I need the answer only discuss in blackboard. it should be one page. no need to write as essay , no need for introduction or conclusion. I want to show as my opinion - use words like this. i believe. i think answer all the parts of the question use t

90

Journal of Hispanic Higher Education

Volume 8 Number 1 January 2009 90-109

© 2009 SAGE Publications 10.1177/1538192708326996

http://jhh.sagepub.com hosted at

http://online.sagepub.com

Latino Immigrant Access to Higher Education in a Bipolar Context of Reception Stella M. Flores Vanderbilt University, Nashville, Tennessee

Jorge Chapa University of Illinois at Urbana Champaign

Abstract: This study evaluates the recent political context in which foreign-born nonci- tizen immigrants in the United States exist, their traditional and new settlement loca- tions of residence, and where they are most likely to use public policies that encourage college enrollment. Legislative trends indicate that state activity continues to evolve around the issue of college access for undocumented immigrants, whereas U.S. con- gressional activity on the federal version of this educational legislation and general immigration policy remains unresolved.

Resumen: Este estudio evalúa el contexto político reciente en el que inmigrantes extranjeros no-ciudadanos se encuentran en los Estados Unidos de América; sus arreg- los tradicionales y nuevos lugares de residencia, y donde existen más posibilidades de utilizar políticas públicas que apoyan inscripciones universitarias. Tendencias legislati- vas indican que la actividad estatal continúa evolucionando alrededor del problema de acceso universitario para inmigrantes sin documentos, mientras que la actividad del congreso estadounidense en la versión federal de esta legislación educacional y política inmigratoria general se mantiene sin resolución.

Keywords: immigrant students; immigration; state policy; financial aid; college access; Latino students

bipolar, adj., having two diametrically opposed natures or views.

Merriam-Webster’s Online Dictionary, 2008

Foreword

In these times of heightened political attention on post-9/11 border security in con- cert with one active segment of the public supporting anti-immigration legislation, Professors Stella M. Flores and Jorge Chapa bring to the table a statistical analysis of

at GEORGIAN COURT UNIV on March 30, 2015jhh.sagepub.comDownloaded from

Flores, Chapa / Latino Immigrant Access to Higher Education 91

access for Latino immigrants to higher education. Recognizing the current political cli- mate, Flores and Chapa explain the bipolarity of policies and laws (many at cross pur- poses) with the Latino immigrant left with few choices if any for a higher education.

I applaud this direct approach to addressing access, a crucial element to educat- ing the fastest growing segment of this nation’s population. By limiting access to higher education or denying it altogether, states across this country are making a significant situation even more dire—they are effectively supporting their own pro- portional decline in knowledge capital. Flores and Chapa look at regions of the country, comparing states with and without in-state tuition policies for Latino immi- grant students. Leverage in the form of “dream” acts, accepted in some states, proves its value in increased access and increased degree attainment by the foreign-born noncitizen Latino student.

Loui Olivas President, AAHHE

Introduction

Immigration is again at the forefront of the nation’s attention, and most of the vociferously expressed sentiments have been against it. The number, variety, and vis- ibility of immigrant groups in general and immigrant Latinos in particular have been increasing. The proportion of the U.S. population consisting of immigrants is now approaching the historically high level of 15% that the United States experienced as the wave of European immigrants crested around the beginning of the 20th century. There is a major difference between the new immigration and the old. Now, most of the immigrants are from Latin America rather than Europe. It is also worthwhile to note that Mexico is the largest single country of origin for the U.S. foreign-born pop- ulation. Most Mexican immigrants to the United States are undocumented, and more than half of undocumented immigrants are from Mexico (Passel, 2005b). For many years now, U.S. policy toward undocumented immigrants could be characterized as bipolar: It has had aspects that simultaneously help and hinder their life chances. The bipolar aspect of contemporary immigration policy can be traced to the Immigration Reform and Control Act (IRCA) of 1986. This law was intended to stop undocu- mented immigrants by eliminating the employment opportunities that attracted them. Under IRCA, employers would be punished or sanctioned for hiring undocumented

Authors’ Note: This scholarly article was commissioned for the 3rd Annual Conference of the American Association of Hispanics in Higher Education, Miami, FL, March 5-8, 2008. The authors would like to thank Jacob Thornton for his assistance on the geographic information system analysis. Please address cor- respondence to Stella Flores, Department of Leadership, Policy and Organizations, Vanderbilt University, Peabody #514, 230 Appleton Place, Nashville, TN 37203; e-mail: [email protected].

at GEORGIAN COURT UNIV on March 30, 2015jhh.sagepub.comDownloaded from

92 Journal of Hispanic Higher Education

workers. However, enforcement of employer sanctions has at best “been at a token level” (Cornelius, 2004). Until recently, employer sanctions have occurred at such extremely low levels that enforcement has often not been taken as a real threat.

IRCA was positive from the perspective of the undocumented immigrants in the United States at the time. Almost 3 million were ultimately able to regularize their status in the United States under IRCA’s amnesty provisions. This may have been one of the factors that led to the virulently anti-Latino politics of California Governor Pete Wilson’s second term from 1994 to 1998. Wilson’s election to a second term was enabled by Proposition 187, which barred undocumented immi- grants from using all state services except emergency health care. It also empowered public employees to police the use of these services and turn in suspected undocu- mented immigrants. Proposition 187 became law by popular vote but was eventually overturned by the courts. However, Proposition187 was followed by a wave of anti- immigrant laws as noted in the following:

Following passage of California’s Proposition 187, many states, including Florida, Arizona, and Texas, considered introducing similar legislation. But passage of the fed- eral Personal Responsibility and Work Opportunity Act of 1996 ended the need for states to establish individual policies. . . . The bill duplicated many of the provisions of Proposition 187. (Ono & Sloop, 2002, pp. 4-5)

The 1996 federal Illegal Immigration Reform and Immigrant Responsibility Act (IIRIRA) was also deeply influenced by the anti-immigrant politics of Proposition 187 (see Tichenor, 2002, chap. 9). Section 505 of IIRIRA is of particular importance to this article’s analysis regarding the college access opportunities of undocumented students, a majority of whom are of Latino origin. This section specifies that unau- thorized aliens

shall not be eligible on the basis of residence within a State (or a political subdivision) for any postsecondary education benefit unless a citizen or national of the United States is eligible for such a benefit (in no less an amount, duration, and scope) without regard to whether the citizen or national is such a resident. (Feder, 2006)

Some members of Congress attempted to negate Section 505 with the introduc- tion of the Student Adjustment Act in 2001. It did not pass but led to the introduc- tion of the DREAM Act (Development, Relief and Education for Alien Minors Act), which has been introduced in the Senate every year since 2003. Although the federal DREAM Act remains to be passed, state legislatures have found a way to temporar- ily mediate access barriers to higher education in the form of in-state resident tuition policies, also known as state “dream” acts (Flores, 2007). The state dream acts were passed with the goal of making undocumented immigrants eligible to pay in-state resident tuition for public colleges and universities without violating Section 505.

at GEORGIAN COURT UNIV on March 30, 2015jhh.sagepub.comDownloaded from

This study assesses the recent political context in which undocumented immigrants in the United States find themselves, their locations of residence, and estimations of where they are most likely to use public policies that encourage college enrollment. Regardless of location implementation, legislative trends to date indicate that state activity continues to evolve around the issue of access to higher education for undoc- umented immigrants, whereas U.S. congressional activity on the federal version of this educational legislation and general immigration policy remains unresolved.

Policy Context at the Turn of the Century

Latino Immigration After September 11, 2001

Despite the anti-immigrant legislation, the population of undocumented Latino immigrants grew throughout the 1990s, especially during the economic boom between 1997 and 2000. The midpoint of one set of estimates of the undocumented Mexican population in the United States as of mid-2001 was 4.5 million. The median of the esti- mates of the undocumented Central American immigrants was an additional 1.5 mil- lion (Bean, Van Hook, & Woodrow-Latfield, 2002). One of the top policy priorities of the then recently elected presidents of the United States and of Mexico was the reso- lution of some of the problems associated with this large-scale undocumented migra- tion. During the first week of September 2001, Mexican President Vicente Fox was the first foreign head of state to have an official state visit with U.S. President George W. Bush. Fox had a highly visible, successful trip to Washington that included White House meetings and an address to Congress. Bush and Fox were jointly developing an initiative that would either “legalize” the undocumented immigrants, that is, give them immigrant visas, or “regularize” their status through a formal, de jure guest worker program. The September 11, 2001, attacks occurred just days after Fox’s visit, and they completely halted any movement toward either regularizing or legalizing undocu- mented immigrants. As a result of these attacks, U.S. borders have become more closed and difficult to cross, and security across the country has become extremely keen. In his comprehensive analysis of U.S. immigration policies and politics, Dividing Lines, Tichenor (2002) argues that episodic international crises or threats serve as important catalysts for major immigration reform. In the United States, since 9/11, it appears that the threat of further terrorist attack has been used to stymie the previously conceptualized reform of undocumented immigration from Mexico.1

State Dream Acts and Undocumented Immigration Settlement

In 2001, Texas passed the first in-state resident tuition legislation targeted at undocumented high school graduates. The discount to attend public colleges and universities at an in-state resident price is particularly significant to undocumented

Flores, Chapa / Latino Immigrant Access to Higher Education 93

at GEORGIAN COURT UNIV on March 30, 2015jhh.sagepub.comDownloaded from

94 Journal of Hispanic Higher Education

students because they do not qualify for federal aid to finance a postsecondary edu- cation. Moreover, this group of students, particularly immigrants from Latin America and the Caribbean, are more likely than the general population to live in poverty (Erisman & Looney, 2007), making paying for a higher education nearly impossible without financial assistance. A typical discount offered per academic year by the in-state resident tuition policy in Texas is $2,000 at the public commu- nity college level and almost $8,000 at the 4-year public college level (McGee. P. 2005, July 24.)

In 2003, a federal version of the in-state resident tuition policy was introduced in the U.S. Congress. Called the DREAM Act, the act would allow, in addition to an in-state resident tuition benefit, certain undocumented students to start on the path toward citi- zenship if they went to college or served in the U.S. military (National Immigration Law Center, 2006; Olivas, 2004). Although the legislation has failed to pass over a number of congressional sessions, the dream terminology used with the original federal version of the proposed act has since been adopted by researchers, advocacy groups, and media outlets across the country to describe the multiple in-state resident tuition policies (Immigrant Legal Resource Center, 2008; Rincon, 2005). Since 2001, nine other states in addition to Texas have passed similar versions of dream act legislation, including California, Illinois, Kansas, Nebraska, New Mexico, New York, Oklahoma, Utah, and Washington. Although much more limited in scope and level of benefits than the pro- posed federal legislation, the state dream acts have not been free from controversy (Olivas, 2004). Of the states that have a tuition policy, Kansas and California have been challenged (the Kansas case unsuccessfully) in federal court and Oklahoma recently repealed its in-state resident tuition benefit in 2007. More recently, states including Arizona and Virginia have passed legislation to ban the tuition benefit for undocu- mented immigrants (Redden, 2007). A particularly interesting phenomenon of state activity regarding these dream acts is that the location of where they are passed and implemented is not particularly predictable. Whereas the adoption of the tuition policies in the traditional migration settlement states of California, Texas, New York, and Illinois may not be surprising, the institution of these policies in Kansas, Nebraska, Utah, and Washington seem less so at first glance.

However, migration activity to different regions of the United States continues to disperse to nontraditional migration states adding some insight as to why states that have not historically experienced continuous waves of migration might have consid- ered such legislation. Increasing numbers of cohorts of immigrant children continue to enter the K-12 system in states with and without a tuition policy. Some state sys- tems are familiar with this migration, whereas others are not. Ironically, all in-state resident tuition policies, with the exception of the Texas policy, passed state legisla- tures after September 11, 2001. However, “counter movements” to this activity in the form of legislative bans to any form of state incentive to enroll undocumented immi- grant students who have completed their high school diploma in the United States were also initiated shortly after this time period, often in areas where Latino popu- lations, particularly of Mexican origin, are new.

at GEORGIAN COURT UNIV on March 30, 2015jhh.sagepub.comDownloaded from

To assess the status of state and federal legislation regarding immigrant student access to higher education, we provide the following analyses. First, we assess whether there are any demographic differences by age, gender, marital status, employment participation, and educational completion rates among U.S. citizen Latino and foreign-born noncitizen (FBNC) Latinos in three distinct migration set- tlement areas before and after September 11, 2001. Second, we provide a visual rep- resentation of the states with and without in-state resident tuition policies in relation to where all Latino-origin individuals and all FBNC Latinos reside as a percentage of the U.S. population. We provide this visual representation to situate the educa- tional prospects for undocumented Latino students in the United States. We then assess whether Latino students likely to be undocumented were more likely to enroll in college in some migration settlement regions over others. We end with a reevalu- ation of the issue of undocumented student access to higher education in relation to U.S. citizen Latino student access to higher education. Table 1 displays the states with an in-state resident tuition policy as of 2006, whether the legislation offers addi- tional state financial aid assistance, and the region of the United States in which they are located according to migration pattern status.

Empirical Assessment of State Tuition Policies Under Current Federal Immigration Policies

The Data

For this analysis, we incorporate two versions of the current population survey (CPS), a nationally representative sample sponsored by the U.S. Census Bureau and the U.S. Bureau of Labor Statistics for the years 1998 to 2005 to assess the college enrollment rates of Latino FBNCs across different geographic areas of the United States. We also incorporate data from the 2006 American Community Survey to visually assess the location of FBNC and citizen Latinos in the United States. The primary data set, the Merged Outgoing Rotation Group (MORG) files, have a more desirable sample size in that they have approximately 30,000 individuals nationally per monthly extract to answer how the in-state resident tuition policies affected college enrollment.2

To display the effect of particular college access policies on the Latino immigrant and citizen population we use two methods. First, we use geographic information system mapping to visually capture the location of these groups in relation to where relevant public policies exist. Second, we use logistic regression to estimate the impact of the tuition policies on students likely to be undocumented immigrants across different immigrant settlement regions across the country. A description of the motivation for each regional analysis is included below.

Flores, Chapa / Latino Immigrant Access to Higher Education 95

at GEORGIAN COURT UNIV on March 30, 2015jhh.sagepub.comDownloaded from

96 Journal of Hispanic Higher Education

Sample

The sample for this analysis includes any individual who has self-identified as Hispanic in the survey, has listed a Latin American country of national origin, and has at least one parent who is foreign-born and has listed a Latin American country of origin. The sample is limited according to the requirements of the in-state resident tuition policy of each state, such as required years of residency and high school/GED completion status. For example, in New York this may be 2 years, whereas in California the residency requirement is 3 years, meaning that the student would have had to enter this state by approximately 1998 as the policy passed in 2001. Table 2 provides a description of all Latin American origin respondents in the sample by geographic region and migration period of interest for this analysis by select demographic characteristics.

Table 1 States That Allow Undocumented Students to Gain Resident Tuition Status as of 2006

State Financial Aid for

Undocumented State Bill Number Date Passed Students Regional Area

Texas H.B. 1403 June 16, 2001 Yes Traditional settlement (77th Leg.)

California A.B. 540 (2001-02 October 12, 2001 No Traditional settlement Cal. Sess)

Utah H.B. 144 (54th Leg., March 6, 2002 Partial New non-Southern Gen Session)

New York S.B. 7784 (225th Leg., June 25, 2002 No Traditional 2001 NY Sess)

Washington H.B. 1079 (H.B. 1079, May 7, 2003 No New non-Southern 58th Leg. Reg. Sess)

Oklahoma S.B. 596 (49th Leg, May 12, 2003 Yes New non-Southern 1st Sess)

Illinois H.B. 60 (93rd Leg. May 18, 2003 No Traditional settlement Reg. Sess.)

Kansas K.S.A. 76-731A May 20, 2004 No New non-Southern New Mexico N.M.S.A. 1978 April 5, 2005 No Traditional settlement

Ch. 348, Sec. 21-1-1.2 (47th Leg. Sess.)

Nebraska L.B. 239 (99th Leg. April 2006 No New non-Southern 2nd Sess.)

NOTE: Texas amended the statute in their 2005 legislative session. Source: Some information from Olivas (2004) and Rincon (2005).

at GEORGIAN COURT UNIV on March 30, 2015jhh.sagepub.comDownloaded from

97

T ab

le 2

S u

m m

ar y

S ta

ti st

ic s

B ef

or e

an d

A ft

er S

ep te

m b

er 1

1, 20

01 a

U .

S .

C it

iz en

L at

in os

F or

ei gn

-B or

n N

on ci

ti ze

n L

at in

os

T ra

di ti

on al

N

ew N

on -S

ou th

er n

S ou

th er

n T

ra di

ti on

al

N ew

N on

-S ou

th er

n S

ou th

er n

P re

- P

os t-

P re

- P

os t-

P re

- P

os t-

P re

- P

os t-

P re

- P

os t-

P re

- P

os t-

9/ 11

( 1)

9/ 11

( 2)

9/ 11

( 3)

9/ 11

( 4)

9/ 11

( 5)

9/ 11

( 6)

9/ 11

( 7)

9/ 11

( 8)

9/ 11

( 9)

9/ 11

( 10

) 9/

11 (

11 )

9/ 11

( 12

)

A ge

20 .8

8 20

.8 7

20 .9

5 21

.0 4

21 .0

1 21

.0 6

21 .1

7 21

.2 2

21 .0

2 21

.1 8

21 .0

2 21

.2 4

(. 21

34 )

(0 .0

20 8)

(0 .0

34 4)

(0 .0

40 3)

(0 .0

35 3)

(0 .0

43 0)

(0 .0

25 6)

(0 .0

26 1)

(0 .0

34 1)

(0 .0

40 6)

(0 .0

35 4)

(0 .0

39 8)

F em

al e

0. 50

74 0.

51 14

0. 47

52 0.

48 96

0. 47

56 0.

48 48

0. 45

30 0.

45 01

0. 46

29 0.

45 08

0. 46

42 0.

44 70

M ar

ri ed

0. 21

54 0.

18 09

0. 24

80 0.

25 08

0. 24

58 0.

25 23

0. 27

97 0.

28 66

0. 26

14 0.

27 17

0. 24

96 0.

27 87

E m

pl oy

ed 0.

65 27

0. 61

97

0. 66

93 0.

68 12

0. 66

08 0.

67 09

0. 65

70 0.

64 97

0. 65

31 0.

68 91

0. 66

18 0.

69 74

(. 00

51 )

(0 .0

04 9)

(0 .0

07 8)

(0 .0

09 2)

(0 .0

08 3)

(0 .0

10 4)

(0 .0

05 9)

(0 .0

06 2)

(0 .0

08 0)

(0 .0

09 3)

(0 .0

08 3)

(0 .0

09 5)

L iv

e in

m et

ro a

re a

0. 91

38 0.

91 91

0. 87

47 0.

82 05

0. 89

88 0.

86 33

0. 93

92 0.

94 03

0. 88

50 0.

84 19

0. 89

51 0.

84 36

P er

ce nt

ag e

hi gh

s ch

oo l

0. 63

06 0.

67 60

0. 56

95 0.

59 91

0. 56

12 0.

58 73

0. 47

91 0.

47 18

0. 52

49 0.

51 21

0. 54

15 0.

51 38

di pl

om a

an d

no B

A P

er ce

nt ag

e w

it h

0. 03

28 0.

03 47

0. 02

70 0.

03 41

0. 02

67 0.

04 26

0. 01

89 0.

02 50

0. 02

40 0.

02 76

0. 02

44 0.

03 01

B A

o r

hi gh

er O

bs er

va ti

on s

11 ,2

80 12

,3 97

4, 04

0 2,

98 6

3, 66

5 2,

37 0

8, 07

2 7,

59 6

3, 87

1 2,

75 7

3, 69

0 2,

75 6

N O

T E

: D

at a

in t

hi s

sa m

pl e

in cl

ud e

in di

vi du

al s

ag ed

1 8

to 2

4 w

ho e

nt er

ed t

he U

ni te

d S

ta te

s be

fo re

1 99

8 fo

r al

l ed

uc at

io na

l co

m pl

et io

n ra

te s.

R ob

us t

st an

- da

rd e

rr or

s ar

e gi

ve n

in p

ar en

th es

es .

S ou

rc e:

U .S

. C

ur re

nt P

op ul

at io

n S

ur ve

y, m

er ge

d ou

tg oi

ng r

ot at

io n

gr ou

ps (

au th

or s’

ca lc

ul at

io ns

). a.

S am

pl e:

A ll

L at

in o

ci ti

ze n

an d

fo re

ig n-

bo rn

n on

ci ti

ze ns

, ag

ed 1

8 to

2 4.

at GEORGIAN COURT UNIV on March 30, 2015jhh.sagepub.comDownloaded from

Data Considerations and Limitations

Individual and group-level data on undocumented immigrants in the United States is an unresolved data conflict. No government agency in the United States directly counts the undocumented immigrant population, leading to some uncer- tainty in capturing their exact numerical presence (Passel, 2005a; Passel, Van Hook, & Bean, 2004). However, data from the U.S. Census Bureau and Department of Labor Statistics offer some of the closest level of detail on citizenship status cur- rently available. These data are used as a principal source of information for calcu- lating estimates of the unauthorized population in the United States and are used in these analyses (U.S. Department of Labor, 2002). Using the FBNC category of the CPS, which includes both the undocumented and legal permanent residents and not naturalized citizens, allows us to more accurately estimate the effect on individuals likely to be undocumented rather than applying these conclusions to all individuals who are foreign-born but residing in the United States.

Analytic Strategy

To estimate the impact of a financial aid policy targeted at undocumented students on their college enrollment by immigrant settlement regions in the United States, we use the passing of each in-state resident tuition policy as a source of exogenous varia- tion by employing a differences-in-differences strategy (Dynarski, 2003, 2004; Kane, 1994, 2003; Long, 2004). That is, we compare college enrollment rates of students likely to be undocumented after the implementation of a tuition policy with the college enrollment rates of a similar cross-section of students also likely to be undocumented before the policy implementation in the same state. We further apply this “before-and- after” strategy to a set of comparison states, identified as control group states with sim- ilar demographic and historical migration pattern characteristics, that have not passed an in-state resident tuition policy. Previous work has suggested that Latino FBNC students living in states with a tuition policy are more likely to enroll in college than similar students living in states without such legislation (Flores, 2007). We hypothe- size that among states with an in-state resident tuition policy, those with longer histories of migration settlement patterns (traditional locations) such as California and Texas are more likely to have higher college enrollment rates for students likely to be undocumented than states with more recent migration patterns.

Rationale for Treatment and Control Group Selection

From 1971 to the early 1990s, almost half of all immigrants settled in the top five urban areas and six particular states (California, Florida, Illinois, New Jersey, New York, and Texas; Massey & Capoferro, 2008). This geographic concentration pattern typical of the last quarter of the 20th century was disrupted by a new deconcentration

98 Journal of Hispanic Higher Education

at GEORGIAN COURT UNIV on March 30, 2015jhh.sagepub.comDownloaded from

and dispersion of primarily Latin American populations to new regions of the coun- try for a variety of reasons still under debate (Massey & Capoferro, 2008). These included areas in the Northeast such as Pennsylvania and some areas of the Midwest that had not seen significant immigrant incorporation since earlier European waves and the sudden appearance of mostly Latino-origin migration in the South, as noted by various scholars since the turn of the century (Massey & Capoferro, 2008; Millard & Chapa, 2004). We use developing demographic trends and a geographic assessment of where Latinos, both FBNC and citizen above the age of 18, were located in 2006 to estimate the impact of the most significant college access policy to benefit undoc- umented immigrants, the in-state resident tuition policies in 10 states across the nation. To evaluate the potential impact of the policies by different regional settlement patterns, we divide the treatment and control groups into three different settlement regions: traditional, new non-Southern, and Southern destinations. Treatment states are states that have adopted an in-state resident tuition policy, whereas control states are locations with similar demographic and historical settlement pattern characteris- tics as the treatment states without an in-state resident tuition policy benefiting undoc- umented immigrant students.

Within each settlement region, we test whether students who live in states that have implemented an in-state resident tuition policy have higher college enrollment rates than students living in states in the same settlement region without the presence of an in-state resident tuition policy. For example, in the traditional settlement region, we compare the treatment states of California, Illinois, New Mexico, New York, and Texas with the control states of Arizona, Colorado, Florida, Massachusetts, Nevada, and New Jersey. The new non-Southern destination treatment states include Kansas, Oklahoma, Utah, and Washington and are compared with similarly located control states of Idaho, Indiana, Minnesota, Nebraska, and Oregon.3 No Southern state in the nation has adopted an in-state resident tuition policy. To evaluate the differences between traditional settlement versus new Southern settlement regions, we estimate the college enrollment effects in treatment traditional settlement regions (California, Illinois, New Mexico, New York, Texas) compared with all Southern states that have experienced recent high rates of undocumented immigrant settlement, including Alabama, Arkansas, Georgia, Kentucky, Mississippi, North Carolina, South Carolina, Tennessee, and Virginia. Although this third comparison is not in terms of demo- graphic and historical migration settlement patterns, the comparison is useful in deci- phering potential educational, time-related, and labor market characteristic differences among the oldest and newest destination choices for Latino immigrants. The logistic regression analysis answers whether undocumented students residing in states with a tuition policy in a particular settlement region of the United States experienced increased college enrollment rates compared with similar students living in states without a tuition policy. A detailed explanation of the logistic regression model with associated variables is provided in the appendix.

Flores, Chapa / Latino Immigrant Access to Higher Education 99

at GEORGIAN COURT UNIV on March 30, 2015jhh.sagepub.comDownloaded from

Results

Individual Characteristics Pre- and Post-September 11, 2001

As the late 1990s marked an unforeseen and new dispersion of immigrant groups across the United States, whether by increased internal migration from FBNC resi- dents or altogether new crossings of similar individuals in the United States, the events of 2001 instituted another set of contextual policies, rules, and reactions to the migra- tion landscape. Table 2 shows summary statistics for citizen versus FBNC Latinos in the set of years preceding and immediately after September 2001 by the three identi- fied relevant settlement destinations: traditional, new non-Southern, and Southern regions. The characteristics measured are age, female status, whether an individual has ever been married, employment, residence in a metro area, percentage with a high school diploma and no BA degree, and percentage with a BA degree or higher.

Table 2 displays particularly interesting characteristics in regard to employment, residence in a metro area, and level of educational attainment. In regard to age, female status, and marital status, there are few differences between the citizen groups (FBNC versus citizen Latinos) and by regions, with two exceptions. The pop- ulation of U.S. citizen Latinos is on average more likely to be female, whereas the opposite is true for FBNC Latinos across the three different settlement regions exam- ined. Although the differences are minimal, the pattern of a male dominance in migration to the United States appears to hold in a post-9/11 context. Marital status, measured by whether an individual has ever been married, increased after September 2001 in all regions and across citizenship groups with the exception of U.S. citizen Latinos in traditional migration states. Among this group, U.S. citizen Latinos were less likely to have ever been married after September 2001 than before this date. Whether an individual was employed at the time of the survey and resided in a metro area yields differential results across the regions and citizenship groups. Individuals residing in the traditional migration states, regardless of citizenship status, were less likely to be employed after September 2001 compared with individuals residing in the new non-Southern and Southern settlement regions. In these latter two destina- tions, individuals were more likely to indicate they were employed after September 2001 than before this date, especially among the FBNC groups not living in a tradi- tional settlement region. Finally, the likelihood of living in a metropolitan area appears to be significantly lower in all regions across citizen groups, with the excep- tion of the traditional settlement regions. Thus, only in the traditional migration des- tinations were individuals aged 18 to 24 less likely to indicate that they were employed and indicate that they were living in a metropolitan area. In terms of edu- cational attainment as measured by completion of a high school diploma but no BA degree, Latino U.S. citizens saw an increase in the percentage of individuals who had completed a high school diploma across all three different regions. In contrast, the percentage of Latino FBNCs with a high school degree was actually lower after

100 Journal of Hispanic Higher Education

at GEORGIAN COURT UNIV on March 30, 2015jhh.sagepub.comDownloaded from

September 2001. Interestingly, the percentage of Latinos of all citizenship categories with a BA degree or higher, after September 2001, was higher among all regional groups examined, with the highest increase in the new non-Southern region for Latino U.S. citizens and almost equal proportions of increase in the traditional and Southern destinations among Latino FBNCs.

Geographic Analysis

Recent research on the effect of in-state resident tuition policies indicates that Latino FBNCs are more likely to enroll in college after the implementation of an in- state resident tuition policy than similar students in states without a tuition policy (Flores, 2007). However, it is unclear whether particular regions with tuition policy states experience a stronger college enrollment impact than other areas of the United States with distinct migration patterns and residence duration. In deciphering these regional puzzles, we therefore ask where Latinos reside according to citizenship status and whether they are located in areas with an in-state resident tuition policy?

Figures 1A and 1B display two maps of the United States Latino population by cit- izenship status as of 2006. States labeled in red represent a location with an in-state resident tuition policy. States labeled in blue are states without an in-state resident tuition policy. The map on the left represents the percentage of the citizen Latino pop- ulation as a proportion of the total U.S. population, whereas the map on the right rep- resents the proportion of the Latino FBNC population, also as a percentage of the total population. States with darker shades of green represent the highest percentages of the selected Latino population, whereas the lightest shades of green represent the lowest percentage of the selected measure. Figure 1A indicates that the U.S. citizen Latino population is generally located in the states where an in-state resident tuition policy is present, with the exceptions of Arizona and Florida. That is, in regard to sheer numbers of Latinos, almost all the traditional migration settlement states have implemented an in-state resident tuition policy signaling increased opportunity for undocumented students, a majority of whom settle in these locations. Colorado, Massachusetts, Nevada, and New Jersey are the states with the next largest share of Latinos that do not have a tuition policy (second category of states). However, a number of states in the next significant share of Latino state composition (the third category of states) do have a tuition policy. These include Illinois, Kansas, Utah, and Washington.

Figure 1B represents a different picture of access to in-state resident tuition leg- islation for the Latino FBNC population as a percentage of the total U.S. population. States with the highest percentage of Latinos likely to be undocumented are those located in the southeastern United States, none of which have adopted an in-state resident tuition policy. The category of states with the second highest percentage of this population (those likely to be undocumented), does include a number of states with a state dream act, which include Illinois, Nebraska, Oklahoma, Utah, and Washington. The most notable trend, however, is the expansion of the undocumented

Flores, Chapa / Latino Immigrant Access to Higher Education 101

at GEORGIAN COURT UNIV on March 30, 2015jhh.sagepub.comDownloaded from

102 Journal of Hispanic Higher Education

Latino population in the South, where instead of activity to adopt an in-state resident tuition policy, there has been legislative activity in the exact opposite direction to ban undocumented students from even enrolling in postsecondary institutions indepen- dent of whether they can afford the required tuition, as in Mississippi, North Carolina, Virginia, and most recently Tennessee (Redden, 2007).

Logistic Regression Analysis

Table 3 presents results measuring the impact of the in-state resident tuition poli- cies on the college enrollment of Latino FBNCs by destination region comprising (a) traditional migration states, (b) new non-Southern migration states, and (c) Southern states. The table shows the estimated odds ratios of enrolling in college and associ- ated robust standard errors for Latino FBNCs who have completed a high school diploma or GED. The data indicate that Latino foreign-born students living in dream act states in the traditional migration regions are 1.69 times more likely to enroll in college than similar students in other states that comprise part of that region. That is,

Figure 1 Two Worlds of College Access for Latinos: Are College Access Opportunities Equal Across the United States

at GEORGIAN COURT UNIV on March 30, 2015jhh.sagepub.comDownloaded from

Flores, Chapa / Latino Immigrant Access to Higher Education 103

even within the traditional migration settlement regions where educational and other social services related to immigrant incorporation have had a longer time to develop, states with a tuition policy in this region are more likely to enroll students likely to be undocumented in college. This indicates that states such as California, Illinois, New York, and Texas are successfully increasing the college enrollment rates of individuals likely to be undocumented (column 1). Column 2 examines states that

Table 3 Impact of In-State Resident Tuition Policies on College Enrollment

of Latino Foreign-Born Noncitizens by Migration Destinations, 1998-2005a

New Non- Traditional Southern Southern

Destinations, Destinations, Destinations, Full Model Full Model Full Model

CA, NM, TX, NY, CA, NM, TX, NY, IL UT, KS, WA, IL Versus AR, TN, Versus AZ, NV, CO, OK Versus ID, VA, NC, SC, GA,

FL, MA,NJ OR, NE, IN, MN AL, MS, GA, KY (1) (2) (3)

Effect of tuition policy states 1.6933*** 0.2882* 1.7933*** (0.2714) (0.1941) (0.3320)

Age 0.8551*** 0.9142** 0.8565*** (0.0090) (0.0342) (0.0102)

Female 1.5220*** 1.5275*** 1.5086*** (0.0606) (0.2212) (0.0676)

Ever married 0.2451*** 0.3736*** 0.2435*** (0.0170) (0.0736) (0.0190)

Metro 1.8271*** 1.4242** 2.1687*** (0.1842) (0.2555) (0.2386)

State unemployment rate 1.0156 0.9978 0.9437 (0.0684) (0.1522) (0.0761)

Month fixed effects 1.3751*** 1.5342 1.4397*** (0.1412) (0.5797) (0.1705)

State and year fixed effects, 1.1393 0.9292 1.0977 including FBNC interaction (0.1178) (0.3684) 0.9791

Observations 18,517 1,789 14,485

Note: States with policy intervention include: TX (2001); CA, and UT (2002); NY, WA, OK, and IL (2003); KS (2004); and NM (2005). NE, passed in 2006, is not included in this analysis. Robust standard errors were calculated to account for clustering within person over time, so that standard error estimates would reflect the structure of the data. Source: U.S. Current Population Survey, merged outgoing rotation groups (authors’ calculations.). a. Sample: Latino high school GED completers, ages 18 to 24; outcome: enrolled in college; destinations: traditional, new non-Southern, and new Southern migration states. Robust standard errors are given in parentheses. *Significant at 10%. **Significant at 5%. ***Significant at 1%.

at GEORGIAN COURT UNIV on March 30, 2015jhh.sagepub.comDownloaded from

104 Journal of Hispanic Higher Education

comprise a more recent migration settlement region labeled here as new non- Southern. Within this region alone, four states have adopted an in-state resident tuition policy (Kansas, Oklahoma, Utah, and Washington). Does the adoption of a state policy then predict a significant increase in college enrollment of Latino FBNCs in comparison to similar states without a tuition policy in this particular region? The data in column 2 indicate that there is no effect on college enrollment for individuals residing in this region of the United States. One important note is that the sample for this region is comparatively smaller than the population in the larger traditional migration settlement states. Although we control for a number of factors, including state and time indicators, the data suggest that college enrollment of undocumented students nationwide may be driven by a particular region of the coun- try with a particular immigration history and settlement period in the United States as of now. The new non-Southern states appear to be at a relative time and infra- structure disadvantage because of time and resources in relation to new immigrant populations. Finally, column 3 shows that, unsurprisingly, when compared with the Southern states, students in the traditional migration states are significantly more likely (1.79 times) to enroll in college after the adoption of an in-state resident tuition policy than similar students living in the Southern states without a tuition policy. The odds ratios for the results in columns 1 and 3 are all significant at the p < .01 level.

Discussion

Latino Access to College by Citizenship Status: Is the DREAM Act Sufficient?

Previous research and the data presented in this analysis indicate that the state dream acts are successful in increasing the college enrollment rates of Latino students likely to be undocumented, with greater success in some regions of the country than others. As a proportion of the population, we see that Latinos and most immigrants reside in the states where these tuition policies are available. However, the dispersion of the Latino immigrant (primarily undocumented) population into new settlement regions where there are no state dream acts poses new challenges to higher education and political systems that are not familiar with their presence or where they have little to no representation at the legislative or institutional level. These are but a few of the challenges looming ahead across the United States.

The larger context of Latino access to higher education, even for those with U.S. citizenship, however, is no less bleak given the privilege of citizenship. Table 4 pro- vides a 1-year snapshot of estimates of the number of U.S. high school students who graduated in 2005 by citizenship status. The dream act–eligible students should be a substantial proportion of the 124,072 FBNCs who graduated from high school. Note that 18% of all Latino high school graduates dealt with in this table are FBNCs, and

at GEORGIAN COURT UNIV on March 30, 2015jhh.sagepub.comDownloaded from

Flores, Chapa / Latino Immigrant Access to Higher Education 105

less than 5% of all high school graduates are FBNCs. Dream acts apply to a small portion of Latino and total high school graduates. In regard to college enrollment, Table 5 shows the rate or percentage of all high school graduates aged 18 to 24 who were enrolled in college in 2005. Note that the percentage of FBNC Latinos enrolled in college (59%) is higher than that of U.S.-born Latinos (50%). These estimated rates must be used with a caveat noted regarding data as we cannot directly distin- guish the documented immigrants in this group from the undocumented. Although efforts to pass the federal DREAM Act are well noted and justified given the grow- ing demographic implications of undocumented students for communities across the country, the data clearly indicate that attention to the U.S. citizen Latino population must also remain a high educational priority.

Complex portraits of the Latino community by citizenship status and national ori- gin presented signal urgent calls for additional research on these topics. For example, it would be interesting to conduct more detailed analyses of the data presented in Tables 4 and 5 over time by Latino student national origin and income status. In addition, further analyses of Latino subgroups by region of the country and time spent in the United States would add critical information on educational trajectories by state and the outcomes of local educational infrastructure for these populations.

Table 4 Noncitizen High School Graduates

2005 High School Graduates (HSGs) Total Latino Mexican

Native, Born in the United States 2,479,443 301,776 221,701 Foreign-born, not a U.S. citizen 124,072 72,016 51,897 Total 2,692,476 392,112 281,521

NOTE: 18% of Latinos HSGs are foreign-born, not a U.S. citizen. Less than 5% of all HSGs are foreign- born, not a U.S. citizen. Source: Analysis of November 2005 Current Population Survey data (authors’ calculations).

Table 5 Percentage of High School Graduates 18 to 24 Enrolled in College

2005 Total Latino Mexican

Native, born in the United States 67 50 51 Foreign-born, noncitizen 68 59 48 Total 67 53 49

NOTE: 72% of foreign-born Latinos, not a U.S. citizen are of Mexican origin; 42% foreign-born, not a U.S. citizen are of Mexican origin. Source: Analysis of November 2005 Current Population Survey data (authors’ calculations).

at GEORGIAN COURT UNIV on March 30, 2015jhh.sagepub.comDownloaded from

106 Journal of Hispanic Higher Education

Finally, more research on how labor market opportunities coincide with educational options by citizenship status and geographic region would be useful to examine a more accurate state of decision making for immigrant students to date.

Conclusion

A contemporary review of particular facets of U.S. immigration policy yields inconsistent and inconclusive data in relation to labor and educational opportunities of undocumented immigrant students, a number of whom entered the country ille- gally without consent (Olivas, 2004). The introduction of the state dream acts have created a form of educational opportunity for students who might not have been able to afford college otherwise. However, the opportunity is limited to school attendance and not the use of a postsecondary degree in the workforce until a federal version of a DREAM Act is passed in Congress. In the interim, efforts to overturn some of the 10 state dream acts have continued, with additional laws, either by legislative action or voter referenda, proposed to counteract any postsecondary assistance and even admission to undocumented high school graduates. The bipolar nature of immigra- tion policy thus continues to exist between federal and state jurisdictions as well as across state levels. The evidence shows that immigrant students, when given the opportunity, are likely to take advantage of policies that work to improve their human capital potential. The nation must now decide whether it will appropriately capitalize and build on these valuable and effective investments or waste them. In the interim, this is ultimately one of a number of multidimensional sets of policies that need to be put in place to truly bring the participation of all Latinos and Latinas in higher education up to parity with the total population.

Appendix

To answer whether students located in states with a tuition policy in a particular settle- ment region experienced increased college participation compared with similar students living in states without a tuition policy, we estimated the following model separately for each of the three settlement regions identified: traditional, new non-Southern, and Southern destinations:4

LOGISTIC (INCOLL = 1) = β0 + β1DIMMIGTUITION + β2FBNC + β3(DIMMIGTUITION × FBNC) + β4STATEDUMMIES + β5YEARDUMMIES +

β6(STATEDUMMIES × FBNC) + β7(YEARDUMMIES × FBNC) + β8X + ε, (1)

where FBNC refers to foreign-born noncitizen.

(continued)

at GEORGIAN COURT UNIV on March 30, 2015jhh.sagepub.comDownloaded from

Flores, Chapa / Latino Immigrant Access to Higher Education 107

Appendix (continued)

1. Traditional settlement destinations Treatment States: CA, NM, TX, NY, IL Control States: AZ, NV, CO, FL, MA, and NJ Sample: Latino high school graduates

2. New non-Southern settlement destinations Treatment States: UT, KS, WA, OK Control States: ID, OR, NE, IN, MN5

Sample: Latino high school graduates 3. Southern versus traditional settlement destinations

Treatment States: CA, NM, TX, NY, IL Control States: AR, TN, VA, NC, SC, GA, AL, MS, GA, KY Sample: Latino high school graduates

where INCOLL is a binary variable and a measure of 18- to 24-year-old Latino FBNCs who were enrolled in college as of the week prior to being surveyed. DIMMIGTUITION is a binary variable equal to 1 in states, months, and years with an in-state tuition policy for undoc- umented immigrants. This variable captures the policy interventions in each state by destina- tion region described above by month and year of the policy enactment date and also includes year of entry restriction for each state’s residency requirement. FBNC is a binary variable set to 1 if a Latino student is classified as a FBNC in the survey data.

β3 represents the coefficient of interest and is the interaction term of DIMMIGTUITION and FBNC. If β3 is nonzero, positive, and statistically significant, we can reject the null hypoth- esis that the tuition policy has no effect, suggesting that states in a particular destination group with a tuition policy have a positive effect on the college enrollment rates of FBNC Latino students compared with similar states in that destination group without a tuition policy. Coefficients β4 to β7 are dummy variables accounting for all states and years in each destina- tion group (traditional, new non-Southern, Southern) and act as state and year fixed effects and selected interactions. State fixed effects are included to account for intrastate variation and con- trol for the average differences in any observable or unobservable predictors not explained by other covariates, such as state unemployment rate and year-specific trends in the outcome. Year fixed effects are included to control for general trends over time in the outcome variable for all states in the sample. As immigration trends vary by state and year, we also include an interac- tion term between state and year fixed effects with FBNCs (as seen in β6 and β7).

X captures the effect of relevant demographic characteristics available in the CPS corre- lated with educational attainment as well as local economic conditions that may affect an indi- vidual’s schooling decisions (age, gender, living in a metropolitan area, and state unemployment rate). Unemployment rate, for example, is included to account for state-specific economic shocks in the various state labor markets for each of the destination groups.

Notes

1. For additional legal analysis of the context of immigration law after September 11, 2001, see Olivas, 2004.

2. Because the merged outgoing rotation group data set has multiple observations for most individuals over time, we calculate robust standard errors to account for clustering of observations at the individual level (within person) and so that standard error estimates reflect the structure of the data.

at GEORGIAN COURT UNIV on March 30, 2015jhh.sagepub.comDownloaded from

108 Journal of Hispanic Higher Education

3. Nebraska is not included as a treatment state as the data extend only until 2005. Nebraska passed an in-state resident tuition policy in 2006.

4. Subscripts are suppressed. 5. (See Note 3.)

References

Bean, F. D., Van Hook, J., & Woodrow-Latfield, K. (2002). Estimates of numbers of unauthorized migrants residing in the United States: The total, Mexican, and non-Mexican Central American unau- thorized populations in mid-2001. Special report. Washington, DC: Pew Hispanic Center.

Cornelius, W. A. (2004). Controlling “unwanted” immigration: Lessons from the United States, 1993-2004 (Paper wrkg92). University of California, San Diego: Center for Comparative Immigration Studies.

Dynarski, S. (2003). Does aid matter? Measuring the effect of student aid on college attendance and com- pletion. American Economic Review, 93, 279-288.

Dynarski, S. (2004). The new merit aid. In C. Hoxby (Ed.), College choices: The economics of where to go, when to go, and how to pay for it. Chicago: University of Chicago Press.

Erisman, W., & Looney, S. (2007). Opening the door to the American Dream: Increasing higher educa- tion and success for immigrants. Washington, DC: Institute for Higher Education Policy.

Feder, J. (2006). Unauthorized alien students, higher education, and in-state tuition rates: A legal analy- sis. Washington DC: Congressional Research Service. p.1.

Flores, S. M. (2007). The effect of in-state resident tuition policies on the college enrollment of undocu- mented Latino students in Texas and the United States. Unpublished doctoral dissertation, Harvard University, Cambridge, MA.

Flores, S. M., Horn, C. L., & Crisp, G. (2006). Community colleges, public policy, and Latino student opportunity. New Directions for Community Colleges, 133(2), 71-80.

Hebel, S. (2007, November 2). “Arizona’s Colleges Are in the Crosshairs of Efforts to Curb Illegal Immigration.” The Chronicle of Higher Education, 54(10), A15.

Immigrant Legal Resource Center. (2008, July). Promoting, educating, and empowering immigrants and their advocates. Retrieved November 23, 2008 from http://www.ilrc.org/dreamact.php.

Kane, T. J. (1994). College attendance by Blacks since 1970: The role of college cost, family background and the returns to education. Journal of Political Economy, 102, 878-911.

Kane, T. J. (2003). A quasi-experimental estimate of the impact of financial aid on college-going (Working Paper 9703). Cambridge, MA: National Bureau of Economic Research.

Long, B. T. (2004). Does the format of an aid program matter? The effect of in-kind tuition subsidies. Review of Economics and Statistics, 86, 767-782.

Massey, D., & Capoferro, C. (2008). The geographic diversification of American immigration. In D. S. Massey (Ed.), New faces in new places: The changing geography of American immigration (pp. 25-50). New York: Russell Sage Foundation.

McGee, P. (2005, July 24). More illegal immigrants in colleges: Enrollment has increased ninefold since the state allowed them to pay lower tuition. Fort Worth Star Telegram, p. B6.

Merriam Webster’s Online Dictionary. (2008). Retrieved May 14, 2008, from http://www.merriam- webster.com/dictionary/bipolar

Millard, A., & Chapa, J. (2004). Apple pie and enchiladas: Latino newcomers in the rural Midwest. Austin: University of Texas Press.

National Conference of State Legislators. (2006). In-state tuition and unauthorized immigrant students. Retrieved November 15, 2008, from http://www.ncsl.org/programs/immig/immig_InStateTuition 0706.htm

at GEORGIAN COURT UNIV on March 30, 2015jhh.sagepub.comDownloaded from

National Immigration Law Center. (2006). DREAM Act Summary. Retrieved November 15, 2008, from http://www.nilc.org./immlawpolicy/DREAM/dream_act_06_summary_2006-04.pdf

NEB. LEG. J., 99th Leg., 2d Sess., No. 60 (Apr. 13, 2006) (final reading of L.B. 239). Olivas, M. (2004). IIRIRA, the dream act, and undocumented college student residency. Journal of

College and University Law, 30, 435-464. Ono, K. A., & Sloop, J. M. (2002). Shifting borders: Rhetoric, immigration, and California’s Proposition

187. Philadelphia: Temple University Press. Passel, J. S. (2005a). Estimates of the size and characteristics of the undocumented population. (Pew

Hispanic Center). Retrieved September 1, 2005, from http://pewhispanic.org/files/reports/44.pdf Passel, J. S. (2005b). Unauthorized migrants: Numbers and characteristics—Background briefing pre-

pared for Task Force on Immigration and America’s Future. Retrieved September 1, 200 from http:// pewhispanic.org/files/reports/46.pdf

Passel, J. S., Van Hook, J., & Bean, F. D. (2004). Estimates of legal and unauthorized foreign born pop- ulation for the United States and selected states, based on Census 2000 (Report to the Census Bureau). Washington, DC: Urban Institute.

Redden, E. (2007). An in-state tuition debate. Retrieved March 1, 2007, from http://www.insidehigh- ered.com/news/2007/02/28/immigration

Rincon, A. (2005). Paying for their status: Undocumented immigrants and college access. Unpublished doctoral dissertation, University of Texas at Austin.

Tichenor, D. J. (2002). Dividing lines: The politics of immigration control in America. Princeton: Princeton University Press.

U.S. Census Bureau and U.S. Bureau of Labor Statistics. U.S. Current Population Survey, Merged Outgoing Rotation Groups, 1998 to 2005. Washington, D.C.

U.S. Department of Labor. (2002). Technical paper 63RV: Current population survey—design and methodology. Washington, DC: Bureau of Labor Statistics.

Stella M. Flores is Assistant Professor of Public Policy and higher education at Peabody College of Vanderbilt University.

Jorge Chapa is Director and Professor of Sociology and Latino studies at the University of Illinois at Urbana Champaign.

Flores, Chapa / Latino Immigrant Access to Higher Education 109

at GEORGIAN COURT UNIV on March 30, 2015jhh.sagepub.comDownloaded from