Natural Science	257	3.15	1250	364	3.18	1188

2000

	Male			Female
	N	GPA	SAT	N	GPA	SAT
Business	389	3.25	1239	512	3.29	1194
Engineering	433	3.21	1257	167	3.27	1232
Natural Science	320	3.20	1258	456	3.26	1173

PERSISTENCE

One of the concerns expressed during emotional debates over HB 588, which in some quarters continues to this day, was the concern over whether large numbers of students were encouraged to reach beyond their academic capacity and apply to UT Austin. This often led to dire predictions of retention problems. In a survey conducted last year of enrolled UT freshmen, a surprising number of students expressed the concern that HB 588 allowed for the admission of many students who were otherwise unqualified to attend UT. At the University of Texas one-year persistence is defined as the percent of students who returned for their sophomore year. And of course, two-year persistence is the percent of students enrolling the following fall.

TABLE X
PERSISTENCE/GRADUATION RATES OF TOP 10% STUDENTS
1996-2000

Freshman Class

One-Year Persistence

Two-Year Persistence

Three-Year Persistence

Four-Year Persistence

Five-Year Persistence

Sum/Fall 96

Enrolled

2428

100.00

2268

93.41

2178

89.70

2060

84.84

1078

44.40

282

11.61

Graduated

0.04

2.27

973

40.07

1698

69.93

Sum/Fall 97

Enrolled

2332

100.00

2218

95.11

2132

91.42

2008

86.11

1050

45.03

Graduated

0.09

2.49

935

40.09

Sum/Fall 98

Enrolled

2513

100.00

2387

94.99

2293

91.25

2130

84.76

Graduated

0.08

1.87

Sum/Fall 99

Enrolled

2925

100.00

2774

94.84

2639

90.22

Graduated

Sum/Fall 00

Enrolled

3346

100.00

3149

94.11

Graduated

PERSISTENCE/GRADUATION RATES OF NON-TOP 10% STUDENTS
1996-2000

Freshman Class

One-Year Persistence

Two-Year Persistence

Three-Year Persistence

Four-Year Persistence

Five-Year Persistence

Sum/Fall 96

Enrolled

3375

100.00

3061

90.70

2829

83.82

2641

78.25

1677

49.69

560

16.59

Graduated

0.03

1.24

847

25.10

1813

53.72

Sum/Fall 97

Enrolled

4033

100.00

3616

89.66

3303

81.90

3061

75.90

1951

48.38

Graduated

0.12

1.21

950

23.56

Sum/Fall 98

Enrolled

3597

100.00

3308

91.97

3073

85.43

2795

77.70

Graduated

0.08

1.42

Sum/Fall 99

Enrolled

3596

100.00

3267

90.85

2942

81.81

Graduated

0.14

Sum/Fall 00

Enrolled

3713

100.00

3382

91.09

Graduated

For top 10% students, one-year persistence slightly surpassed the 1996 level that utilized a classic admissions model relying heavily on SAT scores, high school-generated data, and including affirmative action. It is more accurate to state that persistence rates remained stable.