<%BANNER%>

Language and Earnings of Latinos in Florida

Permanent Link: http://ufdc.ufl.edu/UFE0022196/00001

Material Information

Title: Language and Earnings of Latinos in Florida The Effect of Language Enclaves
Physical Description: 1 online resource (77 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: bilingualism, enclaves, florida, hispanic, hispanics, immigrant, incorporation, language, latino, latinos, linguistic, miami, spanish
Latin American Studies -- Dissertations, Academic -- UF
Genre: Latin American Studies thesis, M.A.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Language ability has assumed priority in current studies of the economic success of immigrants and minority language speakers. Past studies have shown that language ability, a key human and cultural capital trait, tends to be positively associated with earnings. Building on this past research, the goal of this study is to examine how the effect of English language proficiency on earnings of Hispanic men in Florida varies by labor market context. Specifically, it aims to compare the difference in the effect of English language proficiency on earnings in areas densely populated by Spanish speakers to the effect of language on earnings in areas dominated by English speakers. I predict that English language proficiency will have a greater impact on earnings in areas where Spanish is not widely spoken. In areas where there are large enclaves of Spanish-speakers, English will likely be a less important determinant of earnings. The effect of bilingualism on earnings is also analyzed in this manner. To test my hypotheses, my study consists of two parts: the first based on statistical analysis of US census data and the second based on qualitative interviews. Findings show that English language ability is indeed an important determinant of earnings both in areas with a high proportion of Spanish-speakers and in areas with a low proportion of Spanish-speakers. However, results from the statistical analysis show that English language ability has a greater impact on earnings in areas with a high proportion of Spanish-speakers. While English language proficiency yields greater earnings in these areas, Spanish language proficiency also has a positive effect on earnings. Fully bilingual Hispanics earn more than their English only counterparts in these areas.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (M.A.)--University of Florida, 2008.
Local: Adviser: Wood, Charles H.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022196:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022196/00001

Material Information

Title: Language and Earnings of Latinos in Florida The Effect of Language Enclaves
Physical Description: 1 online resource (77 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: bilingualism, enclaves, florida, hispanic, hispanics, immigrant, incorporation, language, latino, latinos, linguistic, miami, spanish
Latin American Studies -- Dissertations, Academic -- UF
Genre: Latin American Studies thesis, M.A.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Language ability has assumed priority in current studies of the economic success of immigrants and minority language speakers. Past studies have shown that language ability, a key human and cultural capital trait, tends to be positively associated with earnings. Building on this past research, the goal of this study is to examine how the effect of English language proficiency on earnings of Hispanic men in Florida varies by labor market context. Specifically, it aims to compare the difference in the effect of English language proficiency on earnings in areas densely populated by Spanish speakers to the effect of language on earnings in areas dominated by English speakers. I predict that English language proficiency will have a greater impact on earnings in areas where Spanish is not widely spoken. In areas where there are large enclaves of Spanish-speakers, English will likely be a less important determinant of earnings. The effect of bilingualism on earnings is also analyzed in this manner. To test my hypotheses, my study consists of two parts: the first based on statistical analysis of US census data and the second based on qualitative interviews. Findings show that English language ability is indeed an important determinant of earnings both in areas with a high proportion of Spanish-speakers and in areas with a low proportion of Spanish-speakers. However, results from the statistical analysis show that English language ability has a greater impact on earnings in areas with a high proportion of Spanish-speakers. While English language proficiency yields greater earnings in these areas, Spanish language proficiency also has a positive effect on earnings. Fully bilingual Hispanics earn more than their English only counterparts in these areas.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (M.A.)--University of Florida, 2008.
Local: Adviser: Wood, Charles H.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022196:00001


This item has the following downloads:


Full Text





LANGUAGE AND EARNINGS OF LATINOS IN FLORIDA:
THE EFFECT OF LANGUAGE ENCLAVES




















By

MOLLY DONDERO


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF ARTS

UNIVERSITY OF FLORIDA

2008


































2008 Molly Dondero



































To my parents, Marian and Lawrence









ACKNOWLEDGMENTS

Many thanks go to the members of my supervisory committee for their guidance and

accessibility. I appreciate both their assistance with this thesis and the positive influence they

had on my overall academic development. I am especially grateful to my committee chair, Dr.

Charles Wood, for helping me to develop this project and for reminding me to "keep it simple."

His direction has been invaluable. I thank Dr. Carmen Carri6n-Flores for lending her expertise in

labor economics to this project. I thank Dr. Efrain Barradas for his support and advice.

I owe a great deal of gratitude to the Center for Latin American Studies for providing me

with the Interdisciplinary Field Research Grant that made this study possible. I also thank my

informants who so generously took time out of their schedules to participate in this study.

For their unconditional love and support, I thank my parents, Marian and Lawrence

Dondero. They have fostered my love of learning, and are a great source of inspiration for me.

Finally, my heartfelt thanks go to Werllayne Nunes for his constant encouragement and moral

support. On countless occasions, he selflessly put aside his own work to help me move toward

completing this thesis.









TABLE OF CONTENTS

page

A CK N O W LED G M EN T S ................................................................. ........... ............. .....

L IST O F T A B L E S ...................... ............... ....................................................... . 7

LIST OF FIGURES .................................. .. ..... ..... ................. .8

LIST OF A BBREV IA TION S ...................... ............................................................9

A B S T R A C T ................................ ............................................................ 10

CHAPTEr

1 INTRODUCTION ............... ................. ........... .............................. 12

O v erv iew ................... ...................1...................2..........
B background ................................................................................ 13
Spanish-Speaking Population in the United States ............................... ............... .13
Spanish-Speaking Population in Florida ...................................................................... 14
R research Q u estion s..................................................... ................. 15

2 L ITE R A TU R E R E V IE W ........................................................................ .. ....................... 17

C onceptual F ram ew ork ............ .................................................................... ...... .. ... 17
E conom ics of Language ........................................................ ......... ............ 17
Form s-of-Capital M odel of Incorporation.................................................................... 18
L language and E arnings................ ................................ .......... ........................ ....22
Effect of Language Enclaves on Language and Earnings ..................................................24
Value of Bilingualism in the Labor M arket................................................. ....... ........ 28

3 D A TA A N D M ETH O D O LO G Y ........................................ ............................................33

R e se a rc h D e sig n ............................................................................................................... 3 3
Quantitative Analysis............. .. ..... .............. ............... 34
U S C en su s D ata .............................................................................................3 4
M easuring the Language Enclave ............................................................................35
Sam ple D description ................. .................................... ...... ........ .. .............36
D description of V ariables .................. .................................... ................. 37
D dependent variable ..................................................... ... .. ........ .... 37
Independent variable .................................. ... .. ..... ............ 38
Control variables ......................... ..... ..... .. .. ......... ......... 39
Statistical M odel/D ata A naly sis ........................................................... .....................4 1
Qualitative Analysis........... .... ..... .............................. 42
M ethodological Caveats ................................... .. .... ..... .. ............43









4 R E SU L T S ....................................................... 49

Sum m ary Statistics ........................ ........ ...... .......... ......... ....... ......................49
Research Question One: The Effect of Language Enclaves on Returns to English ..............50
Research Question Two: Returns to Bilingualism............................................. 55

5 DISCUSSION ......... .. ....... ........ ... ....................................... 62

T h eoretical Im plication s ........................................ ...................................... ..................... 62
Language, Earnings, and Labor Market Characteristics ...........................................62
Returns to English language proficiency ...................................... ............... 62
R returns to bilingualism ........................ .. ...................... .. ...... .... ...... ...... 65
Im migrant/M minority Incorporation ............................................................................ 66
P o licy Im p lic atio n s ........................................................................................................... 6 7
Suggestions for Future R research .................................................. .............................. 68

6 CON CLU SION .......... ................................................................. ............. ... 70

L IST O F R E F E R E N C E S .............................. ........................................... ....................................72

B IO G R A PH IC A L SK E T C H .............................................................................. .....................77









LIST OF TABLES


Table page

3-1 Language spoken by concentration of Spanish-speakers, Florida 2000 ..........................47

4-1 Mean and standard deviation of variables used in Low Concentration model ................58

4-2 Mean and standard deviation of variables used in High Concentration model ..............59

4-3 Returns to English-speaking ability ...................................................... ............... 60

4-4 R returns to bilingualism ...................... ...... ................ ..................... .. .....61









LIST OF FIGURES

Figure page

3-1 Florida linguistic com position. ........................................ ............................................48











HC

LC

Super-PUMA


LIST OF ABBREVIATIONS

High concentration of Spanish-speakers

Low concentration of Spanish-speakers

Super Public Use Microdata Area









Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Arts

LANGUAGE AND EARNINGS OF LATINOS IN FLORIDA:
THE EFFECT OF LANGUAGE ENCLAVES

By

Molly Dondero

May 2008

Chair: Charles Wood
Major: Latin American Studies

Language ability has assumed priority in current studies of the economic success of

immigrants and minority language speakers. Past studies have shown that language ability, a key

human and cultural capital trait, tends to be positively associated with earnings.

Building on this past research, the goal of this study is to examine how the effect of

English language proficiency on earnings of Hispanic men in Florida varies by labor market

context. Specifically, it aims to compare the difference in the effect of English language

proficiency on earnings in areas densely populated by Spanish speakers to the effect of language

on earnings in areas dominated by English speakers. I predict that English language proficiency

will have a greater impact on earnings in areas where Spanish is not widely spoken. In areas

where there are large enclaves of Spanish-speakers, English will likely be a less important

determinant of earnings. The effect of bilingualism on earnings is also analyzed in this manner.

To test my hypotheses, my study consists of two parts: the first based on statistical

analysis of US census data and the second based on qualitative interviews. Findings show that

English language ability is indeed an important determinant of earnings both in areas with a high

proportion of Spanish-speakers and in areas with a low proportion of Spanish-speakers.

However, results from the statistical analysis show that English language ability has a greater









impact on earnings in areas with a high proportion of Spanish-speakers. While English language

proficiency yields greater earnings in these areas, Spanish language proficiency also has a

positive effect on earnings. Fully bilingual Hispanics earn more than their English only

counterparts in these areas.









CHAPTER 1
INTRODUCTION

Overview

Language ability has assumed priority in current studies of the economic success of

immigrants and minority language speakers. Language often serves as a key human capital and

cultural capital trait that facilitates incorporation into the host country. Many scholars cite the

ability of immigrants to effectively communicate with members of the receiving country as the

most important alterable factor that affects their integration into their country of destination and

their absorption into the labor market (Dustmann and van Soest 2001, 2002).

The goal of this study is to examine the effect of English language proficiency on

earnings among Latinos1 in Florida. Specifically, it aims to compare the difference in the effect

of English language proficiency on earnings in areas densely populated by Spanish speakers to

the effect of language on the earnings in areas dominated by English speakers. I predict that

English language proficiency will have a greater impact on earnings in areas where Spanish is

not widely spoken. In areas such as ethnic enclave economies, where Spanish is commonly

spoken, English will likely be a less important determinant of earnings. Simply stated, the more

Spanish that is spoken in a given area, the less important English is to earnings in that area. I

also expect to find the inverse to be true of Spanish language skills; in a Spanish-language

enclave, Spanish proficiency will be more important to earnings than it is outside the enclave.

Results from this study offer valuable insight about the factors that affect the economic

integration of immigrants and non-native speakers into the labor market. Findings also have

important policy implications for English language and bilingual training programs in Florida.


1 Following current US Census Bureau terminology, the terms "Latino" and "Hispanic" are used
interchangeably throughout this paper.









Background

Spanish-Speaking Population in the United States

"The political and social controversies surrounding the position of English versus other

languages in the United States has endured since the founding of the nation" (Stevens 1999).

The recent growth of the Spanish-speaking population in the US has reignited the language issue

and transformed it into a symbolic battleground in the current immigration debates. Spanish now

ranks as the second most common language spoken in the US (Shin and Bruno 2003). In the

Census 2000, over 60% of respondents who speak a non-English language at home reported

speaking Spanish (Mora 2003). This represents a 10% increase in the number of Spanish-

speakers reported from the 1990 Census (Mora 2003).

The rise of Spanish language use in the US stems from the rapid growth of the country's

Latino population. While not all Latinos speak Spanish, nearly all Spanish speakers in the US

are Latinos (Santiestevan 1991). The latest updates of the 2000 Census estimate the total Latino

population to be 41.3 million, or 14% of the total US population, making Latinos the country's

largest minority group (US Census Bureau 2005). According to US Census Bureau projections,

these numbers show no signs of waning. On the contrary, estimates predict the Latino population

to more than double to 102.6 million by 2050 (US Census Bureau 2005). Among the current

Latino population, nearly 31 million people over age 5 report speaking Spanish at home,

constituting a ratio of more than 1-in-10 ratio of household residents in the US (US Census

Bureau 2005). Debate remains over whether Spanish language use in the US will continue to

grow at a rate as fast as that of the Latino population. Some believe that Spanish-speakers will

prove to be a unique group in US linguistic history by maintaining their native tongue for several

generations. Others contend that as English language fluency increases across generations,

Spanish-speakers will gradually abandon Spanish in favor of English, as have many other









immigrant and minority language groups before them. No matter what the fate of Spanish

language in the US proves to be, the sheer number of Spanish-speakers in the US today has

undeniably reshaped the country's current linguistic landscape.

Spanish-Speaking Population in Florida

Florida's Latino population is the third largest in the nation, behind California and Texas,

respectively (Pew Hispanic Center 2006). Of the state's more than 18 million-plus residents,

over 3.2 million-or more than 19% of the state's total population-are Latino (US Census

Bureau 2008). Over 75% of these residents report speaking Spanish at home, making Florida the

state with the fifth largest proportion of residents who speak Spanish at home (US Census

Bureau 2007; Viglucci 2001). Even though a 1988 law designated English as the state's official

language, these numbers show that Spanish-speakers in Florida continue to exert a powerful

presence.

The uneven distribution of the Spanish-speaking population in Florida makes it an ideal

setting for a comparative research design. Although recent research highlights growing pockets

of Latinos in Central and North Florida, most of Florida's Spanish-speaking population is

concentrated in South Florida, and in Miami-Dade County in particular (Duany and Matos-

Rodriguez 2006). Miami-Dade is home to over 1.3 million people of Hispanic origin. With

Hispanics representing 61.3% of the total population of the county, Miami-Dade has the

distinction of being one of 50 counties nationwide in which Hispanics create the majority (US

Census Bureau 2008). Wilson and Portes (1980) have well documented the growth of the Cuban

enclave economy in the city of Miami and the ways in which it stimulated the development of

the mainstream economy in Miami (Portes 1987). The latest US Census estimates that Hispanic-

owned businesses account for 54.9% of all businesses in Miami-Dade County (US Census

Bureau 2008). In addition, Miami has emerged as a hub for international business, particularly









for multinational corporations, financial institutions, and Spanish-language media conglomerates

from Latin America and the Caribbean. Its large Spanish-speaking population and its status as a

prominent international business center have prompted many to refer to the city as "the financial

capital of Latin America." Such characteristics make it likely that English language ability will

have a lesser effect on earnings in Miami.

Research Questions

Having established the context and setting of this study, I turn now to the specific research

hypotheses. This study is premised on the theoretical proposition that labor market context will

dictate the value that the market places on certain forms of human and cultural capital, in this

case language ability. Specifically, it asks: Does the presence of a large Spanish language

enclave alter earnings returns to English language proficiency? The corresponding research

hypotheses predict 1) that English will be a less important determinant of earnings in areas with a

high concentration of Spanish speakers and 2) that bilingual English-Spanish skills will be more

important in areas with a high concentration of Spanish-speakers. To test these hypotheses, my

project consists of two parts: the first based on statistical analysis of census data and the second

based on qualitative interviews.

Three features set this study apart from others. First, it is the first study of this nature to

focus specifically on the Florida labor market; other studies on this topic have used data from

either the Southwest US, the US in general, or Canada. Second, it provides empirical evidence

of an unofficial, minority language-Spanish- as potential asset in the labor market (Pendakur

and Pendakur 2002). Third, the qualitative component of this study gives a unique perspective

on the issue of language and earnings, thereby going beyond previous research based exclusively

on quantitative methodologies.









This study is nonetheless firmly rooted in traditional concepts in sociology and labor

economics. Chapter 2 offers a literature review that details the established theoretical and

empirical work that informs this study. Chapter 3 describes the design and methodology of the

study. Chapter 4 presents results, and Chapter 5 discusses the theoretical and practical

implications of the findings.









CHAPTER 2
LITERATURE REVIEW

The increased flows of immigration to the US over the past several decades spawned an

abundance of studies exploring the role of English-language proficiency in the US labor market.

Yet, despite the current influx of immigrants from Spanish-speaking Latin America, and the

continued growth of linguistic pluralism in the United States, scholars have produced little recent

research on the topic. Most of the published literature dates back to over a decade ago and, in

many cases, relies on data that are even older (Mora 2003). Within these studies, only a small

body of research addresses the relationship between language, earnings, and linguistic

concentrations. These few empirical studies have, for the most part, produced opposing and

inconsistent conclusions about the nuanced ways in which the linguistic profile of the labor

market affects the economic value of language proficiency.

In this chapter, I review the existing literature on the topic, and highlight areas that merit

additional research. The first section presents the conceptual framework for the study. It

examines the economics of language and uses a "forms-of-capital" model to establish the

importance of language in the economic incorporation of non-native speakers. The second

section reviews the empirical studies that address the relationship between language and

earnings. The third section summarizes the studies that specifically treat the relationship between

language, earnings, and linguistic concentrations. The final section turns to the existing literature

on the value of bilingualism in the workplace.

Conceptual Framework

Economics of Language

What role does language play in the economic incorporation of immigrants and minority

language speakers? Although the answer to this question may seem transparent, a closer analysis









reveals that language functions as one of the most versatile and valuable traits in the labor

market. As such, it plays a diverse and significant role in the economic incorporation of minority

language speakers. For this reason, it is useful to outline the theories that hypothesize the ways

in which language operates in the labor market.

Beginning with J. Marschak's pioneering 1965 study on economic approaches to

language, the economics of language is a relatively recent area of specialization within the social

sciences (Grin 1994). It refers to an interdisciplinary field of research that examines the

relationship between linguistic and economic variables (Grin 1994). Prior to the publication of

Marschak's work, mainstream economics largely overlooked this relationship, believing

language and other "ethnic" characteristics to be of little overall importance in the labor market

(Grin 1994). However, increased linguistic, ethnic, and cultural pluralism in contemporary

society, and rising awareness of social problems resulting from such pluralism, stimulated the

growth of studies in this field (Grin 1994).

Forms-of-Capital Model of Incorporation

The economic consequences of language proficiency are of particular importance to

research on the incorporation of immigrants and minorities. The recognition of language as a

key component in labor market outcomes contributed to the current shift in theories of immigrant

incorporation. Traditional assimilation theories no longer suffice as an accurate model of

incorporation of contemporary immigrants, as they fail to account for the diversity of

incorporation experiences by different groups (Tienda 1983). Rather, the contemporary

incorporation experience is better framed within the "forms-of-capital model," developed by Nee

and Sanders (2001). This model privileges the factors that explain the diversity in the modes of

incorporation of immigrants. It views incorporation mainly as a function of the human, social,

and cultural capital that immigrants possess and accrue. The role of language in









immigrant/minority incorporation can be best understood through this model since language

serves simultaneously as multiple forms of capital: human, cultural, social, and linguistic. While

this study borrows the basic framework from Nee and Sanders' model, it modifies the model in

three ways: first, it examines language as forms of capital at the individual, rather than household

or family level; second, it expands the model to include the incorporation not only of

immigrants, but also of minority language speakers in general; and third, it takes into

consideration the ways in which certain labor market characteristics, specifically the linguistic

composition of the labor market, affect the economic value of these forms of capital.

Nee and Sanders' model draws on the concepts of capital developed by Gary Becker

(1993), James Coleman (1988), and Pierre Bourdieu (1986). Bourdieu (1986) theorizes that the

general notion of capital provides a means for understanding most types of interaction in the

social world. "It is what makes the games of society-not the least, the economic game-

something other than simple games of chance offering at every moment the possibility of a

miracle" (Bourdieu 1986: 46). The application of the forms-of-capital model to the study of

language and incorporation is particularly useful if one views linguistic exchange as "an

economic exchange... established within a symbolic relation of power between a producer,

endowed with a certain linguistic capital, and a consumer" (Bourdieu 1991:66). According to

Bourdieu, words represent far more than simple means of communication; they are signs of

authority and wealth that reveal a particular social value. From this perspective, language takes

on a prominent role in the incorporation of minority language speakers. The remaining parts of

this section define the concepts of human, social, cultural, and linguistic capital, and the ways in

which language functions as a form of each type of capital in the labor market.









Most studies of language and immigrant/minority incorporation view language

proficiency primarily as a human capital trait (Park 1999). The concept of human capital offers a

way to understand differences in earnings that are not fully explained by external factors (Becker

1993). Becker (1993) broadly defines human capital as knowledge, skills, and health. As such,

human capital is different from physical or financial capital because, unlike these tangible forms

of capital, it cannot be separated from an individual (Becker 1993). In this sense, language can

be considered one of the most basic forms of human capital. Simply stated, language is essential

to effective communication which, in turn, is essential to enhanced productivity. Indeed, many

earnings analyses have shown that language has an effect on earnings that is comparable to that

of some of the most common human capital characteristics, such as education and number of

years in the host country (Park 1999; Chiswick 1991; McManus 1985; Grenier 1984). Chiswick

(1978) was the first researcher to apply the concept of human capital to the economic

achievement of immigrants (Portes 1995). He concluded that individual skills greatly impact the

incorporation experience. Sociologists, however, take issue with this approach to incorporation,

noting that group membership and other social contexts are also at work (Portes 1995). The

notion of social capital has thus emerged as an important concept.

The concept of social capital offers a way to theorize the value of group membership in

the incorporation experience. Broadly defined, social capital refers to benefits or resources

created by and derived from social networks (Portes 1998; 2002). It is the "aggregate of the

actual or potential resources which are linked to possession of a durable network or more or less

institutionalized relationships of mutual acquaintance and recognition-or in other words, to

membership in a group-which provides each of its members with the backing the collectivity-

owned capital" (Bourdieu 1986:51). Language, as the primary means of social interaction,









fosters social capital. A common language cultivates a sense of shared identity. It functions as a

criterion for membership into a given group, and thus, as a networking tool that can facilitate

labor market success.

The social aspect of language relates directly to the third form of capital: cultural capital.

Cultural capital refers to knowledge, skills, education, and values acquired through cultural

transmission resulting from membership in a given class, region, nation, or ethnic group

(Bourdieu 1986). Bourdieu developed the concept of cultural capital as a way to explain unequal

scholastic achievement among students from different class backgrounds. It represented a

departure from past theories that viewed academic success as the result of solely genetic and

human endowments. He hypothesized that students who possess knowledge of the mainstream

culture in which their education system is rooted generally achieve greater academic success.

The same logic can be applied to economic achievement in the labor market. In order to

successfully navigate the labor market, workers must possess knowledge of the cultural norms

and codes that underpin the market. Language, which is culturally transmitted, serves as a

vehicle that allows an individual to understand these norms and operate effectively within them.

Thus, the cultural value of language makes it an essential trait for labor market success

(Pendakur and Pendakur 2002). However, as cultural signifier, language can also be a detriment

in the labor market when used as grounds for economic discrimination based on culture

(Pendakur and Pendakur 2002).

Bourdieu identifies another form of capital specifically related to language skills-

linguistic capital. Linguistic capital represents an embodied form of cultural capital in the sense

that, once acquired, it cannot be separated from the individual. However, the value of linguistic

capital, like the value cultural capital, is relative. The "market" determines its ultimate worth









(Bourdieu 1991). That is to say, knowledge of a minority language may highly valued or even

essential for survival within a given language enclave, but useless in another context. The value

of linguistic capital is therefore contingent on the context in which it is used.

These four forms of capital-human, social, cultural, and linguistic-are primarily

symbolic. Unlike economic capital, the transmission and acquisition of these forms of capital is

not easily recognized. As such, their value is often underestimated or unrecognized (Bourdieu

1986). However, these forms of capital, under certain circumstances, can be converted into

economic capital (Bourdieu 1986), thus making language proficiency an asset in the labor

market. Even so, language and other human, cultural, and social capital variables only partially

explain wage differentials among immigrant/minority language groups. Tienda (1983) argues

that such a model of incorporation needs to be further expanded to account for structural forces

that also affect labor market outcomes. Structural conditions such as labor market characteristics

and public opinion toward immigrants and foreign language use are particularly relevant to a

comprehensive model of incorporation (Tienda 1983). While it is beyond the scope of this

paper to include a comprehensive inventory of labor market characteristics in the empirical

analysis, this study focuses on one defining macro-structural feature-the linguistic profile of the

labor market-and its interplay with the relationship between language and earnings.

Language and Earnings

This section reviews the studies that empirically demonstrate the impact of English

language proficiency on the labor market outcomes of non-native speakers, in particular

Hispanic males. In the early 1990s, earnings analyses of Hispanic males began to devote more

attention to the role of English language proficiency. The introduction of language as a key

variable in the earnings functions offered insight into causes of wage differentials. "Despite

different definitions of English proficiency and a variety of methodological approaches, nearly









all studies have found that workers experience an earnings "penalty" for the lack of the ability to

speak English well" (Mora 2003).

McManus, Gould, and Welch's (1983) seminal study of the role of English language

proficiency on earnings uses data from the 1975 Survey of Income and Education (SIE) to

explore the cost of language disparity among Hispanic males in the US. Regression estimates

from this study suggest that, compared to their fluent Anglo counterparts, English-deficient

Hispanic males experience earnings penalties between 17-30%, depending on the level of

education being compared. Furthermore, their findings show that English language deficiency

reduces earnings returns to education and work experience. Grenier's (1984) study also uses the

1975 SIE to estimate the effect of language characteristics on earnings of Hispanic-American

men with limited English proficiency. His regression analysis finds that 1) Hispanic males earn

significantly less than their Anglo counterparts, and 2) language characteristics explain up to one

third of these earnings differentials. Subsequent studies report similar findings. Chiswick and

Miller's (1992) statistical analysis shows that immigrants who speak English well or very well

earn on average 17% more than those who are English-deficient. Using a different sample and

reference group, Borjas (1994) concludes Hispanic immigrants who do not speak English earn

17% less than those who speak English. In keeping with these results, other studies have shown

that the rate of immigrant wage assimilation is proportional to increase in English-speaking skills

and time spent in the US (Funkhouser 1996; Carliner 1995; Gonzalez 2000).

Such research leads Chiswick and Miller (1992, 1995, 2002) to argue that English

language proficiency ranks as one of the most important determinants of earnings of non-native

speakers. They conclude that "the acquisition of English language skills clearly pays in the labor

market" (2002: 42). Despite strong empirical evidence from Chiswick and Miller, literature on









the topic remains divided with regard to the extent to which language affects earnings (Park

1999). While language ability is positively associated with earnings, other research suggests that

it is not a major determinant of earnings. Such research points to other factors, such as national

origin, length of US residence, and education, as the primary predictors of earnings of non-native

speakers (Borjas 1982; Tienda 1983).

Much of this previous research invokes a design that focuses attention almost exclusively

on individual level characteristics. In doing so, this research overlooks the importance of

context. Specifically, as I argue in this study, income returns to language are not uniform across

different contexts. Instead, the relationship between proficiency and earnings is contingent upon

the linguistic profile of the labor market in which people work. In this case, a high proportion of

Spanish-speakers in a given area will lower income returns to English proficiency in that area.

Thus, this study will show that the inclusion of a minority language enclave measure can provide

important insights about the extent to which language explains variance in earnings.

Effect of Language Enclaves on Language and Earnings

While most prior research agrees that greater English language proficiency yields greater

earnings, this consensus disappears when analyzing the effect of language on earnings in regions

with strong minority language foundations, such as ethnic enclave economies (Davila and Mora

2000). Theoretical formulations that treat the topic tend to agree that the presence of a large

immigrant/minority population in a given area will alter returns to various determinants of

earnings, such as language, ethnicity, and other variables. "The returns accruing to migrants who

move to areas of high ethnic density should differ from those accruing to migrants who move to

areas of low ethnic density, although it is not obvious whether these returns will be positive or

negative" (Tienda 1992: 661). McManus (1990) theorizes that large ethnic enclaves lower

earnings returns to English language proficiency. That is to say, the market in these areas places









less value on English language. Such thinking follows the forms-of-capital model, which would

assume that the labor market value of a given language depends on its demand in the

marketplace (Bourdieu 1986), and that structural forces, such as the existence of a minority

language enclave, would indeed alter value of human and cultural capital in a given area. Such a

model helps explain why a given language is important to earnings in one area, but less

important in another.

These conceptualizations find their roots in the ethnic enclave economy theories

developed by Wilson and Portes (1980). They define the ethnic enclave economy as a

concentrated area of businesses owned by employers of immigrant or minority ethnic

backgrounds that employ co-ethnic workers. Groups in these areas tend to maintain their

cultural customs and language to a greater extent than comparable groups outside the enclave.

Wilson and Portes (1980) theorize that the ethnic enclave economy functions as a mode of

immigrant incorporation by providing provides access to jobs, opportunities and resources that

might not otherwise be as available for minorities outside the enclave. However, debate exists

over the extent to which the ethnic enclave economy works as a successful incorporation

technique. Wilson and Portes (1980) theorize that returns to human capital brought from the

home country are higher in the enclave economy than in the mainstream economy. While this

may create initial earnings advantages for immigrants and minorities in these areas, others

contend that the ethnic enclave economy acts as a mobility trap in which earnings eventually

plateau (Sanders and Nee 1987). (Such traps generally refer to employees, not employers, in the

enclave economy). This study does not focus specifically on ethnic enclaves economies per se;

that is to say, it does not use the number of immigrant-owned businesses in an area as measure of

enclave. Rather, it focuses on a broader definition of enclave-the linguistic enclave-measured









by the number of Spanish-speakers in a given area. Nevertheless, it is important to highlight the

concept of the ethnic enclave economy because its basic theoretical assumptions apply directly to

this study.

Despite these concordant theoretical assumptions, empirical analyses have produced

inconsistent results. These inconsistencies may in part be due to the fact that only a small

number of studies consider the effects of language enclaves on earnings returns to language

skills. Results from these studies can be roughly classified into four main categories: 1) those

that find that the presence of a large minority language enclave lowers returns to English

(McManus 1990; Chiswick and Miller 2002); 2) those that find that the English deficiency

earnings penalty is greater inside the linguistic enclave (Bloom and Grenier 1992); 3) those that

find mixed results (Davila and Mora 2003; Hand 2006); and 4) those find no evidence that the

existence of a linguistic enclave alters earnings returns to language skills (Fry and Lowell 2003).

In his analysis of earnings regressions for Hispanic males based on 1980 US Census data,

McManus (1990) finds that the enclave reduces earnings penalties associated with English

language deficiency. He invokes the Hispanic ethnicity variable to operationally define the

enclave. He concludes that the greater the Hispanic population, the lower the returns to English.

Chiswick and Miller's (2002) subsequent study utilizes 1990 US Census data to test similar

hypotheses. They conclude that, all things being equal, average earnings tend to be lower for

immigrants that live in an area with a large minority language concentration. They find that the

English-deficient wage penalty is also smaller in these areas.

When Bloom and Grenier (1992) apply similar regression techniques to 1970 and 1980

US Census data, they find that the language-based earnings differentials between Hispanics and

Whites are actually greater in areas with a large Spanish-speaking population. To determine









areas of high and low concentrations of Spanish-speakers, Bloom and Grenier, like McManus,

use Hispanic origin as a proxy for Spanish language and White of non-Hispanic origin as a proxy

for English.

Davila and Mora's (2000) study, which uses 1990 US Census data, seems to support

Bloom and Grenier's results to a large extent and contradict McManus's (1990) and Chiswick

and Miller's (2002) findings. Employing the US-Mexico border as a minority language enclave,

they compare the effect of English language proficiency on earnings of Mexican-Americans

along the US-Mexico border relative to their non-border counterparts in the rest of the US. Their

regression results suggest that the English deficiency earnings penalty is slightly greater for

Mexican-American males in border cities than in non-border cities. However, in the case of

Mexican immigrants, there is no significant difference in the English deficiency earnings penalty

between border and non-border cities.

Critics cite Davila and Mora's measure of minority language enclave as a possible source

of error in their analysis. Hand (2006) believes that their use of border cities and non-border

cities as proxies for enclaves and non-enclaves may have skewed results. He maintains that

border cities cannot necessarily be described as enclaves solely on the basis of their status as

traditional immigrant-receiving destinations. Furthermore, he notes that Davila and Mora make

no attempt to account for language enclaves or ethnic enclave economies that may exist in non-

border metropolitan areas in the US (Hand 2006). Hand's own empirical study of the role of

linguistic enclaves in wage determination of minority speakers in the Southwest US attempts to

remedy this perceived error by using 2000 US Census data to measure the density of Spanish-

speakers throughout defined areas in New Mexico and Arizona. However, his analysis also

produces mixed results, though they run contrary to Davila and Mora's findings. Like Davila









and Mora, Hand runs separate regression models for Mexicans and Mexican-Americans. He

finds that English-deficiency earnings penalties are indeed reduced in the linguistic enclave but

only for foreign-born respondents. Results for native-born Spanish-speaking respondents do not

prove statistically significant and are thus inconclusive.

Value of Bilingualism in the Labor Market

If earnings returns to English decrease as the size of the Spanish-speaking population

increases, do earnings returns to Spanish then increase? Simple supply and demand theory

would assume that areas with a large population of Spanish-speakers would have an increased

demand for Spanish-speaking workers. As discussed in the previous section, past research has

shown that, bilingualism is essential for non-native speakers to successfully integrate into and

compete in the US labor market. However, for native speakers of English, monolingualism in

English is not generally perceived as a disadvantage in the labor market. However, this study

asks: do bilingual speakers earn more than their monolingual English counterparts in areas with

a large proportion of Spanish-speakers? In other words, are there earnings penalties for not

having Spanish language proficiency in those areas?

Examined from the forms-of-capital perspective, bilingual abilities theoretically represent

a greater amount of human, cultural, and social capital than monolingual abilities. Bilingualism

offers the distinct advantage of being able to access dual cultures, networks, and other such

resources. Thus, knowledge of an additional language, even a minority language, should

improve labor market outcomes (Pendakur and Pendakur 2002). Yet, despite this logic and the

well-established body of research on the economics of language, a relatively small amount of

research addresses the value of bilingualism in the labor market.

Educational research, on the other hand, has devoted much attention to the effect of

bilingualism in the classroom. Prior to 1960, mainstream belief in the fields of education and









psychology argued that bilingualism caused academic failure, mental confusion, and

psychological damage (Portes and Schauffler 1996). However, a crop of sound methodological

studies, starting in the 1960s and continuing to the present, reversed this belief after showing

that, all other factors being equal, bilingualism is associated with higher scholastic achievement,

greater cognitive flexibility, and a better capacity to deal with abstract concepts (Portes and

Schauffler 1996). These studies have shown that "instead of creating 'confusion,' having two

symbols for each object enhanced understanding" (Portes and Schauffler 1996: 11).

Just as these enhanced abilities have been shown to increase academic achievement, so

too should they reap positive outcomes in the labor market, or at least in areas with a large

population of non-native English speakers. Yet, past research in the field of labor economics and

sociology has produced inconsistent results on this topic. Qualitative findings from Portes and

Stepick's (1993) study on the transformation of Miami provide evidence of the perceived

advantages of bilingualism, especially in labor markets in areas with a large Spanish-speaking

population. A quote from one of their interviews with a Cuban civic activist and head of a

multiethnic community organization in Miami illustrates this point. He observes:

Language has great importance because if an individual owns a store whose clients come
from Latin America, he will need bilingual employees. During Christmastime, ninety
percent of the stores advertise for bilingual employees. To a person who does not know
the language, this situation represents an economic problem because he knows that, unless
he knows Spanish, he would not compete successfully in the labor market (Portes and
Stepick 1993: 12).

In other words, the large number of Spanish-speaking consumers in Miami gives bilingual

workers an advantage over their English-only counterparts in the labor market.

However, such opinions have received little empirical backing from other studies. Other

researchers contend that bilingual language skills present no additional earnings advantage in the

labor market. Carliner (1981) theorizes that the earnings of bilinguals and monolingual are









likely to be roughly equivalent. He posits, "In multilingual societies, if labor demand for

speakers of one language exceeds the supply of native speakers, bilingual workers will generally

come from other language groups. There will be a wage premium for speaking the "excess

demand" language but no additional premium for being bilingual" (Carliner 1981: 384). He finds

evidence of this pattern in data from the 1970 Canadian Census.

Findings from Fry and Lowell's (2003) analysis of the value of bilingualism in the U.S.

labor market support this theory. In their regression of earnings on a variety of language

variables from the 1992 National Adult Literacy Survey, they find that in the general US labor

market, bilingual workers have a marginally significant earnings advantage over their

monolingual English counterparts. However, these earnings returns disappear after they remove

the effects of education, age, and other control variables. Even after inserting a measure for

geographic linguistic concentrations, they find no evidence that language enclaves alter the

returns to bilingualism. They conclude that higher returns to bilingualism are likely limited to

specialized jobs that deal primarily in the international labor market.

Pendakur and Pendakur's (2002) study of the economic impact of bilingualism in the

Canadian labor market produces slightly different and more varied results. Using a language-as-

human-capital perspective, they predicate their study on three main theoretical assumptions that

follow a logic similar to that employed in this study: 1) polyglotss should earn more than

unilinguals" 2) "different cities (with different populations of majority and minority language

speakers) should have different patterns of returns to language knowledge" and 3) "these returns

should be correlated with the size of the linguistic communities" (Pendakur and Pendakur 2002:

150). Applying regression techniques to data from Canadian census, they find that those

respondents who are fluent in Canada's two official languages-English and French-tend to









earn more than their monolingual (English or French) peers. However, when assessing returns to

knowledge of unofficial languages, the story changes. Their analysis finds no earnings

advantages for possessing knowledge of one official and one or more unofficial languages. In

fact, those who speak one or more unofficial languages (in addition to one official language)

actually earn less than those fluent in only one official language. While earnings returns to

language skills do vary slightly according to the size of the corresponding linguistic community,

in no instance do bilinguals fluent in an unofficial language earn more than monolinguals fluent

in one of the official languages. They attribute these rather counterintuitive findings to

discrimination caused by the ethnic and cultural dimensions of language.

In contrast to these findings, evidence from the 1990 US Census points to a possible shift

in the economic value of English-Spanish bilingualism in the US labor market. In their study of

income patterns of bilingual and English-only Hispanics in the US, Boswell and Fradd (1999)

find that bilingual language skills have a greater economic value than monolingual English skills

in select US cities with large Hispanic populations. Their simple cross-tabulations of mean

earnings by language ability of Hispanics clearly show that bilinguals earn more than their

monolingual English counterparts in certain areas of the US. These results hold true for

Hispanics in three metropolitan areas in the US: Miami, El Paso, and, to a smaller extent,

Chicago. While their findings leave little doubt of the positive association between bilingualism

and earnings in these areas, there is a need to go beyond this general observation by controlling

for the effects of education, citizenship status, years in the US, and other select variables in order

to observe pure effect of bilingualism on earnings.

As this literature review illustrates, there is a large body of research dedicated to the

study of the economic impact of language proficiency. However, inconsistencies within this









research about the effect of language enclaves on returns to English-proficiency and bilingualism

make these topics ripe for further study. Data from the 2000 US Census provide a number of

variables that permit detailed exploration of these topics. The methodology and variables used to

achieve this analysis are described in detail in the next chapter.









CHAPTER 3
DATA AND METHODOLOGY

Research Design

Building on the theoretical foundations and substantive research highlighted in the

previous chapters, my hypothesis predicts that English language ability will be a strong

determinant of earnings in both areas with high concentrations and areas with low concentrations

of Spanish-speakers, even after controlling for education and other variables that may influence

earnings. However, adopting McManus's (1990) theoretical framework and reasoning, my

primary hypothesis asserts that the labor market context will affect the relationship between

language and earnings. That is to say, among Hispanics in Florida, English language ability will

have a stronger impact on earnings in areas with a low percentage of Spanish-speakers. In areas

with large minority language enclaves, where there are high percentages of Spanish-speakers,

English language ability will be less relevant to earnings; the labor market in these areas will

place less value on English language ability.

My corollary hypothesis follows this same line of reasoning to test the effect of bilingual

English-Spanish language skills on earnings. It posits that the inverse holds true for bilingual

language skills. In areas with large Spanish language enclaves, returns to bilingualism will be

higher than returns to bilingualism in areas dominated by the majority language.

To test these hypotheses, my study consists of two parts: the first based on statistical

analysis of census data and the second based on qualitative interviews. The following sections

detail this methodology.









Quantitative Analysis


US Census Data

The US Census 2000 serves as the dataset for this study. Specifically, the study utilizes

the 5% Public Use Microdata Sample (PUMS) for the state of Florida. This random sample

contains nearly 796,500 cases, weighted to represent the entire population. The census collects

information on both households and individuals on topics related to income, language ability,

ethnicity, occupation, and other variables such as age, race, education, that provide a basis for

testing variations in the relationship between language abilities and earnings.

Since 1890, every decennial US census (with the exception of the 1950 census) has

included at least one question pertaining to respondents' language characteristics. Stevens (1999)

categorizes the history of census language questions into three clusters: the earliest cluster,

focusing on English proficiency; the middle cluster, focusing on mother tongue, and the current

cluster, focusing again on English proficiency. The shifting focus of the questions reflects

changing perceptions of the relationship between language and ethnicity (Stevens 1999). For

example, the first two clusters of language questions were primarily designed to elicit

information on the ethnic and racial characteristics of growing immigrant populations. For that

reason, the language questions on the early and mid-century censuses pertain only to white

foreign-born respondents and/or members of select immigrant groups (Stevens 1999). Only the

most recent censuses (1980-2000) record both the native language characteristics and English

proficiency levels of the entire population.

The Census 2000's questions on language are particularly useful for this study because

they focus on both English language proficiency and language spoken at home, and they include

the language characteristics of both native and foreign-born respondents. The Census 2000 uses a

three-part series of questions to gather information on respondents' language abilities: 1) Does









this person speak a language other than English at home? 2) What is this language? 3) How well

does this person speak English? (Very well, Well, Not well, Not at all). The Census Bureau

included this series of questions specifically to identify geographic areas with a large population

of people with limited English language abilities in order to better assess needs for bilingual

education and other social services (Stevens 1999). The following section details how I use

these language variables to construct the linguistic concentrations for this study.

Measuring the Language Enclave

Due to the comparative nature of this study, it was first necessary to identify areas with

high and low proportions of Spanish-speakers. The Census 2000 offers several levels of

geographic disaggregation, ranging in smallest to largest order from block, block groups, and

census tracts to county, state, regional, and national divisions. However, in the 5% PUMS files,

the Census Bureau collapses the smaller geographic units in order to maintain the anonymity of

respondents. Thus, in the 5% sample, the Public Use Microdata Area (PUMA) is the smallest

identifiable geographic unit, containing 100,000 or more respondents. The next level of

aggregation is the Super-Public Use Microdata Area (Super-PUMA). The state of Florida

contains 32 Super-PUMAs, each containing populations of 400,000 or more.

The Super-PUMAs serves as the primary geographic unit of interest for this study. While

they do not offer analysis at the most disaggregate level, they serve the basic purpose of this

study, and their larger size offers a distinct advantage over the PUMA. Since the Super-PUMAs

typically cover an area that encompasses one to four PUMAs, and corresponds to one or more

counties as shown in the map in Figure 3-1, the chances that a respondent's place of work lies

within the Super-PUMA in which he lives are greater than the chances that a respondent's place

of work lies within the smaller PUMA in which he lives. This reduces the need to construct a









variable that proxies place of work and partly circumvents potential bias that may arise from

wage differentials due to location.

To determine areas with a high proportion of Spanish-language speakers, I selected

broadly for both Spanish-speakers and monolingual English speakers and then cross-tabulated

language by Super-PUMA. This tabulation allowed me to identify which Super-PUMAs contain

large populations of Spanish-speakers and which contain populations that are primarily English-

speaking. I then collapsed the Super-PUMAs into two categories: 1) high concentration of

Spanish-speakers (HC) and 2) low concentration of Spanish-speakers (LC). I assigned Super-

PUMAs with a Spanish-speaking population of 40% or less to the Low Concentration category

and Super-PUMAS with a Spanish-speaking population of 60% or more to the High

Concentration category. Table 3-1 shows a simple cross-tabulation of language spoken at home

by proportion of Spanish-speakers (High or Low). In the LC area, less than 10% speak Spanish;

the remaining 90% are English-speakers. In HC area, almost 63% speak Spanish, while only

37% speak English. To give a geographic representation of the linguistic concentrations, Figure

3-1 presents a map outlining the Florida Super-PUMAs, color-coded to show the linguistic

concentrations used in this study. It is obvious from the map that the Spanish-language enclave,

or HC area, corresponds exactly to the Super-PUMAS that compose Miami-Dade County. The

remaining Super-PUMAs in Florida constitute the LC area. With the linguistic concentration

measurement in place, I then invoked several criteria to select the cases to be included in the

regression analysis.

Sample Description

The sample used in the regression analysis consists of 16,611 cases (n=16,611) of Hispanic

males aged 18-65. The age restrictions account for those respondents in their prime working

years, while the sex restriction eliminates the bias created by a gender-related wage gap. I further









limit the sample to respondents who report themselves as employees, due to the well-

documented complications with measuring the wages of the self-employed (McManus 1990). In

order to prevent potential bias caused by the inclusion of wages from part-time work or

intermittent stints in the labor force, the sample includes only full-time employees, defined as

those who reported working an average of 35 hours or more per week and 45 weeks or more in

1999. Finally, I select for employees who work within three broad categories of occupations:

Management and Professional Positions; Sales; and Service, since occupations within these

categories rely heavily on on-the-job language use. (The criteria for these occupational

categories are detailed in a later section of this chapter).

To secure a sample of Spanish-speakers, I selected those respondents who reported

speaking a language other than English at home, and specifically, those who reported speaking

Spanish at home. For the second research question, which compares the earnings of bilingual

Spanish-English speakers and monolingual English speakers, I expanded the above sample to

include those respondents who reported speaking English only. I define bilinguals as those

respondents who report 1) speaking Spanish at home and 2) speaking English "very well."

Description of Variables

Dependent variable

The Census 2000 includes information on several different types of income, such wage and

salary, interest, dividend, and rental income as well as total personal income. I select only wage

and salary income as the dependent variable for two reasons. First, wage and salary income

refers specifically to income earned by working. Second, income from wages and salary tends to

be less subject to underreporting than other forms of income (US Census Bureau 2005). Since

wage and salary income tends to be documented and is generally received in consistent amounts









throughout the year, a respondent is more likely to accurately report this type of income as

opposed to income earned through other sources.

Following standard practice, I use the natural log of wage and salary income as the

dependent variable in the regression analysis. The log form of income is generally preferred in

regression analyses for two reasons. First, the logged value, by bringing outlier values closer to

the regression line, generates better estimates. Second, when using the natural log of income, the

'b' coefficients in the regression analysis can be interpreted as percentages of returns to earnings

(Lovell 1989; Hardy 1993).

Independent variable

English proficiency is the primary independent variable. Respondents self-report their

English-speaking ability by choosing from four possible responses: not at all; not well; well; and

very well. In the regression model, rather than inserting the English-language variable as a

single variable in the form of an index (0-3), I code those responses into three dummy variables:

Not well (not well=1; otherwise=0); Well (well=1; otherwise=0), and Very well (very well=1;

otherwise=0). Those who reported not speaking any English-the Not at all group-are used as

the comparison group. Converting the English-language ability variable into multinomial

dummy variables allows me to observe how the effect of English language ability on earnings

varies by proficiency level. In the second regression analysis, which seeks to compare the

earnings of bilinguals and monolinguals, I insert monolingual (English-only) Hispanics into the

sample, and use them as the reference group. I then use the following four dummy variables for

English speaking ability: Not at all (not at all=l; otherwise=0); Not well (not well=1;

otherwise=0); Well (well=l; otherwise=0), and Very well (very well=l; otherwise=0).









Control variables

Selection of the control variables used in the analysis accounts for personal characteristics,

human capital traits, migration circumstances, and labor characteristics that are likely to

influence wage and salary income.

The first set of control variables represents common human capital traits, including age,

educational attainment, work experience, and work experience-squared. Age refers to the

respondent's age in years. Educational attainment represents the respondent's highest level

education completed. An individual's labor market experience also tends to be a strong human

capital trait that is positively correlated with earnings (Chiswick 1991). However, since the

Census does not provide a variable that explicitly measures the number of years the respondent

has been in the labor force, I construct the standard proxy measure for work experience by

computing the respondent's age less his educational attainment less 6 years (age-educational

attainment-6) (Mincer 1974).

The next set of control variables corresponds to personal characteristics, and includes

race, nationality, and linguistic isolation. The Census Bureau records seven mutually exclusive

categories for race, including a Two or more races category introduced in the 2000 Census. For

the purposes of this study, I have collapsed these categories into "White" and "Non-white" in

order to use race as a binomial dummy variable (White=l and Non-white=0).

Since previous research has shown that nationality can act as a social capital trait that

strongly affects earnings (Borjas 1982, Tienda 1983), I incorporate the following multinomial

dummy variables for national origin: Puerto Rican (Puerto Rican=l and otherwise=0); Mexican

(Mexican=l and otherwise=0); and Other Hispanic (Other Hispanic=l and otherwise=0). Since

Cubans comprise the largest group of Hispanics in Florida, they are used as the reference group

for comparison. Finally, I include linguistic isolation, a variable that indicates a respondent's









exposure to English in the home. The Census Bureau defines a linguistically isolated household

as one in which no members aged 14 or older speak English only and no members aged 14 or

older speak a non-English language and speak English "very well." That is to say, all members

aged 14 or older in linguistically isolated households experience some degree of difficulty with

English. I code this as a binomial dummy variable (not linguistically isolated=1 and

linguistically isolated=0).

The control variables for migration characteristics are bit t/lait. e, citizenship status, and

years in the US. Prior research finds that US origin, US citizenship, and length of residence in

the US are positively associated with earnings of non-native English speakers (Chiswick and

Miller 1992; 2002). I thus insert dummy variables for hbit tll/q/ie (US=1 and abroad=0) and

citizenship status (US citizen=1 and non-US citizen=0). Years in the US represents the total

number of years that a respondent has lived in the US.

The final set of control variables accounts for variations in respondents' labor

characteristics. Since wage income often varies by number of hours and weeks worked, I

include the usual number of hours worked per week and the total number of weeks worked in

1999. The third labor characteristic variable accounts for occupation-related wage differentials.

The Census Bureau recorded 992 different types of occupations in the 2000 Census. In order to

convert these into more manageable categories, I follow the Census Bureau's classification

scheme and collapse the occupations into the seven principal categories: Management and

Professional; Service; Sales and Office; Farming, Fishing, and Forestry; Construction,

Extraction, and Maintenance; Production and Transportation; and Military. However, as noted in

the sample description, I select the three occupational categories that most use language on the

job: management and professional positions; sales; and service. I then construct multinomial









dummy variables for these occupations by coding sales (sales=l and otherwise=0) and service

(service= 1 and otherwise=0). Professional and management positions, which tend to be the

highest-paid occupations of the three categories, are left out of the equation and used as the

comparison group.

Statistical Model/Data Analysis

A simple Ordinary Least Squares (OLS) regression analysis allows me to observe the

effect of English language ability on the natural log of income, net of the effects of education,

age, sex, race, hours and weeks worked, occupation, work experience, national origin,

citizenship status, years in the U.S., and linguistic isolation. The analysis of the relationship

between earnings and the aforementioned wage-predictors is modeled in Equation 3-1.

Y =a+ Plage + p22racei + 3educ+ 04exp + 15exp2 + 36weeks + p7hours+ 3Pocci + 39bpl +
Piocitizen + p11years+ Pl21ingisol + P13hispani + P14engabili (3-1)

In this equation, Y represents the natural log of wage and salary income; age is the

respondent's age in years; race, 's refers to binary racial categories; educ represents highest level

of education attained; exp is the proxy for work experience; exp is work experience-squared;

occ, 's refer to different occupational categories; weeks is the number of weeks worked in 1999;

hours is the usual number of hours worked per week; bpl indicates a respondent's birthplace;

citizen indicates a respondent's citizenship status; years represents the number of years a

respondent has lived in the US; lingisol indicates whether or not the respondent lives in a

linguistically isolated household; hispan, 's represent national origin categories; and engabil, 's

represent different levels of English-speaking ability.

Regressions are conducted separately for the two categories of Super-PUMAs. A

comparison of the regression models for the high and low concentrations will allow me to

determine if the effect of English language ability on earnings varies according to the percentage









of Spanish-speakers in the area. Since I argue that English language ability will be a more

important determinant of earnings in areas where there are fewer Hispanics, I expect the effect of

English language ability on earnings to be greater in areas with a low concentration of Spanish-

speakers than in areas with a high concentration of Spanish-speakers. Specifically, I predict that,

net of the controls for occupation and human capital variables, the 'b' coefficient for English

language ability will be significantly higher in the low concentration model than in the high

concentration model. In the second regression analysis, which examines the effect of

bilingualism, I expect the 'b' coefficient for to be higher for bilinguals in areas with a high

concentration of Spanish-speakers than in areas with a low concentration of Spanish-speakers.

Qualitative Analysis

Valuable as the empirical analysis may be in terms of the magnitude of effects, and how

the effects vary by labor market context, it is also useful to examine the behavioral and

attitudinal models that underlie my hypothesis. Specifically, it is important to verify that

employers give less importance to workers' language abilities in enclave economies. In order to

do this, I added a qualitative component to this study.

The qualitative analysis consisted of interviews with employers and managers who make

hiring and promotion decisions and frontline workers from employment and staffing agencies

familiar with the demands of the local labor markets. The sample consisted of informants

associated with businesses in the service sector, since such businesses rely heavily on oral

communication with the public. After establishing initial contacts, I used snowball sampling to

identify subsequent key informants.

Through semi-structured interviews, I gained information about the behavioral and

attitudinal patterns associated with language and earnings. Specifically, I strove to elicit

information about how employees' language proficiency influences employers' decisions









regarding earnings and how and why criteria for such decisions vary by labor market context.

The general interview guide consisted of questions related to employees' language ability and

work performance; language-based delegation of positions and tasks; payment, raise, and bonus

criteria; language proficiency vs. language use; linguistic composition of workforce and

clientele; language-based discrimination in the workplace; and bilingualism in the workplace.

The conversational nature of the semi-structured interviews facilitated the flow of ideas

regarding these topics and allowed others to emerge. I conducted a total often interviews in

Miami-Dade and Broward counties.

Although the qualitative analysis played a secondary role in this study, it nonetheless

provided further insight into the relationship between language and earnings and was particularly

helpful in identifying new variables to include in the quantitative analysis. Consequently, I do

not discuss or analysis the interviews at length in the proceeding chapters, but instead include

small excerpts as supporting material for the quantitative findings. To maintain anonymity, the

names of the informants have been changed and their place of employment has been concealed.

Methodological Caveats

Despite the high quality of the Census 2000 (which is reportedly the most accurate US

census in history) and the applicability of its questions to this study, it is nevertheless important

to interpret findings with the following caveats in mind.

Wage and salary income, the dependent variable, is self-reported by respondents. Since

answers to the income questions are generally based on memory, the Census Bureau (2005)

cautions that respondents often forget precise amounts of income received. This is especially

true in the case of undocumented and informal earnings, such as tips. As a result, wage and

salary income tends to be underreported.









English speaking ability, the primary independent variable of interest, also suffers from

three main limitations. First, like income, language ability is self-reported by the respondent. The

census operationalizes English language ability as speaking ability, a definition that is broad and

subject to interpretation. Since there are no explicit and uniform criteria for gauging language

ability, a respondent may claim any level of proficiency he wishes. Thus, English language

ability, unlike wage and salary income, tends to be overestimated by respondents (Siegel 2001).

The over-reporting of English language ability is more pronounced in the very well and well

categories (Stevens 1999).

The second limitation of the English-language ability variable is that it refers only to

English speaking proficiency. As with most recent surveys and censuses, the 2000 Census does

not ask about the ability to read English. The inclusion of a question on reading ability would

offer a more precise analysis since most occupations typically require both reading and speaking

skills (Chiswick 1991).

Third, the fact that the questionnaire is written in English presents an inherent problem

for respondents with limited English abilities. Each household must have at least one member

that has some English literacy in order to fill out the questionnaire or understand what alternate

resources the Census Bureau provides to assist non-English language speakers with the

questionnaire. The Census Bureau has implemented various procedural measures to reduce

undercount resulting from such language issues. For example, the census questionnaires are now

available in six languages, including Spanish. Language assistance guides are also available to

provide assistance with the completion of the questionnaire in 49 languages. Nevertheless, for

obvious reasons, this issue lingers as a potential source of undercount or misreporting.









The measurement of bilingualism also warrants a word of caution. Stevens (1999) notes

that many people who report speaking English very well and report speaking another language at

home, may not necessarily be completely fluent in the non-English language. This pertains

especially to children of immigrants, who often experience first-language attrition, or loss of a

second language. In addition, data from past censuses reveal that respondents have a tendency to

overreport the use of a non-English language at home. Over-reporting of language ability tends

to be more common among native-born Americans than among foreign-born residents (Stevens

1999). Problems with reporting the use of non-English languages may arise from ambiguities in

the phrasing of the census question "Does this person speak a language other than English at

home?" The question does not specify the extent and frequency with which the non-English

language is used at home. Additionally, "at home" may also be a source of ambiguity, especially

for immigrants who may interpret "home" as home country (Siegel 2001).

The issue of undercount again arises with the Hispanic variable. The census undercount

of Hispanics is well-documented (Duany 1992; Evans 2001). However, the 2000 Census boasts

the lowest reported levels of undercount of minority groups (Evans 2001). Indeed, the

undercount estimates of Hispanics dropped from 4.99% on the 1990 Census to 2.58% on the

2000 Census (Evans 2001). Researchers have theorized about a variety of cultural and

behavioral factors that may contribute to this undercount. Duany (1992) posits that undercount of

Hispanics often stems from five main causes:

(1) disbelief in the confidentiality of the census
(2) distrust of government authorities
(3) fear of losing public assistance
(4) fear of deportation among undocumented immigrants
(5) cultural differences in defining household structure (p.1)









Other research has shown that the placement of the census question on Hispanic origin also

influences the count. The 2000 Census marked a change in the sequencing of questions on race

and Hispanic origin. In 1990, the question on Hispanic origin directly preceded the question on

race; the 2000 Census reversed the order of these questions (Grieco and Cassidy 2001). Analysis

of this sequencing shows that the order of questions used in 2000 significantly reduced the non-

response of Hispanics (US Census Bureau 1999; Grieco and Cassidy 2001).

Another minor limitation of this study stems from the measurement of linguistic

concentration. While other studies have invoked the Hispanic variable (McManus 1990), the

linguistic isolation variable (Chiswick and Miller 1992), and other methods to identify language

enclaves, this study defines linguistic concentrations more broadly by using simply the

percentage of Spanish-speakers who live in a certain area, regardless of their linguistic isolation

status. While this may be a rather crude measurement of language enclaves, it nevertheless

serves the basic purposes of this study.













Table 3-1. Language spoken by concentration of Spanish-speakers, Florida 2000
Concentration of Spanish-speakers Language spoken Total
Spanish English
Low
N 59,347 546,704 606,051
% 9.8 90.2 100.0

High
N 59,149 35,272 94,421
% 62.6 37.4 100.0

Source: US Census 2000 IPUMS 5% Sample














B" 8T 8s 85 84"' 83 82


12051 A


LEGEND
12102 SuperPut
Slale
ADMS Conty
---- Shwellne


Figure 3-1. Florida linguistic composition.
Census Bureau, Census 2000.


Public Use Microdata Sample (PUMS) files. US


SAreas with low concentration of Spanish-speakers (LC)
Areas with high concentration of Spanish speakers (HC)









CHAPTER 4
RESULTS

This chapter presents the findings from the quantitative analysis and qualitative interviews.

I first outline descriptive statistics for the variables used in the analysis and then report the results

from the OLS regression analyses for both research questions. Wherever relevant, I include

excerpts from the interviews as supporting evidence of the empirical results.

Summary Statistics

Tables 4-1 and 4-2 provide the sample means and standard deviations for the dependent

and independent variables used in the regression analysis for the Low Concentration (LC) and

High Concentration (HC) models, respectively. With the exception of certain variables

discussed in further detail below, the statistics show little variation between the two contexts.

The annual wage and salary income in the LC areas averages $30,717 (natural log

10.101). The mean wage and salary income in the HC area is slightly higher at $35,507(natural

log 10.221) per year. The distribution of English-speaking abilities, however, is fairly similar

across the two concentrations. In both areas, just over half of the respondents report speaking

"very well." Approximately 20% of respondents in each concentration report speaking "well,"

17% report speaking "not well," and about 9% report speaking no English at all.

The mean and standard deviations for the human capital, labor characteristics, and

migration variables are fairly uniform across the concentrations. The Hispanic origin

composition does, however, vary by concentration. Cubans compose the majority (52.3%) of the

sample in HC area, while Mexicans, Puerto Ricans, and other Hispanics rank as the most

populous Hispanic origin groups in the LC areas. The percentage of foreign-born respondents

also is greater in the HC areas.









Research Question One: The Effect of Language Enclaves on Returns to English

The first models regress the log of wage and salary earnings on English-speaking ability of

non-native speakers. Table 4-3 displays the results from the regression models, sorted by

linguistic concentration. The b coefficients, reported in the first and third columns, show the

effect of English language ability in the two linguistic contexts. Since the dependent variable is

logged, the b coefficient is interpreted as percent of change in the dependent variable (Y) for

each unit change in the independent variable (X), when X is a continuous measure such as age or

years of school. In the case of dummy variables, such as those used for English-speaking ability,

the b coefficient is interpreted differently since the dummy variables have discrete values of 0

and 1 (Hardy 1993). In this instance, the antilog of the regression coefficient is used to measure

the percent of change associated with belonging to the category of interest, relative to the omitted

reference category.

As expected, results from these models point to a positive relationship between English-

speaking ability and earnings in both English-dominant and Spanish-dominant areas in Florida,

as indicated by the positive regression coefficients in both models. An initial observation

confirms the importance of treating English language proficiency as a set of discrete dummy

variables, rather than an ordinal variable with values ranging from 0-3. Had I done the latter, and

thereby generated a single 'b' coefficient for language proficiency, it would not be evident that

the returns to language ability differ by proficiency level. For example, the findings show that

there are no significant differences in the earnings returns to language proficiency for

respondents in the "not well" and "not at all" categories. However, when compared to the "not at

all" category, the earnings returns to language proficiency are greater for those in the "well"

category and greater still for those in the "very well" category. In short, the findings clearly









show that the effect of language ability on earnings is nonlinear, a pattern that is evident in both

labor market contexts.

When examined comparatively, the two models show unexpected results. The b

coefficients for English-speaking ability variables in the HC model are higher than the b

coefficients in the LC model. For example, in the LC model, the b coefficient (column 1) for

those respondents who speak English "well" is .052. Those who speak English "well" earn

roughly 5.2% more than the reference group, who speaks no English at all. In the HC model,

the b coefficient (column 3) for those respondents who speak English "well" is .076. This means

that those who speak English "well" earn approximately 8% more than those who speak no

English at all. The comparison of these two regression coefficients indicates that, net of the

effects of the selected control variables, the impact of intermediate English proficiency on

earnings is greater in the HC areas.

The difference in the magnitude of the effect of English ability on earnings is even more

pronounced for those have a higher level of English-speaking proficiency. In the LC areas,

respondents who speak "very well" earn approximately 14.3% more than the reference group, as

indicated by the b coefficient of .134. However, in the HC model, those who speak "very well"

earn nearly 25% more than their English-deficient counterparts, as indicated by the b coefficient

of .219. Only the Not well group is discounted from the analysis since its relationship with

earnings does not prove statistically significant. These findings suggest that, in the case of the

"well" and "very well" proficiency levels, English has a greater effect on earnings in the HC

areas. In other words, in the labor market in Florida, English is actually more important in areas

with a high concentration of Spanish-speakers. Such findings run contrary to my research

hypothesis and contradict McManus's (1990) study that found that larger enclave size lowers









returns to earnings and Chiswick and Miller's (2002) study that concluded that non-native

English speakers have greater earnings opportunities in areas with a high concentration of

Spanish-speakers.

Many comments from interviews in Miami-Dade support these findings. For example,

Dan, a recruiter at a staffing agency in North Miami, notes that, although most of his job

applicants are bilingual:

English is still very important in Miami. A lot of the clients that we deal with are not in
the Latin or Spanish-speaking areas; they're mostly English-speaking, so they need to be
able to communicate with these guys and let them know what needs to get done. If they
can't communicate with them, they are going to send them off the job.

Victoria, a recruiter for a hotel and food service employment agency that staffs companies

in several counties in South Florida, seconds this opinion. "English is a must," she says. She

explains that although all of the companies she staffs require English-speaking workers, her job

applicants are mostly Spanish or Creole-speakers. She adds:

Sometimes they don't even know a little bit of English; they cannot answer me. They have
a companion with them to talk with me. I tell them, how can I communicate with them,
when I will be calling them personally to send them on a job? Usually what I give them are
dishwashing jobs... if they're insistent. And sometimes jobs in the hotels, if the hotel
manager or supervisor speaks Spanish. But if not, they will tell me, 'Please don't send this
person, because we can't communicate'... Those jobs are usually the lowest paid,
minimum wage.

Steve, a recruiter at a staffing agency in Miami, likewise emphasizes the importance of

English in the workplace. In addition to job experience, references, work ethic, and

presentability, he cites English as one of the top qualifications that he seeks in applicants. "They

absolutely have to have some proficiency in English. He explains further that language

proficiency indirectly influences pay because communication affects job performance, and job

performance is what ultimately determines raises and bonuses.









Indeed, relative to the other independent variables in the model, English language

proficiency ranks among the most important determinants of earnings. To demonstrate this, I

call attention to the standardized Beta coefficients, which offer a means of comparing the effect

English ability to the effect of the other independent variables on earnings. Since these variables

serve primarily as controls, I do not discuss their impact at length. Rather, I mention briefly the

magnitude of their effects in order to give a general idea of the relative impact of English

proficiency on earnings.

Similar to findings from past studies (eg. Chiswick and Miller 2001, Tienda 1983, Borjas

1982), both models in Table 4-3 confirm the positive correlations between earnings and the

standard human capital variables: age, education, and work experience-squared1. However, in

the HC model, the ability to speak English very well actually exerts a greater influence on

earnings than educational attainment. In fact, with a Beta coefficient of .165, the ability to speak

English very well functions as the fourth most important determinant of earnings, behind age

(Beta=.731); work experience-squared (Beta= -.655); and hours worked (Beta=. 177). Though

its relative impact is lesser in the LC model, the ability to speak English very well still ranks as

an important determinant of earnings.

When compared to English-speaking ability, the other control variables exert a less

powerful, but nonetheless significant, influence on earnings. For example, US nativity, US

citizenship, and length of residence in the US are all positively associated with earnings. Type

of occupation also acts as a significant predictor of earnings. Workers in sales and in the service

industry tend to earn less than those in professional positions. Such findings resonate with the

aforementioned excerpt from the interview with Victoria, in which she indicates that workers


1 Collinearity exists between the age and work experience variables.









with low levels of English proficiency are largely relegated to the lowest-paying jobs in the

service industry.

The models also show evidence of wage discrimination along racial and ethnic lines. In

both the HC and LC areas, White Hispanics tend to earn about 4.7% more than Black or other

Hispanics. However, in relative terms, race is one of the least important determinants of

earnings, as indicated by the Beta values of .037 and .023 in the second and fourth columns,

respectively. Ethnic networks, however, do appear to have a strong impact on earnings. Indeed,

the perceived effect of ethnic networks emerged as a recurrent theme in several interviews. The

following quote from Raul, an administrative personnel manager for a local university, illustrates

one possible rationale for the relationship between immigrant and/or ethnic networks and

earnings and how language proficiency factors into that relationship. When asked about his

willingness to hire workers with limited English proficiency, his rationale recalls his own

immigrant past:

I look at myself: when I came to this country, someone gave me a chance and I was able to
move up through the system, and I like to do the same for people that are out there in the
same position that I was thirty-some years ago.

National origin-which serves as an operational definition of ethnic networks- in

particular stands out as a significant determinant of earnings. The results from both regression

models suggest that, among Hispanics, being Cuban is an added advantage in the labor market in

Florida. The negative regression coefficients for the Hispanic origin variables indicate that, all

other things being equal, Cubans earn more than other Hispanics in Florida. Compared to Puerto

Ricans, Florida's second most populous Hispanic group, Cubans earn about 10% more in the HC

areas and 14% more in the LC areas. Anna, a former career counselor and a current immigration

lawyer at a legal services organization in Downtown Miami, observes:









Sometimes I see Cubans come in; they hardly speak any English and they have only been
here for a short time, but they already have great jobs, working at FIU, at hospitals,
offices...

Her comments hint at the ways in which national solidarity can function as a social capital trait

that yields positive labor market outcomes.

Together with the English-speaking ability variables, these additional independent

variables form a statistical model that explains almost 36% of the variance in earnings in the LC

areas (R2=.356) and 33% of the variance in earnings in the HC areas (R2=.331). Both the

empirical evidence from these models and qualitative support from the interviews underscore the

importance of English proficiency in the labor market, regardless of linguistic concentration.

However, many interviewees, both in HC and LC areas, also highlighted the simultaneous

importance of Spanish in the labor force. The next section presents the findings that from the

models comparing the effects of bilingualism on earnings in each linguistic concentration.

Research Question Two: Returns to Bilingualism

The second set of regression analyses incorporates monolingual English-speaking

Hispanics in the sample in order to examine how the effect of bilingualism on earnings varies by

linguistic concentration. This analysis uses the same regression equation used to test the first

research hypothesis, but uses monolinguals as the reference group. The Very Well category in

both tables refers to bilingual speakers, since bilingual respondents are defined as those

respondents who 1) speak Spanish at home and 2) speak English "very well." Table 4-4

presents the results, sorted by linguistic concentration. In this analysis, the b coefficient again

serves as the main tool for comparison between the two concentrations. The LC model suggests

that monolingual English speakers earn more than their bilingual counterparts, as indicated by

the b coefficient of -.03 for the Very Well category in Column 1. Yet, findings in this case are









not statistically significant, and thus cannot offer valid suggestions about the effect of

bilingualism on earnings in LC areas in Florida.

The HC model, however, shows strikingly different results. As predicted, bilingual

language skills are positively associated with earnings in the HC area, as indicated by the b

coefficient of .066 in the Very Well category in Column 3. All other factors being equal,

bilingual Hispanics earn more than their monolingual English counterparts in the enclave.

However, the positive relationship between bilingualism and earnings only holds true for the

fully bilingual speakers. Spanish-speakers who have intermediate English proficiency earn less

than monolingual English Hispanics, as shown by the negative b coefficient for the Well

category in column 3.

Comments from the qualitative interviews echo these empirical findings. For example,

Kevin, a manager of an administrative and technical staffing agency in Hialeah, FL, notes that

"more often than not" clients (employers) in the South Florida area specifically request bilingual

English-Spanish speakers for positions. The applicant pool at his agency consists of

monolingual English speakers, monolingual Spanish-speakers, and bilingual English/Spanish-

speakers. He explains that forjobs that require little interaction with the public, monolingual

language skills, Spanish-only or English-only, suffice. However, for most positions, bilingual

skills are preferred since the businesses he staffs deal not only with bilingual clients in South

Florida, but also with monolingual Spanish-speakers in Latin America and the Caribbean. He

adds:

In Miami, bilingualism is pervasive. A good command of the English and Spanish
languages is necessary. Generally, bilingual positions pay higher because there is more
responsibility involved. Sometimes it creates issues with our recruiting efforts... we end up
ignoring a lot of folks who do not speak Spanish. There are those people that are English-
only who would not qualify. I had to learn Spanish myself when I came to Miami... I use
it everyday and it allows me to perform my job better. Yet, when you are limited to the









minority language, (which I don't know if Spanish is anymore in Miami), but, in my
experience, Spanish-only would generally pay lower than English-only. So, it goes both
ways, and may be something that is unique to this area [Miami-Hialeah].

This excerpt highlights the labor market advantages bilingualism and disadvantages of

monolingualism in the language enclave.

Anecdotal evidence from a recent Miami Herald also points directly to the advantage that

bilinguals have in the Miami-Dade labor market. The article quotes a Miami business executive

about the language abilities one of his former Cuban-American employees "Professionally, she

was very good. But, she was almost incapable of writing Spanish." He eventually replaced her

with a fully bilingual Puerto Rican secretary (Fernandez 2008). His experience shows the need

for employees with high proficiency levels of both English and Spanish.

Even Victoria, whose quote in the previous section stressed the importance of English in

the labor market, and who later expressed her belief that non-native speakers "should speak

English because they are in America," acknowledges the growing importance of Spanish in the

labor market in South Florida. She says, "I know a little bit of Spanish. You have to at least

learn the basics, because they [Spanish-speakers] are everywhere."

Thus, in this case, findings partially support the research hypothesis that predicted that

bilinguals would earn more than monolinguals in the enclave. Such findings run contrary to past

studies (Pendakur and Pendakur 2002) that conclude that knowledge of an unofficial, minority

language, garners no additional earnings advantage in the labor market. However, because

findings from the LC model were not significant, I cannot compare the effects of bilingualism on

earnings in the two linguistic concentrations. Nevertheless, these findings, as well as those from

the first research question, have important implications that I discuss further in the next chapter.











Table 4-1. Mean and standard deviation of variables used in Low Concentration model. Hispanic
males, age 18-65.


Variable
Wage & salary Income


Log of wage &
salary income


Age
Race


Description
Total annual wage &
salary income in dollars
Natural log of wage
& salary income


Mean
30717.2

10.101


Age of respondent
Dummy variable
(ref=Black/Other)


Educational attainment

Work experience
Work experience squared

Occupation
Management
Sales
Service


Weeks worked

Hours worked

Birthplace


Citizenship status


Years in the US

Linguistic isolation



Hispanic origin
Cuban
Mexican
Puerto Rican
Other Hispanic


Years of school

Age-education-6
(Age-education-6)2

Dummy variables
(ref=Management)



Total annual weeks
worked
Typical hours worked per
week

Dummy variable
(ref=Abroad)

Dummy variable
(ref=Non-US citizen)

Total years lived in the US

Dummy variable
(ref=Linguistically
isolated)

Dummy variables
(ref=Cuban)


32.784
1360.255


0.1831
0.1494
0.1777


51.18

45.12


12.32


.134
.299
.267
.301


Speaks English Dummy variables
None (ref=None)
Not well
Well
Very well
Source: 2000 US Census IPUMS 5% Sample: Florida
N=8848; Ref=Reference category; coded as 0


SD
30770.02


0.631


10.828
0.484


3.454

10.864
898.73


0.387
0.357
0.382


8.507

0.417


0.489


12.545

0.447


0.34
0.458
0.442
0.458


0.271
0.382
0.409
0.499











Table 4-2. Mean and standard deviation of variables used in High Concentration model. Hispanic
males, age 18-65.


Variable
Wage & salary Income


Log of wage &
salary income


Age
Race


Educational attainment

Work experience
Work experience squared

Occupation
Management
Sales
Service


Weeks worked

Hours worked

Birthplace


Citizenship status


Years in the US

Linguistic isolation



Hispanic origin
Cuban
Mexican
Puerto Rican
Other Hispanic


Description
Total annual wage &
salary income in dollars

Natural log of wage
& salary income


Age of respondent
Dummy variable
(ref=Black/Other)


Years of school

Age-education-6
(Age-education-6)2

Dummy variables
(ref=Management)



Total annual weeks
worked
Typical hours worked per
week

Dummy variable
(ref=Abroad)

Dummy variable
(ref=Non-US citizen)

Total years lived in the US

Dummy variable
(ref=Linguistically
isolated)

Dummy variables
(ref=Cuban)


Speaks English Dummy variables
None (ref=None)
Not Well
Well
Very Well
Source: 2000 US Census IPUMS 5% Sample: Florida, N:


Mean
35507.16


10.221


39.42
0.87


10.49


34.934
1360.255


0.253
0.2361
0.1261


51.14

44.68


15.80


.5228
.033
.0652
.38


.0920
.1735
.223
.511
=7763;


SD
37176.76


0.662


11.437
0.34


3.072

11.825
898.73


0.434
0.425
0.33198


1.875

8.143

0.332


0.497


12.624

0.446


0.5
0.179
0.247
0.485


0.289
0.379
0.416
0.499
Ref=reference category; coded as 0











Table 4-3. Returns to English-speaking ability. Log of income regressed on English ability and
other selected variables.


Independent variables

Constant

Age

Race (ref=Black/other)

Educational Attainment

Work Experience
Work Experience Squared

Occupation
Service
Sales

Weeks Worked

Hours Worked per Week

Citizenship Status

Birthplace (ref=Abroad)

Years in the US

Linguistic Isolation (ref=Not Isolated)

Hispanic Origin (ref=Cuban)
Mexican
Puerto Rican
Other Hispanic

English Ability (ref=Not at all)
Not Well
Well
Very Well


Low concentration
of Spanish-speakers

b Beta

6.904

.044* 0.762

.048* 0.037

.021* 0.117


0*


-.158**
-.038*

.026*

.011*

.060*

.100*

.005*

0.028*


-.133*
-.153*
-.60*


0.012
.052**
0.134*


-0.64


-0.096
-0.022

0.077

0.153

0.047

0.066

0.108

0.02


-0.096
-0.107
-0.044


0.007
0.034
0.106


High concentration
of Spanish-speakers

b Beta

6.957

.042* 0.731

0.044** 0.023

0.018* 0.082


0*


-.241*
-.098*

0.022*

.014*

.091*

.135*

.007*

0.042**


0.024
-.106*
-.038*


-.039
.076*
.219*


-0.656


-0.121
-0.063

0.062

0.178

0.068

0.067

0.14

0.028


0.006
-.039
-0.028


-0.022
0.048
0.165


R2 0.356 0.331
N 8847 7763
Source: 2000 Census 5% IPUMS Sample: Florida; *Significant at .001 or less; **Significant at
.05 or less. Sample includes Hispanic males, age 18-65, who reported speaking Spanish at home











Table 4-4. Returns to bilingualism. Log of income regressed on language ability and other
selected variables.
Low concentration High concentration
of Spanish-speakers of Spanish-speakers

Independent variables b Beta b Beta
Constant 7.056 7.053

Age 0.036* .611 .031* 0.539

Race (ref=Black/other) .050* 0.038 0.042** 0.021

Educational Attainment 0.020* 0.108 0.030* 0.14


Work Experience
Work Experience Squared

Occupation
Service
Sales

Weeks Worked

Hours Worked per Week

Citizenship Status

Birthplace (ref=Abroad)

Years in the US

Linguistic Isolation (ref=Not Isolated)

Hispanic Origin (ref=Cuban)
Mexican
Puerto Rican
Other Hispanic

English Ability (ref=English only)
Not at all
Not Well
Well
Very Well


0*


-.158*
-.03**

.028*

.011*

.063*

.104*

.005*

0.028


-.118*
-.160*
-.063*


-.121*
-.113*
-.079*
-.03


-0.588


-0.094
-0.017

0.081

0.152

0.047

0.075

0.101

0.019


-0.084
-0.111
-0.046


-0.048
-0.63
-0.048
-.002


0*


-.240*
-.102*

0.022*

.015*

.099*

.144*

.007*

0.041**


0.024
-.097*
-.033*


-.138*
-.181*
-.072*
.066*


-0.466


-0.12
-0.065

0.063

0.179

0.074

0.075

0.143

0.027


0.007
-.037
-0.024


-0.059
-0.102
-0.044
0.05


R2 0.36 0.332
N 10404 8084
Source: 2000 Census 5% IPUMS Sample: Florida; *Significant at .001 or less; ** Significant at
.05 or less Sample includes Hispanic males, age 18-65, who reported speaking Spanish at home
or speaking English only.









CHAPTER 5
DISCUSSION

Chapter 4 presented regression estimates that suggest: first, that English is more important

to earnings in the minority language enclave; and second, that bilinguals earn more than their

monolingual English counterparts in the minority language enclave. The opinions expressed in

the qualitative interviews also largely support the empirical evidence. Together, these findings

offer valuable insight into the relationship between language, labor markets, and

immigrant/minority incorporation. Consequently, they pose important theoretical and policy

implications for these topics. This chapter discusses those implications in the following order:

implications for labor markets and incorporation theories; application of findings to current

policy issues; and suggestions for further research.

Theoretical Implications

Language, Earnings, and Labor Market Characteristics

Returns to English language proficiency

This study identifies the Super-PUMAS in Miami-Dade County as a Spanish language

enclave. The remaining Super-PUMAS in Florida qualify as areas with a low concentration of

Spanish speakers. Given that the majority (62%) of inhabitants in Miami-Dade County speaks

Spanish at home, intuitive reasoning would suggest that English is less important in the labor

market in this area. However, results reveal otherwise. The regression estimates indicate that

English proficiency has a greater impact on earnings in this area. Such findings contradict

McManus's (1990) and Chiswick and Miller's (2002) observed reduction in returns to English

language proficiency in areas with a high minority language concentration, but support Davila

and Mora's (2003) findings that English is more important to earnings in the labor market in

minority language concentrations. Although these results run contrary to my research hypothesis









and are seemingly counterintuitive, they are nevertheless statistically significant, and thus

provide insight into the relationship between language, earnings, and labor market

characteristics. They suggest that the relationship between language proficiency and earnings

does indeed vary by context, but not always in the predicted ways. The main question raised by

such findings then becomes: why does English language proficiency have a greater effect on

earnings in areas with a large Spanish language enclave?

One possible explanation could be rooted in labor economics theory. As the supply of

Spanish-speaking workers increases, the demand for their labor, and consequently, the value of

their labor, decreases (Bloom and Grenier 1992). Similarly, as the population of Spanish-

speakers increases in a given area, so does the competition among them in the labor market. In

an applicant pool filled with Spanish-speaking candidates with similar education and experience

qualifications, high English literacy may be the most important distinguishing factor among

them. English language proficiency would then assert itself as a more marketable trait in areas

with a high concentration of Spanish-speakers, and would thus be worth more within the enclave.

In areas with a low concentration of Spanish-speakers, where there is a smaller pool of Spanish-

speaker workers, the labor market cannot be as discriminating with regard to language skills.

Bloom and Grenier (1992) find a similar effect in their study of earnings of Hispanic males in the

US in 1970 and 1980.

A second, related explanation could point to effects of the Cuban ethnic enclave economy

in Miami. As previously mentioned, this study does not utilize the ethnic enclave economy as a

unit of analysis; rather, it focuses more broadly on linguistic enclaves. However, given the

prominence of the Cuban ethnic enclave economy in Miami, its presence may affect findings.

Some research suggests that while the ethnic enclave economy may have an insulating effect for









new immigrants, it may eventually serve as a mobility trap for immigrant employees (Sanders

and Nee 1987; Booth 1998). Assuming that the ethnic enclave economy acts as a mobility trap

and that it exists as a smaller unit within a larger linguistic or ethnic enclave, one could argue

that English is a more valuable asset in the areas with a high minority language concentration

because it serves as way to access better paying jobs outside the ethnic enclave economy. An

alternate strategy for future analyses may be to control for the presence of the ethnic enclave

economy within a larger ethnic or linguistic enclave. This measurement would require the

simultaneous inclusion of language and place of work variables to determine the chances that the

respondent works in the ethnic enclave economy (Zhou and Logan 1989).

A third possible explanation could be related to an element that, for lack of a

corresponding variable, is not considered by this study-public opinion of immigrants/foreign

language speakers. Past analyses have shown that public opinion toward immigrants affects the

incorporation process (Tienda 1983). Public sentiment toward immigrants/foreign language

speakers often varies according to the percentage of the population that is foreign-born in a given

area (Stolzenberg 1990). De Jong and Tran (2001) find Miami residents to be largely receptive

towards immigrants. Other studies, however, contend that the large foreign-born population in

Miami is a source of contempt among many city residents. For example, Portes and Stepick

(1993), through their use of competing discourses from the city's main ethnic groups, provide

ample evidence of the deep-rooted tensions between Hispanics (particularly Cubans), Blacks,

and Whites that have existed in Miami, at least in past decades. Given the high symbolic value

attached to English in the US, prejudiced employers may use lack of English language skills as a

basis for culture-based wage discrimination (Pendakur and Pendakur 2002). Given the large

foreign born population in Miami-Dade, it is possible that this type of discrimination occurs









more frequently in this area than in areas with a smaller percentage of immigrants. Further

research is needed to validate this line of reasoning.

Finally, methodological decisions may also explain results. First, the decision to include

foreign and native-born Hispanic males in my sample may impact results. Other similar studies

(Davila and Mora 2000; Hand 2006) run separate regression models for immigrants and natives

and find that the earnings returns to English language proficiency differ for the two groups. I

defend my decision to include both groups simultaneously since my primary focus is language,

not birthplace. Additionally, OLS regression analysis allows me to control for the effects of

birthplace. Nevertheless, I acknowledge that separate regressions for immigrants and native may

alter results. Second, results may also be sensitive to the linguistic concentration measurement.

As detailed in the literature review, past analyses have invoked other methods to measure the

linguistic enclave, such as Hispanic origin variable (McManus 1990). Again, since language is

the focal point of this study, I choose to define the enclave by the language spoken at home

variable. It is possible that different definitions of the enclave will produce different results.

Returns to bilingualism

Findings from second set of regression models support the hypothesis that bilinguals earn

more than their monolingual counterparts in the enclave. The importance of bilingualism in

Miami-Dade County also arose as a consistent theme in the qualitative interviews. Since findings

from the LC bilingualism model were not statistically significant, I cannot draw a comparison

between the two models regarding the effects of bilingualism on earnings in each area.

Nevertheless, the significant findings from the HC model on their own are unique and important

for two main reasons. First, most prior research has not shown that knowledge of an unofficial,

minority language to be positively associated with earnings (Pendakur and Pendakur 2002).









Second, they show that dual forms of linguistic capital are needed to maximize labor market

outcomes in the minority language enclave.

Immigrant/Minority Incorporation

This study finds that English is positively associated with earnings of Hispanics both inside

and outside the Spanish language enclave in Florida. It makes clear that throughout the state of

Florida there are economic incentives to learning English. This reinforces theories of majority

language proficiency as a valuable human, social, and cultural capital trait in the labor market;

for non-native speakers, an investment in learning English will clearly reap monetary rewards.

Additionally, as an aside, these findings may also serve to quell any nativist fears that English is

losing its importance in areas with a large minority language population, such as Miami (Portes

and Rumbault 1996).

The second set of findings does not fit as easily into traditional incorporation theories.

They show that although English matters in the enclave, Spanish does, too. In other words, there

are also economic incentives to maintaining or acquiring Spanish language skills. Since these

results contradict most previous studies of the economic value of bilingualism (Fry and Lowell

2003), they may indicate a shift in the value of bilingualism in the US labor market. In addition,

they highlight one way in which minority language speakers have at once adapted to and

transformed the linguistic landscape in Miami. These findings directly echo Portes and Stepick's

(1993) discovery of the impact that immigrants had on Miami. They note:

As sociologists, our principal focus was the adaptation of foreign-born minorities to their
new environment. As time passed, however, it became clear that the environment itself
was changing in ways that we could not have anticipated. The immigrants were
transforming not only themselves, but also the city around them (Portes and Stepick 1993:
xi).









Policy Implications

Two primary policy implications stem from this research. First, findings offer helpful

information about where English language training programs would be most valuable for non-

native speakers of English. Adding to the urgency of this issue, a recent New York Times article

underscored the growing need for government-funded English language training programs,

particularly in areas with large immigrant populations. As the foreign-born population increases

in many states, the waitlists for admission into free or low-cost government funded English

classes range from several months to two years (Santos 2007). In some cases, frustrated business

owners have taken matters into their own hands, teaching English to immigrant employees. As

Tara Colton, author of the Center for an Urban Future report, notes, "The issue of English

proficiency has become an issue of economic development" (Santos 2007: 4).

Secondly, there are important implications for bilingual education initiatives in Miami.

Findings from this study show that Spanish language skills are important resources in the labor

market in Miami. However, educational research links recent English-only initiatives and

federal legislation such as the No Child Left Behind to a decline in bilingual language skills and

a de-emphasis on bilingual education in Florida (SSTESOL 2005). A recent Miami Herald

article also laments the perceived decline of bilingualism in Miami. The article views the decline

of Spanish language skills among Hispanics in Miami as the "loss of an asset." A telling quote

from the article further highlights the importance of bilingualism in Miami. University of Miami

linguist Andrew Lynch observes, "Miami grew as a city along with the Spanish and bilingualism.

Bilingualism was the foundation of Miami as a global city" (Fernandez 2008). This study

provides additional evidence of the need to place greater emphasis on bilingual and Spanish

language education in Miami.









Suggestions for Future Research

The currency of the language and immigration debates in the US and the inconsistencies in

the research make this topic ripe for further exploration. This particular study could be further

expanded to explore a variety of angles. For example, the inclusion of non-Hispanic

monolingual English speakers into the sample would allow for deeper comparative analysis of

the economic value of bilingualism. Bilingual Hispanics may earn more than their monolingual

English Hispanic counterparts, but does the same pattern hold when comparing bilinguals to non-

Hispanics monolinguals?

Larger scale modifications of this study could also prove interesting. For example, the

application of these methods to a nationwide analysis would help draw conclusions about the

between language, earnings, and enclaves in the general US labor market. While language and

the labor market may interact in one way in Florida, their relationship may be different in other

areas in the US. This is particularly important since Miami is often viewed as a unique case in

urban sociology; it should not be considered "a microcosm of the American city" (Portes and

Stepick 1993). Thus, a nationwide analysis would be allow to see if such findings are unique to

Florida or if the pattern repeats itself in other states in the US and on a national scale.

In addition, longitudinal studies on the topic will offer further insight into the ways in

which the linguistic composition of the population affects the relationship between language and

earnings. Since census language questions have not been uniform throughout the last century,

there are certain complications involved conducting such a study (Stevens 1999). However,

since 1980, the language questions have remained the same and would thus be suitable for an

analysis of how the effect of English on earnings has changed in accordance with changes in the

immigrant population over the last several decades.









Longitudinal studies should also be used monitor the future effect of bilingualism on

earnings. The aforementioned Miami Herald article already cites anecdotal evidence of a decline

in bilingualism in Miami (Fernandez 2008). Language attrition has indeed revealed itself to be a

common pattern among children of immigrants in the US. As Portes and Rumbault (1996:11)

note, "The United States is unique in the rate at which other languages have been abandoned in

favor of English... in no other country have foreign languages been extinguished with such

speed." Thus, it is possible that Spanish language will lose prominence as second and third-

generation speakers become more acculturated and abandon Spanish in favor of English. If this

proves to be the case, the shift in linguistic preference will likely alter the effect of bilingualism

on earnings in the future.









CHAPTER 6
CONCLUSION

English language proficiency functions as a valuable form of human, cultural, and social

capital in the US labor market. Past research has shown English language proficiency to be a

uniformly positive determinant of earnings in the US. While individual human, cultural, and

social capital characteristics largely influence earnings, context also plays an integral role in

labor market outcomes. The primary theoretical proposition of this study assumed that labor

market context- specifically the linguistic profile of the labor market-determines the value of

language proficiency in the labor market. Despite the linguistic diversity in the US and the rich

data available from the decennial Census and other surveys, few recent studies address this topic.

Using data from the 2000 US Census, this study provided empirical evidence of the ways

in the effect of language proficiency on earnings varies according to the linguistic profile of the

labor market. The central research hypothesis posited that, among Latinos in Florida, English

language proficiency would be less important to earnings in areas with a large percentage of

Spanish-speakers than in areas with a small percentage of Spanish-speakers. The corollary

hypothesis argued that Spanish language skills would have a positive effect on earnings in areas

with a large Spanish language enclave, such as Miami-Dade County. However, results from the

regression models suggest that English language proficiency has a greater impact on earnings of

Latinos in areas with a large proportion of Spanish-speakers. The second set of results show that

while English language proficiency still plays a significant role in areas with a large Spanish

language enclave, Spanish language proficiency also has a positive effect on earnings in these

areas. Although lack of significance in the low concentration model did not permit a comparison

of the effects of bilingualism on earnings between the two linguistic contexts, the high









concentration model shows that, all other factors being equal, bilingual Hispanics earn more than

monolingual English counterparts in these areas.

These findings offer valuable insight into processes of labor market incorporation of

immigrants and non-native speakers. Like past studies, my study reinforces the economic

incentives for learning English. However, it differs from past studies in that it also offers

evidence of the economic value of learning or maintaining Spanish language skills in certain

contexts. Consequently, these findings have important policy implications for English language

training and bilingual education programs.

English remains the most widely spoken language in the US and as such, clearly reaps the

most monetary rewards in the labor market. However, as the Spanish-speaking population in the

US continues to grow, it is likely that, as this study has shown, Spanish will also assert itself as a

prominent force in the labor market. While the value of language proficiency clearly varies by

labor market context, there is no consensus within the literature regarding the ways in which it

varies. Thus, as the linguistic composition of the US continues to evolve, the study of the

relationship between language, earnings, and labor market context remains fertile ground for

additional research.









LIST OF REFERENCES


Becker, Gary. 1993. Human Capital:A Theoretical and Empirical Analysis, I ih Special
Reference to Education. Chicago: University of Chicago Press.

Bloom, David E. and Gilles Grenier. 1992. "Earnings of the French Minority in Canada and the
Spanish Minority in the United States." Pp. 373-409 in Immigration, Language, and
Ethnicity: Canada and the United States, edited by Barry R. Chiswick. Washington
D.C.: AEI Press.

Booth, William. 1998. "One Nation, Indivisible: Is it History?" Washington Post. February 22.
Retrieved March 1, 2008
(http://www.washingtonpost.com/wpsrv/national/longterm/meltingpot/melt0222.tm).

Borjas, George J. 1982. "The Earnings of Hispanic Immigrants." Industrial andLabor
Relations Review 35:343-353.

Bourdieu, Pierre. 1986. "The Forms of Capital." Pp. 241-58 in Handbook of Theory ofResearch
for the Sociology ofEducation, edited by J.E. Richardson. New York: Greenwood Press.

------. 1991. Language and Symbolic Power. Cambridge: Polity Press.

Carliner, Geoffrey. 1981. "Wage Differences by Language Group and the Market for
Language Skills in Canada." The Journal of Human Resources 16:384-399.

Camevale, Anthony P., Richard Fry, and B. Lindsay Lowell. 2001. "Understanding,
Speaking, Reading, Writing, and Earnings in the Immigrant Labor Market." The
American Economic Review 91:159-163.

Chiswick, Barry R. 1978. "The Effect of Americanization on the Earnings of Foreign-Born
Men." Journal of Political Economy 86:897-921.

------. 1991. "Speaking, Reading, and Earnings among Low-Skilled Immigrants." Journal of
Labor Economics 9:149-170.

Chiswick, Barry R. and Paul W. Miller. 1992. "Language in the Immigrant Labor Market." Pp.
229-296 in Immigration, Language, and Ethnicity: Canada and the United States, edited
by Barry R. Chiswick. Washington D.C.: AEI Press.

------. 1995. "The Endogeneity between Language and Earnings: International Analyses."
Journal ofLabor Economics 13:246-288.

------. 2002. "Immigrant Earnings: Language Skills, Linguistic Concentrations, and the Business
Cycle." Journal ofPopulation Economics 15:31-57.

Coleman, James. 1988. "Social capital in the creation of human capital." American
Journal of Sociology 94:95-120.









Davila, Alberto and Marie T. Mora. 2000. "English Skills, Earnings, and the Occupational
Sorting of Mexican Americans along the U.S.-Mexico Border." International
Migration Review 34:133-157.

DeJong, Gordon F. and Quynh-Giang Tran. 2001. "Warm Welcome, Cool Welcome: Mapping
Receptivity Toward Immigrants in the U.S." Washington, DC: Population Reference
Bureau. Retrieved March 1, 2008
(http://www.prb.org/Articles/2001/WarmWelcomeCoolWelcomeMappingReceptivityTo
wardImmigrantsintheUS.aspx).

Duany, Jorge. 1992. "The Census Undercount, the Underground Economy, and
Undocumented Migration: The Case of Dominicans in Santurce, Puerto Rico."
Ethnographic Evaluation of the 1990 Decennial Census Report #17. Prepared under
Joint Statistical Agreement 90-09 with the University of the Sacred Heart. Washington,
D.C.: Bureau of the Census. Retrieved February 22 2008
(http://www.census.gov/srd/papers/pdf/ev92-17.pdf).

Duany, Jorge and Felix Matos-Rodriguez. 2006. "Puerto Ricans in Orlando and Central Florida."
Policy Report 1, No. 1. New York: Centro de Estudios Puertorriquefios, Hunter College,
CUNY. Retrieved March 1, 2008
(http://www.orlando.org/clientuploads/hsummit/hsummitprcentralflorida.pdf).

Dustmann, Christian and Arthur van Soest. 2001. "Language Fluency and Earnings:
Estimation with Misclassified Language Areas." The Review of Economics and
Statistics 83:63-674.

------. 2002. "Language and the Earnings of Immigrants." Industrial and Labor Relations
Review 55:473-492.

Fernandez, Enrique. 2008. "The Spanish Spoken today by bilingual Hispanics in South
Florida has become watered down through generations." Miami Herald. March 1.
Retrieved March 1 2008
(http://www.miamiherald.com/news/breaking_dade/v-print/story/440517.html).

Fry, Richard and B. Lindsay Lowell. 2003. "The Value of Bilingualism in the U.S. Labor
Market." Industrial and Labor Relations Review 57:128-140.

Gonzalez, Arturo. 2000. "The Acquisition and Labor Market value of Four English Skills: New
Evidence from NALS." Contemporary Economic Policy 18:259-269.

Grieco, Elizabeth M. and Rachel C. Cassidy. 2001. "Overview of Race and Hispanic Origin."
Washington, DC: US Census Bureau, Retrieved February 22 2008
(http://www.census.gov/prod/2001pubs/cenbr01-1.pdf).

Grenier, Gilles. 1984. "The Effects of Language Characteristics on the Wages of Hispanic-
American Males. Journal of Human Resources 19:35-52.









Grin, FranCois. 1994. "The Economics of Language: Match or Mismatch?" International
Political Science Review 15:25-42.

Guzman, Betsy. 2001. "The Hispanic Population, Census 2000 Brief." Washington D.C.:
U.S. Census Bureau. Retrieved March 1 2007
(http://www.census.gov/prod/2001pubs/c2kbr01-3.pdf).

Hand, Michael. 2006. "Are enclaves amenities? An Empirical Investigation in the
Southwest United States." Economics Bulletin 10:1-7.

Hardy, Melissa. 1993. Regression i/ ih Dummy Variables. Sage Publications: Newbury Park.

Light, Ivan, Georges Sabagh, Mehdi Bozorgmehr, and Claudia Der-Martirosian. 1994.
"Beyond the Ethnic Enclave Economy." Social Problems 41:65-80.

Lovell, Peggy. 1989. "Racial Inequality and the Brazilian Labor Market." Ph.D. dissertation,
Department of Sociology, University of Florida, Gainesville, FL.

McGroarty, Mary E. 1990. "Bilingualism in the Workplace." Annals of the American
Academy of Political and Social Science 511:159-179.

McManus, Walter S. 1990. "Labor Market Effects of Language Enclaves: Hispanic Men in the
United States." The Journal of Human Resources 25:228-252.

Mincer, Jacob. 1974. Schooling, Experience, and Earnings. New York: National Bureau of
Economic Research.

Mora, Marie T. 2003. "An Overview of the Economics of Language in the U.S. Labor
Market." Presentation Prepared for the American Economic Association Summer
Minority Program. Denver: University of Colorado. Retrieved March 1 2007
(http://www.econ.duke.edu/aeasp/seminarfiles/seminar2003_files/MORA2 .pdf).

Nee, Victor and Jimy Sanders. 2001. "Understanding the Diversity of Immigrant
Incorporation: A Forms-of-Capital Model." Ethnic andRacial Studies 24:386-411.

Park, Jim Heum. 1999. "The Earnings of Immigrants in the United States: The Effect of
English-Speaking Ability." American Journal ofEcnomics and Sociology 58: 43-56.

Pew Hispanic Center. 2006. "Hispanics at Mid-Decade." Washington DC: Pew Hispanic Center.
Retrieved February 25 2008 (http://pewhispanic.org/files/other/middecade/Table-12.pdf).

Portes, Alejandro. 1987. "The Social Origins of the Cuban Enclave Economy ofMiami."
Sociological Perspectives 30:340-372.









------. 1995. "Economic Sociology and the Sociology of Immigration: A Conceptual
Overview." Pp. 1-41 in The Economic Sociology of Immigration, edited by
Alejandro Portes. New York: Russell Sage Foundation.

------. 1998. "Social Capital: Its Origins and Applications in Modern Sociology." Annual Review
of Sociology 24:1-24.

------. 2002. "The Two Meanings of Social Capital." Sociological Forum 15:1-12.

Portes, Alejandro and Richard Schauffler. 1996. "Language and the Second Generation:
Bilingualism Yesterday and Today." Pp. 8-29 in The New Second Generation, edited by
Alejandro Portes. New York: Russel Sage Foundation.

Portes, Alejandro and Alex Stepick. 1993. City on the Edge: The Transformation of Miami.
Berkley: University of California Press.

Ruggles, Steven, Matthew Sobek, Trent Alexander, Catherine A. Fitch, RonaldGoeken,
Patricia Kelly Hall, Miriam King, and Chad Ronnander. 2004. IntegratedPublic
Use Microdata Series: Version 3.0 [Machine-readable database]. Minneapolis,
MN: Minnesota Population Center [producer and distributor],
http://usa.ipums.org/usa/

Sanders, Jimy M. and Victor Nee. 1987. "Limits of Ethnic Solidaryity in the Enclave
Economy." American Sociological Review 52:745-773.

Santiestevan, Stina. 1991. "Use of the Spanish Language in the United Sates: Trends,
Challenges, and Opportunities." ERIC Digest. Retrieved March 1, 2008
(http://www.ericdigests.org/pre-9221/spanish.htm).

Santos, Fernanda. 2007. "Demand for English Lessons Outstrips Supply." New York Times.
January 27. Retrieved February 22, 2007
(http://www.nytimes.com/2007/02/27/education/27esl.html?_r=1&oref=slogin).

Shin, Hyon B. and Rosalind Bruno. 2003. "Language Use and English-Speaking Ability."
Washington, DC: US Census Bureau. Retrieved February 22 2008
(http://www.census.gov/prod/2003pubs/c2kbr-29.pdf).

Siegel, Paul, Elizabeth Martin and Rosalind Bruno. 2001. "Language Use and Linguistic
Isolation: Historical Data and Methodological Issues." Pp. 167-190 in Statistical Policy
Working Paper 32: 2000 Seminar on Integrating Federal Statistical Information and
Processes. Washington DC: Federal Committee on Statistical Methodology, Office of
Management and Budget. Retrieved February 22 2008
(http://www.census.gov/srd/www/abstract/ssm2007-02.html).

Stevens, Gillian. 1999. "A Century of U.S. Censuses and the Language Characteristics of
Immigrants." Demography 36:387-397.









Stolzenberg, Ross M. 1990. "Ethnicity, Geography, and Occupational Achievement of Hispanic
Men in the United States. American Sociological Review 55:143-154.

Sunshine State TESOL of Florida. 2005. "Position Statement on Bilingual Education."
Florida: SSTESOL. Retrieved March 1 2008
(http://www.clas.ufl.edu/users/rthompso/advocacybilingualed.html).

Tienda, Marta. 1983. "Market Characteristics and Hispanic Earnings: A Comparison of
Natives and Immigrants." Social Problems 31:59-72.

US Census Bureau. 1999. "Findings on Questions on Race and Hispanic Origin Tested in the
1996 National Content Survey." Washington, DC: U.S. Census Bureau. Retrieved
February 22 2008
(http://www.census.gov/population/www/documentation/twps0016/report.html#intro).

------. 2006. "Facts for Features: Hispanic Heritage Month 2005." Washington, DC: U.S. Census
Bureau. Retrieved March 1, 2008. (http://www.census.gov/Press-
Release/www/releases/archives/cb05ff- 14-3.pdf).

------. 2007. "America Speaks: A Demographic Profile of Foreign-Language Speakers for the
United States: 2000." Washington, DC: U.S. Census Bureau. Retrieved March 1, 2008
(http://www.census2010.gov/population/www/socdemo/hh-fam/AmSpks.html).

------. 2008. "State and County QuickFacts: Miami-Dade County and Florida." Washington, DC:
U.S. Census Bureau. Retrieved March 1, 2008
(http://quickfacts.census.gov/qfd/states/12/12086.html).

Viglucci, Andres. 2001. "Spanish is Spreading Northward in Florida." Miami Herald.
November 22. Retrieved March 1 2008 (http://www.puertorico-
herald.org/issues/2001/vol5n49/SpanishNorth-en.html).

Wilson, Kenneth L. and Alejandro Portes. 1980. "Immigrant Enclaves: An Analysis of
the Labor Market Experiences of Cubans in Miami." The American Journal of Sociology
86:295-319.

Zhou, Min and John R. Logan. 1989. "Returns of Human Capital in Ethnic Enclaves: New York
City's Chinatown." American Sociological Review 54:809-820.









BIOGRAPHICAL SKETCH

Molly Dondero was born in Philadelphia, Pennsylvania in 1981. She graduated from the

Pennsylvania State University in 2004 with a B.A. in Spanish and English. Before attending the

University of Florida, Molly worked as a Fulbright teaching assistant in the English Department

at the Universidad Nacional de Villa Maria, in C6rdoba, Argentina. Upon graduating in May

2008 with an M.A. in Latin American Studies, Molly plans to pursue a Ph.D. in Sociology at the

University of Texas at Austin.





PAGE 1

1 LANGUAGE AND EARNINGS OF LATINOS IN FLORIDA: THE EFFECT OF LANGUAGE ENCLAVES By MOLLY DONDERO A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Molly Dondero

PAGE 3

3 To my parents, Marian and Lawrence

PAGE 4

4 ACKNOWLEDGMENTS Many thanks go to the m embers of my s upervisory committee for their guidance and accessibility. I appreciate both their assistance with this thesis and the positive influence they had on my overall academic development. I am especially grateful to my committee chair, Dr. Charles Wood, for helping me to develop this pr oject and for reminding me to keep it simple. His direction has been invaluable. I thank Dr. Carmen Carrin-Flores for le nding her expertise in labor economics to this projec t. I thank Dr. Efran Barrada s for his support and advice. I owe a great deal of gratitude to the Center for Latin Amer ican Studies for providing me with the Interdisciplinary Field Research Grant that made this study possible. I also thank my informants who so generously took time out of their schedules to part icipate in this study. For their unconditional love a nd support, I thank my pare nts, Marian and Lawrence Dondero. They have fostered my love of learning, and are a great source of inspiration for me. Finally, my heartfelt thanks go to Werllayne Nu nes for his constant encouragement and moral support. On countless occasions, he selflessly pu t aside his own work to help me move toward completing this thesis.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES.........................................................................................................................8 LIST OF ABBREVIATIONS.......................................................................................................... 9 ABSTRACT...................................................................................................................................10 CHAP TEr 1 INTRODUCTION..................................................................................................................12 Overview....................................................................................................................... ..........12 Background.............................................................................................................................13 Spanish-Speaking Populati on in the United States ......................................................... 13 Spanish-Speaking Population in Florida......................................................................... 14 Research Questions............................................................................................................. ....15 2 LITERATURE REVIEW.......................................................................................................17 Conceptual Framework........................................................................................................... 17 Economics of Language.................................................................................................. 17 Forms-of-Capital Model of Incorporation....................................................................... 18 Language and Earnings...........................................................................................................22 Effect of Language Enclaves on Language and Earnings......................................................24 Value of Bilingualism in the Labor Market............................................................................ 28 3 DATA AND METHODOLOGY........................................................................................... 33 Research Design.....................................................................................................................33 Quantitative Analysis..............................................................................................................34 US Census Data...............................................................................................................34 Measuring the Language Enclave...................................................................................35 Sample Description.........................................................................................................36 Description of Variables.................................................................................................. 37 Dependent variable...................................................................................................37 Independent variable................................................................................................38 Control variables...................................................................................................... 39 Statistical Model/Data Analysis...................................................................................... 41 Qualitative Analysis........................................................................................................... .....42 Methodological Caveats......................................................................................................... 43

PAGE 6

6 4 RESULTS...............................................................................................................................49 Summary Statistics.................................................................................................................49 Research Question One: The Effect of La nguage Enclaves on Returns to English ............... 50 Research Question Two: Returns to Bilingualism..................................................................55 5 DISCUSSION.........................................................................................................................62 Theoretical Implications....................................................................................................... ..62 Language, Earnings, and Labor Market Characteristics ................................................. 62 Returns to English language proficiency................................................................. 62 Returns to bilingualism............................................................................................65 Immigrant/Minority Incorporation.................................................................................. 66 Policy Implications.................................................................................................................67 Suggestions for Future Research............................................................................................ 68 6 CONCLUSION..................................................................................................................... ..70 LIST OF REFERENCES...............................................................................................................72 BIOGRAPHICAL SKETCH.........................................................................................................77

PAGE 7

7 LIST OF TABLES Table page 3-1 Language spoken by concentration of Spanish-speakers, Florida 2000............................ 47 4-1 Mean and standard deviation of vari ables used in Low Conc entration model.................. 58 4-2 Mean and standard deviation of vari ables used in High Concentration m odel................. 59 4-3 Returns to English-speaking ability................................................................................... 60 4-4 Returns to bilingualism.................................................................................................... ..61

PAGE 8

8 LIST OF FIGURES Figure page 3-1 Florida linguistic composition........................................................................................... 48

PAGE 9

9 LIST OF ABBREVIATIONS HC High concentration of Spanish-speakers LC Low concentration of Spanish-speakers Super-PUMA Super Public Use Microdata Area

PAGE 10

10 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Arts LANGUAGE AND EARNINGS OF LATINOS IN FLORIDA: THE EFFECT OF LANGUAGE ENCLAVES By Molly Dondero May 2008 Chair: Charles Wood Major: Latin American Studies Language ability has assumed priority in cu rrent studies of the economic success of immigrants and minority language speakers. Past studies have shown that language ability, a key human and cultural capital trait, tends to be positively associated with earnings. Building on this past research, the goal of this study is to examine how the effect of English language proficiency on earnings of Hisp anic men in Florida varies by labor market context. Specifically, it aims to compare the difference in th e effect of English language proficiency on earnings in areas densely populated by Spanish speakers to the effect of language on earnings in areas dominated by English speakers. I predict that Engl ish language proficiency will have a greater impact on earn ings in areas where Spanish is not widely spoken. In areas where there are large enclaves of Spanish-sp eakers, English will likely be a less important determinant of earnings. The effect of bilingualism on earnings is also analyzed in this manner. To test my hypotheses, my study consists of two parts: the first based on statistical analysis of US census data and the second base d on qualitative interviews. Findings show that English language ability is indeed an important determinant of earnings both in areas with a high proportion of Spanish-speakers and in areas with a low proportion of Spanish-speakers. However, results from the statis tical analysis show that Englis h language ability has a greater

PAGE 11

11 impact on earnings in areas with a high proportion of Spanish-speakers. While English language proficiency yields greater earni ngs in these areas, Spanish lan guage proficiency also has a positive effect on earnings. Fully bilingual Hi spanics earn more than their English only counterparts in these areas.

PAGE 12

12 CHAPTER 1 INTRODUCTION Overview Language ability has assum ed priority in cu rrent studies of the economic success of immigrants and minority language speakers. Langua ge often serves as a key human capital and cultural capital tra it that facilitates incorporation into the host country. Many scholars cite the ability of immigrants to effectively communicat e with members of the receiving country as the most important alterable factor that affects their integration into their c ountry of destination and their absorption into the labor market (Dustmann and van Soest 2001, 2002). The goal of this study is to examine the effect of English language proficiency on earnings among Latinos1 in Florida. Specifically, it aims to compare the difference in the effect of English language proficiency on earnings in ar eas densely populated by Spanish speakers to the effect of language on the earnings in areas dominated by English spea kers. I predict that English language proficiency will have a greate r impact on earnings in areas where Spanish is not widely spoken. In areas such as ethnic enclave economies, where Spanish is commonly spoken, English will likely be a less important de terminant of earnings. Simply stated, the more Spanish that is spoken in a given area, the less impor tant English is to earn ings in that area. I also expect to find the inverse to be true of Spanish language skills; in a Spanish-language enclave, Spanish proficiency will be more important to earnings than it is outside the enclave. Results from this study offer valuable insight about the factors that affect the economic integration of immigrants and non-native speakers into the labor market. Findings also have important policy implications for English language and bilingual training programs in Florida. 1 Following current US Census Bureau terminol ogy, the terms Latino and Hispanic are used interchangeably throughout this paper.

PAGE 13

13 Background Spanish-Speaking Population in the United States The political and social controversies surr ounding the position of English versus other languages in the United States has endured sin ce the founding of the nation (Stevens 1999). The recen t growth of the Spanish-speaking populati on in the US has reignited the language issue and transformed it into a symbolic battleground in the current immigration debates. Spanish now ranks as the second most common language spok en in the US (Shin and Bruno 2003). In the Census 2000, over 60% of respondents who spea k a non-English language at home reported speaking Spanish (Mora 2003). This represents a 10% increase in th e number of Spanishspeakers reported from the 1990 Census (Mora 2003). The rise of Spanish language use in the US stems from the rapid growth of the countrys Latino population. While not all Latinos speak Sp anish, nearly all Spanish speakers in the US are Latinos (Santiestevan 1991). The latest updates of the 2000 Census estimate the total Latino population to be 41.3 million, or 14% of the tota l US population, making Latinos the countrys largest minority group (US Census Bureau 2005). According to US Census Bureau projections, these numbers show no signs of waning. On the contrary, estimates pred ict the Latino population to more than double to 102.6 million by 2050 (US Census Bureau 2005). Among the current Latino population, nearly 31 million people over age 5 report speaking Spanish at home, constituting a ratio of more than 1-in-10 rati o of household residents in the US (US Census Bureau 2005). Debate remains over whether Span ish language use in the US will continue to grow at a rate as fast as that of the Latino popul ation. Some believe that Spanish-speakers will prove to be a unique group in US linguistic history by maintaining their native tongue for several generations. Others contend that as English language fluency increases across generations, Spanish-speakers will gradually abandon Spanish in favor of English, as have many other

PAGE 14

14 immigrant and minority language groups before them. No matter what the fate of Spanish language in the US proves to be, the sheer num ber of Spanish-speakers in the US today has undeniably reshaped the countrys current linguistic landscape. Spanish-Speaking Population in Florida Floridas Latino population is the third larges t in the nation, behind California and Texas, respectively (Pew Hispanic Center 2 006). Of th e states more than 18 million-plus residents, over 3.2 millionor more than 19% of the stat es total populationare Latino (US Census Bureau 2008). Over 75% of these residents repo rt speaking Spanish at home, making Florida the state with the fifth largest pr oportion of residents who speak Spanish at home (US Census Bureau 2007; Viglucci 2001). Even though a 1988 la w designated English as the states official language, these numbers show that Spanish-speakers in Florida continue to exert a powerful presence. The uneven distribution of the Spanish-speak ing population in Florida makes it an ideal setting for a comparative research design. Alth ough recent research highlights growing pockets of Latinos in Central and North Florida, most of Floridas Spanis h-speaking population is concentrated in South Florida, and in Miami-Dade County in particular (Duany and MatosRodrguez 2006). Miami-Dade is home to ove r 1.3 million people of Hispanic origin. With Hispanics representing 61.3% of the total population of th e county, Miami-Dade has the distinction of being one of 50 counties nationwid e in which Hispanics create the majority (US Census Bureau 2008). Wilson and Portes (1980) have well documen ted the growth of the Cuban enclave economy in the city of Miami and the wa ys in which it stimulated the development of the mainstream economy in Miami (Portes 1987). Th e latest US Census estimates that Hispanicowned businesses account for 54.9% of all bus inesses in Miami-Dade County (US Census Bureau 2008). In addition, Miami has emerged as a hub for international business, particularly

PAGE 15

15 for multinational corporations, fi nancial institutions, and Spanis h-language media conglomerates from Latin America and the Cari bbean. Its large Span ish-speaking population and its status as a prominent international business center have prompted many to refer to the city as the financial capital of Latin America. Such characteristics make it likely that English language ability will have a lesser effect on earnings in Miami. Research Questions Having established the context a nd setting of this study, I turn now to the specific research hypotheses. This study is prem ised on the theore tical proposition that labor market context will dictate the value that the market places on certain forms of human and cultural capital, in this case language ability. Specifically, it asks: Do es the presence of a large Spanish language enclave alter earnings returns to English la nguage proficiency? Th e corresponding research hypotheses predict 1) that English will be a less important determinan t of earnings in areas with a high concentration of Spanish speak ers and 2) that bilingual Englis h-Spanish skills will be more important in areas with a high concentration of Spanis h-speakers. To test these hypotheses, my project consists of two parts: the first based on statistical analys is of census data and the second based on qualitative interviews. Three features set this study apart from others. First, it is the first st udy of this nature to focus specifically on the Florida la bor market; other studies on this topic have used data from either the Southwest US, the US in general, or Canada. Second, it provid es empirical evidence of an unofficial, minority languageSpanish as pot ential asset in the labor market (Pendakur and Pendakur 2002). Third, the qua litative component of this st udy gives a unique perspective on the issue of language and earni ngs, thereby going beyond previous research based exclusively on quantitative methodologies.

PAGE 16

16 This study is nonetheless firmly rooted in traditional con cepts in sociology and labor economics. Chapter 2 offers a literature review that details the esta blished theoretical and empirical work that informs this study. Chap ter 3 describes the design and methodology of the study. Chapter 4 presents results, and Chapte r 5 discusses the theo retical and practical implications of the findings.

PAGE 17

17 CHAPTER 2 LITERATURE REVIEW The increased flows of i mmigration to the US over the past several decades spawned an abundance of studies exploring the role of Englis h-language proficiency in the US labor market. Yet, despite the current influx of immigrants from Spanish-speaking Latin America, and the continued growth of linguistic pluralism in the United States, scholars have produced little recent research on the topic. Most of the published li terature dates back to over a decade ago and, in many cases, relies on data that are even older (Mora 2003). Within these studies, only a small body of research addresses the relationshi p between language, ear nings, and linguistic concentrations. These few empirical studies have, for the most part, produced opposing and inconsistent conclusions about the nuanced ways in which the linguistic profile of the labor market affects the economic valu e of language proficiency. In this chapter, I review the existing literatu re on the topic, and highlight areas that merit additional research. The first section presents the conceptual framework for the study. It examines the economics of language and uses a forms-of-capital model to establish the importance of language in the economic incor poration of non-native speakers. The second section reviews the empirical studies that address the rela tionship between language and earnings. The third section summarizes the studies th at specifically treat the relationship between language, earnings, and linguistic con centrations. The final section turns to the existing literature on the value of bilingualism in the workplace. Conceptual Framework Economics of Language What role does language play in the econom ic incorporation of immigrants and minority language speakers? Although the answer to this que stion may seem transparent, a closer analysis

PAGE 18

18 reveals that language functions as one of the mo st versatile and valuable traits in the labor market. As such, it plays a diverse and significant role in the economic incorporation of minority language speakers. For this reason, it is useful to outline the theories that hypothesize the ways in which language operates in the labor market. Beginning with J. Marschaks pionee ring 1965 study on economic approaches to language, the economics of language is a relatively recent area of specialization within the social sciences (Grin 1994). It refers to an interdisciplinary field of research that examines the relationship between linguistic and economic variables (Grin 1994). Prior to the publication of Marschaks work, mainstream economics larg ely overlooked this relationship, believing language and other ethnic characteristics to be of little overall importance in the labor market (Grin 1994). However, increased linguistic, et hnic, and cultural plur alism in contemporary society, and rising awareness of social problems resulting from su ch pluralism, stimulated the growth of studies in th is field (Grin 1994). Forms-of-Capital Model of Incorporation The economic consequences of language pr oficiency are of particular importance to research on the incorporation of immigrants a nd minorities. The recognition of language as a key component in labor market out comes contributed to the current shift in theories of immigrant incorporation. Traditional assimilation theories no longer suffice as an accurate model of incorporation of contemporary immigrants, as they fail to account for the diversity of incorporation experiences by different groups (Tienda 1983). Rather, the contemporary incorporation experience is bett er framed within the forms-o f-capital model, developed by Nee and Sanders (2001). This model privileges the fact ors that explain the dive rsity in the modes of incorporation of immigrants. It views incorpor ation mainly as a functio n of the human, social, and cultural capital that immi grants possess and accrue. The role of language in

PAGE 19

19 immigrant/minority incorporation can be best understood through this model since language serves simultaneously as multiple forms of capital: human, cultural, social, and linguistic. While this study borrows the basic fram ework from Nee and Sanders model, it modifies the model in three ways: first, it examines language as forms of capital at the individua l, rather than household or family level; second, it expands the model to include th e incorporation not only of immigrants, but also of minority language speakers in general; and third, it takes into consideration the ways in which certain labor market characteris tics, specifically the linguistic composition of the labor market, affect the ec onomic value of these forms of capital. Nee and Sanders model draws on the concep ts of capital developed by Gary Becker (1993), James Coleman (1988), and Pierre Bourdieu (1986). Bourdieu (1986) theorizes that the general notion of capital provides a means for unde rstanding most types of interaction in the social world. It is what makes the game s of societynot the leas t, the economic game something other than simple games of chance o ffering at every moment the possibility of a miracle (Bourdieu 1986: 46). The application of the forms-of -capital model to the study of language and incorporation is pa rticularly useful if one view s linguistic exchange as an economic exchangeestablished within a symbo lic relation of power between a producer, endowed with a certain linguistic capita l, and a consumer (Bourdieu 1991:66). According to Bourdieu, words represent far more than simple means of communication; they are signs of authority and wealth that reveal a particular social value. Fr om this perspectiv e, language takes on a prominent role in the inco rporation of minority language sp eakers. The remaining parts of this section define the concepts of human, social, cultural, and li nguistic capital, and the ways in which language functions as a form of each type of capital in the labor market.

PAGE 20

20 Most studies of language and immigran t/minority incorpora tion view language proficiency primarily as a human capital trait (Park 1999). The con cept of human capital offers a way to understand differences in earnings that are not fully explained by ex ternal factors (Becker 1993). Becker (1993) broadly defines human capital as knowledge, skills, and health. As such, human capital is different from physical or fina ncial capital because, unl ike these tangible forms of capital, it cannot be separated from an individual (Becker 1993). In this sense, language can be considered one of the most basic forms of hu man capital. Simply stat ed, language is essential to effective communication which, in turn, is es sential to enhanced productivity. Indeed, many earnings analyses have shown that language has an effect on earnings that is comparable to that of some of the most common hu man capital characteristics, su ch as education and number of years in the host country (Park 1999; Chiswi ck 1991; McManus 1985; Grenier 1984). Chiswick (1978) was the first researcher to apply th e concept of human cap ital to the economic achievement of immigrants (Portes 1995). He conc luded that individual skills greatly impact the incorporation experience. Sociologists, however, take issue with this approach to incorporation, noting that group membership and other social contexts are also at work (Portes 1995). The notion of social capital has thus em erged as an important concept. The concept of social capital offers a way to theorize the value of group membership in the incorporation experience. Broadly defined, so cial capital refers to benefits or resources created by and derived from social networks (P ortes 1998; 2002). It is the aggregate of the actual or potential resources whic h are linked to possession of a dur able network or more or less institutionalized relationships of mutual acquaintan ce and recognitionor in other words, to membership in a groupwhich provides each of its members with the backing the collectivityowned capital (Bourdieu 1986:51). Language, as the primary means of social interaction,

PAGE 21

21 fosters social capital. A common la nguage cultivates a sense of shared identity. It functions as a criterion for membership into a given group, and thus, as a networking tool that can facilitate labor market success. The social aspect of language relates directly to the third form of capital: cultural capital. Cultural capital refers to knowledge, skills, education, and values acquired through cultural transmission resulting from membership in a given class, region, na tion, or ethnic group (Bourdieu 1986). Bourdieu devel oped the concept of cultural capit al as a way to explain unequal scholastic achievement among students from di fferent class backgrounds It represented a departure from past theories that viewed academic success as the result of solely genetic and human endowments. He hypothesized that stude nts who possess knowledge of the mainstream culture in which their education system is rooted generally achieve greater academic success. The same logic can be applied to economic achievement in the labor market. In order to successfully navigate the labor market, worker s must possess knowledge of the cultural norms and codes that underpin the market. Language, wh ich is culturally tran smitted, serves as a vehicle that allows an individua l to understand these norms and ope rate effectively within them. Thus, the cultural value of language makes it an essential trait for labor market success (Pendakur and Pendakur 2002). However, as cultural signifier, language can also be a detriment in the labor market when used as grounds for economic discrimination based on culture (Pendakur and Pendakur 2002). Bourdieu identifies another form of capital specifically related to language skills linguistic capital. Linguistic cap ital represents an embodied form of cultural capital in the sense that, once acquired, it cannot be sepa rated from the individual. However, the value of linguistic capital, like the value cultural capital, is relati ve. The market determines its ultimate worth

PAGE 22

22 (Bourdieu 1991). That is to say, knowledge of a minority language may highly valued or even essential for survival within a given language encl ave, but useless in another context. The value of linguistic capital is therefore conti ngent on the context in which it is used. These four forms of capitalhuman, social cultural, and linguisticare primarily symbolic. Unlike economic capital, the transmissi on and acquisition of these forms of capital is not easily recognized. As such, their value is often underestimated or unrecognized (Bourdieu 1986). However, these forms of capital, under certain circumstances, can be converted into economic capital (Bourdieu 1986), thus making la nguage proficiency an asset in the labor market. Even so, language and other human, cultur al, and social capital variables only partially explain wage differentials among immigrant/minority language groups. Tienda (1983) argues that such a model of inco rporation needs to be fu rther expanded to accoun t for structural forces that also affect labor market outcomes. Structur al conditions such as labor market characteristics and public opinion toward immigrants and foreign language use are particularly relevant to a comprehensive model of incorpor ation (Tienda 1983). While it is beyond the scope of this paper to include a comprehensive inventory of labor market characteristics in the empirical analysis, this study focuses on one defining macro-structural featurethe li nguistic profile of the labor marketand its interplay with the relationship between language and earnings. Language and Earnings This section reviews the studies that em pi rically demonstrate the impact of English language proficiency on the la bor market outcomes of non-na tive speakers, in particular Hispanic males. In the early 1990s, earnings anal yses of Hispanic males began to devote more attention to the role of English language prof iciency. The introduction of language as a key variable in the earnings functions offered insight into causes of wage differentials. Despite different definitions of English proficiency and a variety of me thodological approaches, nearly

PAGE 23

23 all studies have found that worker s experience an earnings penalty for the lack of the ability to speak English well (Mora 2003). McManus, Gould, and Welchs (1983) seminal study of the role of English language proficiency on earnings uses data from the 1975 Survey of Income and Education (SIE) to explore the cost of language disparity among Hispanic males in the US. Regression estimates from this study suggest that, compared to their fluent Anglo counterpa rts, English-deficient Hispanic males experience earnings penaltie s between 17-30%, depending on the level of education being compared. Furthermore, their fi ndings show that English language deficiency reduces earnings returns to education and work e xperience. Greniers (1 984) study also uses the 1975 SIE to estimate the effect of language char acteristics on earnings of Hispanic-American men with limited English proficiency. His regre ssion analysis finds that 1) Hispanic males earn significantly less than their Anglo counterparts, and 2) language characte ristics explain up to one third of these earnings differentials. Subsequent studies report similar findings. Chiswick and Millers (1992) statistical analysis shows that immigrants who speak English well or very well earn on average 17% more than those who are Eng lish-deficient. Using a different sample and reference group, Borjas (1994) concludes Hispanic immigrants who do not speak English earn 17% less than those who speak English. In keepin g with these results, ot her studies have shown that the rate of immigrant wage assimilation is proportional to increase in English-speaking skills and time spent in the US (Funkhouser 1996; Carliner 1995; Gonzalez 2000). Such research leads Chiswick and Mi ller (1992, 1995, 2002) to argue that English language proficiency ranks as one of the most im portant determinants of earnings of non-native speakers. They conclude that the acquisition of English language sk ills clearly pays in the labor market (2002: 42). Despite strong empirical ev idence from Chiswick an d Miller, literature on

PAGE 24

24 the topic remains divided with regard to the extent to which language affects earnings (Park 1999). While language ability is positively associat ed with earnings, other research suggests that it is not a major determinant of earnings. Such re search points to other factors, such as national origin, length of US residence, a nd education, as the primary predic tors of earnings of non-native speakers (Borjas 1982; Tienda 1983). Much of this previous resear ch invokes a design that focuse s attention almost exclusively on individual level characterist ics. In doing so, this resear ch overlooks the importance of context. Specifically, as I argue in this study, income returns to language are not uniform across different contexts. Instead, the relationship be tween proficiency and earnings is contingent upon the linguistic profile of the labor market in whic h people work. In this case, a high proportion of Spanish-speakers in a given area will lower income returns to English proficiency in that area. Thus, this study will show that the inclusion of a minority language enclave measure can provide important insights about the ex tent to which language e xplains variance in earnings. Effect of Language Enclaves on Language and Earnings While m ost prior research agr ees that greater English language proficiency yields greater earnings, this consensus disappears when analyzing the effect of language on earnings in regions with strong minority language f oundations, such as ethnic enclav e economies (Dvila and Mora 2000). Theoretical formulations th at treat the topic tend to agr ee that the pres ence of a large immigrant/minority population in a given area will alter returns to vari ous determinants of earnings, such as language, ethnic ity, and other variables. The re turns accruing to migrants who move to areas of high ethnic density should diffe r from those accruing to migrants who move to areas of low ethnic density, although it is not obvi ous whether these returns will be positive or negative (Tienda 1992: 661). McManus (1990) theo rizes that large ethnic enclaves lower earnings returns to English language proficiency. That is to say, the market in these areas places

PAGE 25

25 less value on English language. Such thinking follows the forms-of-capital model, which would assume that the labor market value of a given language depends on its demand in the marketplace (Bourdieu 1986), and th at structural forces, such as the existence of a minority language enclave, would indeed alter value of human and cultural capital in a given area. Such a model helps explain why a given language is important to earn ings in one area, but less important in another. These conceptualizations find their roots in the ethni c enclave economy theories developed by Wilson and Portes (1980). They define the ethnic enclave economy as a concentrated area of businesses owned by em ployers of immigrant or minority ethnic backgrounds that employ co-ethnic workers. Gr oups in these areas te nd to maintain their cultural customs and language to a greater extent than co mparable groups outside the enclave. Wilson and Portes (1980) theori ze that the ethnic enclave econo my functions as a mode of immigrant incorporation by providing provides a ccess to jobs, opportunities and resources that might not otherwise be as available for minorities outside the enclave. However, debate exists over the extent to which the ethnic enclave economy works as a succ essful incorporation technique. Wilson and Portes (1980) theorize th at returns to human cap ital brought from the home country are higher in th e enclave economy than in the ma instream economy. While this may create initial earnings advantages for immi grants and minorities in these areas, others contend that the ethnic enclave economy acts as a mobility trap in which earnings eventually plateau (Sanders and Nee 1987). (Such traps generally refer to employees, not employers, in the enclave economy). This study does not focus sp ecifically on ethnic enclaves economies per se; that is to say, it does not use th e number of immigrant-owned busine sses in an area as measure of enclave. Rather, it focuses on a broader defini tion of enclavethe lingui stic enclavemeasured

PAGE 26

26 by the number of Spanish-speakers in a given area. Nevertheless, it is important to highlight the concept of the ethnic enclave economy because its basic theoretical assumptions apply directly to this study. Despite these concordant theoretical a ssumptions, empirical analyses have produced inconsistent results. These inconsistencies may in part be due to the fact that only a small number of studies consider the effects of language enclaves on earnings returns to language skills. Results from these studies can be roughly classified into four ma in categories: 1) those that find that the presence of a large minority language enclave lowers returns to English (McManus 1990; Chiswick and M iller 2002); 2) those that find that the English deficiency earnings penalty is greater inside the linguistic enclave (Bloom and Grenier 1992); 3) those that find mixed results (Dvila and Mora 2003; Hand 2006); and 4) those find no evidence that the existence of a linguistic enclave alters earnings return s to language skills (Fry and Lowell 2003). In his analysis of earnings regressions for Hispanic males based on 1980 US Census data, McManus (1990) finds that the enclave reduces earnings penalties associated with English language deficiency. He invokes the Hispanic et hnicity variable to operationally define the enclave. He concludes that the greater the Hisp anic population, the lower the returns to English. Chiswick and Millers (2002) s ubsequent study utilizes 1990 US Ce nsus data to test similar hypotheses. They conclude that, all things bein g equal, average earnings tend to be lower for immigrants that live in an area with a large mi nority language concentrat ion. They find that the English-deficient wage penalty is also smaller in these areas. When Bloom and Grenier (1992) apply si milar regression techniques to 1970 and 1980 US Census data, they find that the language-bas ed earnings differentials between Hispanics and Whites are actually greater in areas with a la rge Spanish-speaking population. To determine

PAGE 27

27 areas of high and low concentrations of Span ish-speakers, Bloom and Grenier, like McManus, use Hispanic origin as a proxy for Spanish language and White of non-Hispanic origin as a proxy for English. Dvila and Moras (2000) study, which us es 1990 US Census data, seems to support Bloom and Greniers results to a large extent and contradict McManuss (1990) and Chiswick and Millers (2002) findings. Employing the US-M exico border as a minority language enclave, they compare the effect of English language proficiency on earnings of Mexican-Americans along the US-Mexico border relative to their non-bor der counterparts in the rest of the US. Their regression results suggest that the English defici ency earnings penalty is slightly greater for Mexican-American males in border cities than in non-border cities However, in the case of Mexican immigrants, there is no significant differ ence in the English deficiency earnings penalty between border and non-border cities. Critics cite Dvila and Moras measure of minority language enclav e as a possible source of error in their analys is. Hand (2006) believes that thei r use of border cities and non-border cities as proxies for enclaves and non-enclaves may have skewed results. He maintains that border cities cannot necessarily be described as enclaves solely on the basis of their status as traditional immigrant-receiving destinations. Furthe rmore, he notes that Dvila and Mora make no attempt to account for language enclaves or ethnic enclave economies that may exist in nonborder metropolitan areas in the US (Hand 2006). Hands own empirical study of the role of linguistic enclaves in wage dete rmination of minority speakers in the Southwest US attempts to remedy this perceived error by using 2000 US Census data to measure the density of Spanishspeakers throughout defined areas in New Mexico and Arizona. However, his analysis also produces mixed results, though they run contrary to Dvila and Moras findings. Like Dvila

PAGE 28

28 and Mora, Hand runs separate regression mode ls for Mexicans and Mexican-Americans. He finds that English-deficiency earnings penalties are indeed reduced in th e linguistic enclave but only for foreign-born respondents. Results fo r native-born Spanish-sp eaking respondents do not prove statistically significant and are thus inconclusive. Value of Bilingualism in the Labor Market If earnings returns to Englis h decrease as the size of the Spanish-speaking population increases, do earnings returns to Spanish then increase? Simple supply and demand theory would assume that areas with a large population of Spanish-speak ers would have an increased demand for Spanish-speaking workers. As discussed in the previous section, past research has shown that, bilingualism is essential for non-nati ve speakers to successfu lly integrate into and compete in the US labor market. However, fo r native speakers of English, monolingualism in English is not generally perceived as a disadva ntage in the labor market. However, this study asks: do bilingual speakers earn mo re than their monolingual Englis h counterparts in areas with a large proportion of Spanish-speakers? In other words, are there earnings penalties for not having Spanish language profic iency in those areas? Examined from the forms-of-capital perspectiv e, bilingual abilities theoretically represent a greater amount of human, cultural, and social capital than monolingua l abilities. Bilingualism offers the distinct advantage of being able to access dual cultures, networks, and other such resources. Thus, knowledge of an additiona l language, even a minority language, should improve labor market outcomes (Pendakur and Pe ndakur 2002). Yet, despite this logic and the well-established body of research on the economics of language, a relatively small amount of research addresses the value of bilingualism in the labor market. Educational research, on the other hand, has devoted much a ttention to the effect of bilingualism in the classroom. Prior to 1960, main stream belief in the fields of education and

PAGE 29

29 psychology argued that bilingualism caused ac ademic failure, mental confusion, and psychological damage (Portes and Schauffler 1996). However, a crop of sound methodological studies, starting in the 1960s and continuing to the present, reve rsed this belief after showing that, all other factors being equal, bilingualism is associated with higher scholastic achievement, greater cognitive flexibility, and a better capacity to deal with abstract concepts (Portes and Schauffler 1996). These studies have shown that instead of creating confusion, having two symbols for each object enhanced understa nding (Portes and Schauffler 1996: 11). Just as these enhanced abilities have been shown to increase academic achievement, so too should they reap positive outcomes in the labo r market, or at least in areas with a large population of non-native English speakers. Yet, past research in the field of labor economics and sociology has produced inconsistent results on this topic. Qualit ative findings from Portes and Stepicks (1993) study on the transformation of Miami provide evidence of the perceived advantages of bilingualism, especially in labor markets in areas with a large Spanish-speaking population. A quote from one of th eir interviews with a Cuban civic activ ist and head of a multiethnic community organization in Miami illustrates this point. He observes: Language has great importance because if an individual owns a store whose clients come from Latin America, he will need bilingual employees. During Christmastime, ninety percent of the stores advertise for bilingua l employees. To a person who does not know the language, this situation represents an economic problem because he knows that, unless he knows Spanish, he would not compete succes sfully in the labor market (Portes and Stepick 1993: 12). In other words, the large number of Spanis h-speaking consumers in Miami gives bilingual workers an advantage over their Englishonly counterparts in the labor market. However, such opinions have received little empirical backing from other studies. Other researchers contend that bilingual language skills present no additional earnings advantage in the labor market. Carliner (1981) theorizes that the earnings of bilinguals and monolingual are

PAGE 30

30 likely to be roughly equivalent. He posits, I n multilingual societies, if labor demand for speakers of one language exceeds the supply of na tive speakers, bilingual workers will generally come from other language groups. There will be a wage premium for speaking the "excess demand" language but no additiona l premium for being bilingual (Carliner 1981: 384). He finds evidence of this pattern in data from the 1970 Canadian Census. Findings from Fry and Lowells (2003) analysis of the value of bilingualism in the U.S. labor market support this theory. In their regression of earnings on a variety of language variables from the 1992 National Ad ult Literacy Survey, they find that in the general US labor market, bilingual workers have a marginally significant earnings advantage over their monolingual English counterparts. However, thes e earnings returns disappear after they remove the effects of education, age, and other control variables. Even after inserting a measure for geographic linguistic concentrations, they find no evidence that language enclaves alter the returns to bilingualism. They conclude that high er returns to bilingualis m are likely limited to specialized jobs that deal primarily in the international labor market. Pendakur and Pendakurs (2002) study of th e economic impact of bilingualism in the Canadian labor market produces slightly different and more varied results. Using a language-ashuman-capital perspective, they predicate their study on three main theoretical assumptions that follow a logic similar to that employed in this study: 1) polyglots should earn more than unilinguals 2) different citie s (with different populations of majority and minority language speakers) should have different patterns of retu rns to language knowledge and 3) these returns should be correlated with the size of the lingu istic communities (Pendakur and Pendakur 2002: 150). Applying regression techniques to data from Canadian census, they find that those respondents who are fluent in Canadas two official languag esEnglish and Frenchtend to

PAGE 31

31 earn more than their monolingual (English or French ) peers. However, when assessing returns to knowledge of unofficial languages, the story ch anges. Their analys is finds no earnings advantages for possessing knowledge of one offici al and one or more uno fficial languages. In fact, those who speak one or more unofficial la nguages (in addition to one official language) actually earn less than those fluent in only one official language. While earnings returns to language skills do vary slightly according to th e size of the corresponding linguistic community, in no instance do bilinguals fluent in an unofficial language earn more than monolinguals fluent in one of the official languages. They attr ibute these rather counterintuitive findings to discrimination caused by the ethnic and cultural dimensions of language. In contrast to these findings, evidence from the 1990 US Census points to a possible shift in the economic value of English-Spanish bilingua lism in the US labor market. In their study of income patterns of bilingual a nd English-only Hispanics in the US, Boswell and Fradd (1999) find that bilingual language skills have a greater economic value than monolingual English skills in select US cities with larg e Hispanic populations. Their simple cross-tabulations of mean earnings by language ability of Hispanics clearly show that bilinguals ea rn more than their monolingual English counterparts in certain ar eas of the US. These results hold true for Hispanics in three metropolitan areas in the US: Miami, El Paso, and, to a smaller extent, Chicago. While their findings leave little doubt of the positive associat ion between bilingualism and earnings in these areas, there is a need to go beyond this general observation by controlling for the effects of education, citizenship status, year s in the US, and other se lect variables in order to observe pure effect of bilingualism on earnings. As this literature review i llustrates, there is a large body of research dedicated to the study of the economic impact of language proficiency. However, inconsistencies within this

PAGE 32

32 research about the effect of language enclaves on returns to English-proficiency and bilingualism make these topics ripe for further study. Data from the 2000 US Census provide a number of variables that permit detailed expl oration of these topics. The methodology and variables used to achieve this analysis are described in detail in the next chapter.

PAGE 33

33 CHAPTER 3 DATA AND METHODOLOGY Research Design Building on the theoretical foundations and s ubstantive research highlighted in the previous chapters, m y hypothesis predicts that English language ability will be a strong determinant of earnings in both areas with high co ncentrations and areas with low concentrations of Spanish-speakers, even after controlling for ed ucation and other variables that may influence earnings. However, adopting McManuss (1990 ) theoretical framework and reasoning, my primary hypothesis asserts that the labor market context will affect the relationship between language and earnings. That is to say, among Hispanics in Florida, English language ability will have a stronger impact on earnings in areas with a low percentage of Spanish-speakers. In areas with large minority language en claves, where there are high percentages of Spanish-speakers, English language ability will be less relevant to earnings; the labor market in these areas will place less value on English language ability. My corollary hypothesis follows this same line of reasoning to test th e effect of bilingual English-Spanish language skills on earnings. It po sits that the inverse holds true for bilingual language skills. In areas with large Spanish la nguage enclaves, returns to bilingualism will be higher than returns to bilingualism in areas dominated by the majority language. To test these hypotheses, my study consists of two parts: the first based on statistical analysis of census data and the second based on qualitative interviews. The following sections detail this methodology.

PAGE 34

34 Quantitative Analysis US Census Data The US Census 2000 serves as the dataset for this st udy. Specifically, the study utilizes the 5% Public Use Microdata Sam ple (PUMS) fo r the state of Florida. This random sample contains nearly 796,500 cases, weight ed to represent the entire population. The census collects information on both households and individuals on topics relate d to income, language ability, ethnicity, occupation, and other variables such as age, race, education, that provide a basis for testing variations in the relationship betw een language abilities and earnings. Since 1890, every decennial US census (with the exception of the 1950 census) has included at least one question pert aining to respondents language characteristics. Stevens (1999) categorizes the history of census language questio ns into three clusters: the earliest cluster, focusing on English proficiency; the middle cluster, focusing on mother tongue, and the current cluster, focusing again on English proficiency. The shifting focus of the questions reflects changing perceptions of the relationship between language and ethnicity (Stevens 1999). For example, the first two clusters of language questions were primarily designed to elicit information on the ethnic and racial characteri stics of growing immigrant populations. For that reason, the language questions on the early and mid-century censuses pertain only to white foreign-born respondents and/or members of sel ect immigrant groups (Stevens 1999). Only the most recent censuses (1980-2000) record both th e native language char acteristics and English proficiency levels of the entire population. The Census 2000s questions on language are particularly useful for this study because they focus on both English language proficiency and language spoken at home, and they include the language characteristics of both native and foreign-born res pondents. The Census 2000 uses a three-part series of questions to gather inform ation on respondents language abilities: 1) Does

PAGE 35

35 this person speak a language other than English at home? 2) What is this language? 3) How well does this person speak English? (Very well, Well, Not well, Not at all). The Census Bureau included this series of questions specifically to identify geographic area s with a large population of people with limited English la nguage abilities in order to better assess needs for bilingual education and other social serv ices (Stevens 1999). The follo wing section details how I use these language variables to construct the linguistic concentrations for this study. Measuring the Language Enclave Due to the com parative nature of this study, it was first necessary to identify areas with high and low proportions of Spanish-speakers. The Census 2000 offers several levels of geographic disaggregation, ranging in smallest to largest order from block, block groups, and census tracts to county, state, regi onal, and national divi sions. However, in the 5% PUMS files, the Census Bureau collapses the smaller geographi c units in order to main tain the anonymity of respondents. Thus, in the 5% sample, the Publ ic Use Microdata Area (P UMA) is the smallest identifiable geographic unit, containing 100,000 or more respondents. The next level of aggregation is the Super-Pub lic Use Microdata Area (Super-PUM A). The state of Florida contains 32 Super-PUMAs, each containi ng populations of 400,000 or more. The Super-PUMAs serves as the primary geographic unit of interest for this study. While they do not offer analysis at the most disaggreg ate level, they serve th e basic purpose of this study, and their larger size offers a distinct ad vantage over the PUMA. Since the Super-PUMAs typically cover an area that encompasses one to four PUMAs, and corresponds to one or more counties as shown in the map in Figure 3-1, the chances that a respondents place of work lies within the Super-PUMA in which he lives are greater than the ch ances that a respondents place of work lies within the smaller PUMA in which he lives. This reduces th e need to construct a

PAGE 36

36 variable that proxies place of work and partly circumvents poten tial bias that may arise from wage differentials due to location. To determine areas with a high proportion of Spanish-language speakers, I selected broadly for both Spanish-speakers and monolingua l English speakers and then cross-tabulated language by Super-PUMA. This tabulation allowe d me to identify which Super-PUMAs contain large populations of Spanish-speakers and which contain populations that are primarily Englishspeaking. I then collapsed the Super-PUMAs into two categorie s: 1) high concentration of Spanish-speakers (HC) and 2) low concentrati on of Spanish-speakers (LC). I assigned SuperPUMAs with a Spanish-speaking population of 40% or less to the Low Concentration category and Super-PUMAS with a Spanish-speaking population of 60% or more to the High Concentration category. Table 3-1 shows a simp le cross-tabulation of language spoken at home by proportion of Spanish-speakers (High or Low). In the LC area, less than 10% speak Spanish; the remaining 90% are English-speakers. In HC area, almost 63% speak Spanish, while only 37% speak English. To give a geographic representation of the linguistic concentrations, Figure 3-1 presents a map outlining the Florida Supe r-PUMAs, color-coded to show the linguistic concentrations used in this study. It is obvious from the map that the Spanish-language enclave, or HC area, corresponds exactly to the Super-PUMAS that compose Miami-Dade County. The remaining Super-PUMAs in Florida constitute th e LC area. With the linguistic concentration measurement in place, I then invoke d several criteria to select the cases to be included in the regression analysis. Sample Description The sam ple used in the regression analysis consists of 16,611 cases (n =16,611) of Hispanic males aged 18-65. The age restrictions account for those respondents in their prime working years, while the sex restriction e liminates the bias created by a ge nder-related wage gap. I further

PAGE 37

37 limit the sample to respondents who report th emselves as employees, due to the welldocumented complications with measuring the wages of the self-employed (McManus 1990). In order to prevent potential bias caused by the inclusion of wages from part-time work or intermittent stints in the labor force, the samp le includes only full-time employees, defined as those who reported working an average of 35 hours or more per week and 45 weeks or more in 1999. Finally, I select for employees who work w ithin three broad cate gories of occupations: Management and Professional Positions; Sales; and Service, since occupations within these categories rely heavily on on-the-job language use. (The criteria for these occupational categories are detailed in a late r section of this chapter). To secure a sample of Spanish-speakers I selected those respondents who reported speaking a language other than English at home, and specifically, those who reported speaking Spanish at home. For the second research quest ion, which compares the earnings of bilingual Spanish-English speakers and monolingual Englis h speakers, I expanded the above sample to include those respondents who re ported speaking English only. I define bilinguals as those respondents who report 1) speaking Spanish at home and 2) speaking English very well. Description of Variables Dependent variable The Census 2000 includes inform ation on several different types of income, such wage and salary, interest, dividend, and rental income as well as total personal income I select only wage and salary income as the dependent variable fo r two reasons. First, wage and salary income refers specifically to income earned by working. Second, income from wages and salary tends to be less subject to underreporting than other form s of income (US Census Bureau 2005). Since wage and salary income tends to be documented and is generally received in consistent amounts

PAGE 38

38 throughout the year, a respondent is more likely to accurately repor t this type of income as opposed to income earned through other sources. Following standard practice, I use the natural log of wage and salary income as the dependent variable in the regressi on analysis. The log form of income is generally preferred in regression analyses for two reasons. First, the lo gged value, by bringing outli er values closer to the regression line, generates better estimates. Second, when using the natural log of income, the b coefficients in the regression analysis can be interpreted as percentage s of returns to earnings (Lovell 1989; Hardy 1993). Independent variable English prof iciency is the primary indepe ndent variable. Respondents self-report their English-speaking ability by choosing from four possi ble responses: not at al l; not well; well; and very well. In the regression model, rather th an inserting the Englishlanguage variable as a single variable in the form of an index (0-3), I code those responses into three dummy variables: Not well (not well=1; otherwise=0); Well (well=1; otherwise=0), and Very well (very well=1; otherwise=0). Those who reported not speaking any Englishthe Not at all groupare used as the comparison group. Converting the Englishlanguage ability variable into multinomial dummy variables allows me to observe how the effect of English language ability on earnings varies by proficiency level. In the second re gression analysis, which seeks to compare the earnings of bilinguals and monoli nguals, I insert monolingual (Engl ish-only) Hispanics into the sample, and use them as the reference group. I then use the following four dummy variables for English speaking ability: Not at all (not at all=1; otherwise=0); Not well (not well=1; otherwise=0); Well (well=1; otherwise=0), and Very well (very well=1; otherwise=0).

PAGE 39

39 Control variables Selection of the control variables used in the an alysis accounts for personal characteristics, human capital traits, migration circumstances, and labor characteristics that are likely to influence wage and salary income. The first set of control va riables represents common human capital traits, including age, educational attainment, work experience, and work experience-squared. Age refers to the respondents age in years. Educational attainment represents the res pondents highest level education completed. An individuals labor mark et experience also tends to be a strong human capital trait that is positivel y correlated with earnings (Chiswick 1991). However, since the Census does not provide a variable that explicitly measures th e number of year s the respondent has been in the labor force, I construct the standard proxy measure for work experience by computing the respondents age less his educa tional attainment less 6 years (age-educational attainment-6) (Mincer 1974). The next set of control va riables corresponds to personal characteristics, and includes race, nationality, and linguistic isolation. The Census Bureau records seven mutually exclusive categories for race, including a Two or more races category introduced in the 2000 Census. For the purposes of this study, I ha ve collapsed these categories into White and Non-white in order to use race as a binomial dummy variable (White=1 and Non-white=0). Since previous research has shown that nati onality can act as a social capital trait that strongly affects earnings (Borja s 1982, Tienda 1983), I incorpor ate the following multinomial dummy variables for national origin: Puerto Rican (Puerto Rican=1 and otherwise=0); Mexican (Mexican=1 and otherwise=0); and Other Hispanic (Other Hispanic=1 and otherwise=0). Since Cubans comprise the largest group of Hispanics in Florida, they are used as the reference group for comparison. Finally, I include linguistic isolation, a variable that indicates a respondents

PAGE 40

40 exposure to English in the home. The Census Bu reau defines a linguisti cally isolated household as one in which no members aged 14 or older speak English only and no members aged 14 or older speak a non-English language and speak Englis h very well. That is to say, all members aged 14 or older in linguistically isolated households experience some degree of difficulty with English. I code this as a binomial dummy va riable (not linguistically isolated=1 and linguistically isolated=0). The control variables for mi gration characteristics are birthplace, citizenship status, and years in the US. Prior research finds that US origin, US citizenship, and length of residence in the US are positively associated with earnings of non-native English speakers (Chiswick and Miller 1992; 2002). I thus insert dummy variables for birthplace (US=1 and abroad=0) and citizenship status (US citizen=1 and non-US citizen=0). Years in the US represents the total number of years that a respondent has lived in the US. The final set of control variables acc ounts for variations in respondents labor characteristics. Since wage income often varies by number of hours and weeks worked, I include the usual number of hours worked per week and the total number of weeks worked in 1999. The third labor characteristic variable accounts for occupatio n-related wage differentials. The Census Bureau recorded 992 different types of occupations in the 2000 Census. In order to convert these into more manageable categories, I follow the Census Bureaus classification scheme and collapse the occupations into the seven principal categories: Management and Professional; Service; Sales and Office; Farming, Fishing, and Fore stry; Construction, Extraction, and Maintenance; Produc tion and Transportation; and Mili tary. However, as noted in the sample description, I select the three occupationa l categories that most use language on the job: management and professional positions; sales; and service. I then construct multinomial

PAGE 41

41 dummy variables for these occupations by coding sales (sales=1 and otherwise=0) and service (service=1 and otherwise=0). Professional and management positions, which tend to be the highest-paid occupations of the three categories, are left out of the equation and used as the comparison group. Statistical Model/Data Analysis A sim ple Ordinary Least Squares (OLS) regr ession analysis allows me to observe the effect of English language ability on the natural log of income, net of the effects of education, age, sex, race, hours and weeks worked, occupation, work experience, national origin, citizenship status, years in the U.S., and linguistic isolation. The analysis of the relationship between earnings and the aforementioned wage -predictors is modeled in Equation 3-1. Y = + 1age + 2racei + 3educ+ 4exp + 5exp2 + 6weeks + 7hours+ 8occi + 9bpl + 10citizen + 11years+ 12lingisol + 13hispani + 14engabili (3-1) In this equation, Y represents the natural log of wage and salary income; age is the respondents age in years; raceis refers to binary racial categories; educ represents highest level of education attained; exp is the proxy for work experience; exp2 is work experience-squared; occis refer to different o ccupational categories; weeks is the number of weeks worked in 1999; hours is the usual number of hours worked per week; bpl indicates a respondents birthplace; citizen indicates a respondents citizenship status; years represents the number of years a respondent has lived in the US; lingisol indicates whether or not the respondent lives in a linguistically isolated household; hispanis represent national origin categories; and engabilis represent different levels of English-speaking ability. Regressions are conducted separately fo r the two categories of Super-PUMAs. A comparison of the regression models for the high and low concentrations will allow me to determine if the effect of English language abi lity on earnings varies according to the percentage

PAGE 42

42 of Spanish-speakers in the area. Since I argue that English language ability will be a more important determinant of earnings in areas where there are fewer Hisp anics, I expect the effect of English language ability on earnings to be greate r in areas with a low concentration of Spanishspeakers than in areas with a high concentration of Spanish-speakers. Specifically, I predict that, net of the controls for occupa tion and human capital variables, the b coefficient for English language ability will be significantly higher in the low concentration model than in the high concentration model. In the second regressi on analysis, which examines the effect of bilingualism, I expect the b coefficient for to be higher for bilingual s in areas with a high concentration of Spanish-speakers than in areas with a low concentration of Spanish-speakers. Qualitative Analysis Valuable as the em pirical analysis may be in terms of the magnitude of effects, and how the effects vary by labor market context, it is also useful to examine the behavioral and attitudinal models that underlie my hypothesis. Specifically, it is impor tant to verify that employers give less importance to workers langua ge abilities in enclave economies. In order to do this, I added a qualitativ e component to this study. The qualitative analysis consisted of interv iews with employers and managers who make hiring and promotion decisions a nd frontline workers from employment and staffing agencies familiar with the demands of the local labor markets. The sample consisted of informants associated with businesses in the service sector since such businesses rely heavily on oral communication with the public. Af ter establishing initial contacts, I used snowball sampling to identify subsequent key informants. Through semi-structured interviews, I gain ed information about the behavioral and attitudinal patterns associated with language and earnings. Specificall y, I strove to elicit information about how employees language pr oficiency influences employers decisions

PAGE 43

43 regarding earnings and how and why criteria for such decisions vary by labor market context. The general interview guide cons isted of questions related to employees language ability and work performance; language-based delegation of positions and tasks; payment, raise, and bonus criteria; language proficiency vs. language use; linguistic composition of workforce and clientele; language-based discrimination in the wo rkplace; and bilingualism in the workplace. The conversational nature of th e semi-structured interviews f acilitated the fl ow of ideas regarding these topics and allowed others to emer ge. I conducted a total of ten interviews in Miami-Dade and Broward counties. Although the qualitative analysis played a secondary role in this study, it nonetheless provided further insight into the relationship between language a nd earnings and was particularly helpful in identifying new variables to include in the quantitative analysis. Consequently, I do not discuss or analysis the inte rviews at length in the proceeding chapters, but instead include small excerpts as supporting material for the qua ntitative findings. To maintain anonymity, the names of the informants have been changed and their place of employment has been concealed. Methodological Caveats Despite the high quality of the Census 2000 (w hich is repo rtedly the most accurate US census in history) and the applicability of its que stions to this study, it is nevertheless important to interpret findings with the following caveats in mind. Wage and salary income, the dependent variable, is self-reported by respondents. Since answers to the income questions are generally based on memory, the Census Bureau (2005) cautions that respondents often forget precise amo unts of income received. This is especially true in the case of undocumented and informal earnings, such as tips. As a result, wage and salary income tends to be underreported.

PAGE 44

44 English speaking ability, the primary independent variable of interest, also suffers from three main limitations. First, like income, language ability is self-reported by the respondent. The census operationalizes English language ability as speaking ability, a definition that is broad and subject to interpretation. Sin ce there are no explicit and unif orm criteria for gauging language ability, a respondent may claim any level of prof iciency he wishes. Thus, English language ability, unlike wage and salary income, tends to be overestimated by respondents (Siegel 2001). The over-reporting of English language ability is more pronounced in the very well and well categories (Stevens 1999). The second limitation of the English-language ab ility variable is that it refers only to English speaking proficiency. As with most r ecent surveys and censuse s, the 2000 Census does not ask about the ability to read English. The inclusion of a question on reading ability would offer a more precise analysis since most occupa tions typically require both reading and speaking skills (Chiswick 1991). Third, the fact that the questionnaire is writ ten in English presents an inherent problem for respondents with limited English abilities. Each household must have at least one member that has some English literacy in order to fill out the questionnaire or understand what alternate resources the Census Bureau provides to as sist non-English language speakers with the questionnaire. The Census Bureau has implem ented various procedural measures to reduce undercount resulting from such lang uage issues. For example, th e census questionnaires are now available in six languages, includ ing Spanish. Language assistance guides are also available to provide assistance with the completion of the que stionnaire in 49 languages. Nevertheless, for obvious reasons, this issue lingers as a potential source of undercount or misreporting.

PAGE 45

45 The measurement of bilingualism also warra nts a word of caution. Stevens (1999) notes that many people who report speaking English very well and report speaki ng another language at home, may not necessarily be completely fluent in the non-English language. This pertains especially to children of immigrants, who often experience fi rst-language attrit ion, or loss of a second language. In addition, data from past censu ses reveal that respondents have a tendency to overreport the use of a non-English language at home. Over-reporting of language ability tends to be more common among native-born American s than among foreign-born residents (Stevens 1999). Problems with reporting the use of non-E nglish languages may arise from ambiguities in the phrasing of the census question Does this person speak a language other than English at home? The question does not specify the exte nt and frequency with which the non-English language is used at home. Additionally, at home may also be a source of ambiguity, especially for immigrants who may interpret home as home country (Siegel 2001). The issue of undercount again arises with the Hispanic variable. The census undercount of Hispanics is well-documented (Duany 1992; Evans 2001). However, the 2000 Census boasts the lowest reported levels of undercount of minority groups (Evans 2001). Indeed, the undercount estimates of Hispanics dropped fr om 4.99% on the 1990 Census to 2.58% on the 2000 Census (Evans 2001). Researchers have theorized about a variety of cultural and behavioral factors that may c ontribute to this undercount. Dua ny (1992) posits that undercount of Hispanics often stems from five main causes: (1) disbelief in the confidentiality of the census (2) distrust of government authorities (3) fear of losing public assistance (4) fear of deportation among undocumented immigrants (5) cultural differences in de fining household structure (p.1)

PAGE 46

46 Other research has shown that the placement of the census question on Hispanic origin also influences the count. The 2000 Census marked a change in the sequencing of questions on race and Hispanic origin. In 1990, the question on Hispanic origin directly preceded the question on race; the 2000 Census reversed the order of these questions (Gri eco and Cassidy 2001). Analysis of this sequencing shows that the order of ques tions used in 2000 significantly reduced the nonresponse of Hispanics (US Census Bu reau 1999; Grieco and Cassidy 2001). Another minor limitation of this study stems from the measurement of linguistic concentration. While other st udies have invoked the Hispanic variable (McManus 1990), the linguistic isolation variable (C hiswick and Miller 1992), and othe r methods to identify language enclaves, this study defines linguistic concen trations more broadly by using simply the percentage of Spanish-speakers who live in a certain area, regardle ss of their linguistic isolation status. While this may be a rather crude m easurement of language enclaves, it nevertheless serves the basic purposes of this study.

PAGE 47

47 Table 3-1. Language spoken by concentra tion of Spanish-speakers, Florida 2000 Concentration of Spanish-speakers Language spoken Total Spanish English Low N 59,347 546,704 606,051 % 9.8 90.2 100.0 High N 59,149 35,272 94,421 % 62.6 37.4 100.0 Source: US Census 2000 IPUMS 5% Sample

PAGE 48

48 Figure 3-1. Florida linguistic composition. Public Use Microdata Sample (PUMS) files. US Census Bureau, Census 2000. Areas with low concentration of Spanish-speakers (LC) Areas with high concentration of Spanish speakers (HC)

PAGE 49

49 CHAPTER 4 RESULTS This chapter presents the findings from the qua ntitative ana lysis and qualitative interviews. I first outline descriptive statistics for the variable s used in the analysis an d then report the results from the OLS regression analyses for both rese arch questions. Wherever relevant, I include excerpts from the interviews as supporting evidence of the empirical results. Summary Statistics Tables 4-1 and 4-2 provide the sam ple means and standard deviations for the dependent and independent variables used in the regressi on analysis for the Low Concentration (LC) and High Concentration (HC) models, respectively. With the exception of certain variables discussed in further detail below, the statistics sh ow little variation between the two contexts. The annual wage and salary income in the LC areas averages $30,717 (natural log 10.101). The mean wage and salary income in the HC area is slightly higher at $35,507(natural log 10.221) per year. The distribution of English-speaking abilities, howev er, is fairly similar across the two concentrations. In both areas, ju st over half of the respondents report speaking very well. Approximately 20% of respondents in each concentration report speaking well, 17% report speaking not well, and about 9% report speaking no English at all. The mean and standard deviations for th e human capital, labor characteristics, and migration variables are fairly uniform across the concentrations. The Hispanic origin composition does, however, vary by concentration. Cubans compose the majo rity (52.3%) of the sample in HC area, while Mexicans, Puerto Ricans, and other Hispanics rank as the most populous Hispanic origin groups in the LC areas. The percentage of foreign-born respondents also is greater in the HC areas.

PAGE 50

50 Research Question One: The Effect of Language Enclaves on Returns to English The first m odels regress the log of wage and salary earnings on Englis h-speaking ability of non-native speakers. Table 4-3 displays the results from the regression models, sorted by linguistic concentration. The b coefficients, reported in the fi rst and third columns, show the effect of English language ability in the two linguistic contexts. Since the dependent variable is logged, the b coefficient is interpreted as percent of change in the dependent variable (Y) for each unit change in the independent variable (X), when X is a continuous measure such as age or years of school. In the case of dummy variables, such as those used for English-speaking ability, the b coefficient is interpreted differently since th e dummy variables have discrete values of 0 and 1 (Hardy 1993). In this instan ce, the antilog of the regression coefficient is used to measure the percent of change associated with belonging to the category of interest, relative to the omitted reference category. As expected, results from these models poi nt to a positive relationship between Englishspeaking ability and earnings in both English-domi nant and Spanish-dominant areas in Florida, as indicated by the positive regression coeffi cients in both models. An initial observation confirms the importance of treating English langu age proficiency as a set of discrete dummy variables, rather than an ordina l variable with values ranging from 0-3. Had I done the latter, and thereby generated a single b coefficient for lan guage proficiency, it would not be evident that the returns to language ability differ by proficienc y level. For example, the findings show that there are no significant differences in the ea rnings returns to la nguage proficiency for respondents in the not well and n ot at all categories. However, when compared to the not at all category, the earnings returns to language proficiency ar e greater for those in the well category and greater still for those in the very well category. In shor t, the findings clearly

PAGE 51

51 show that the effect of language ability on earnings is nonlinear, a pattern that is evident in both labor market contexts. When examined comparatively, the two models show unexpected results. The b coefficients for English-speaking ability vari ables in the HC model are higher than the b coefficients in the LC model. Fo r example, in the LC model, the b coefficient (column 1) for those respondents who speak Eng lish well is .052. Those w ho speak English well earn roughly 5.2% more than the reference group, who speaks no English at all. In the HC model, the b coefficient (column 3) for t hose respondents who speak English well is .076. This means that those who speak English well earn appr oximately 8% more than those who speak no English at all. The comparison of these two re gression coefficients indi cates that, net of the effects of the selected control variables, the impact of intermediate English proficiency on earnings is greater in the HC areas. The difference in the magnitude of the effect of English ability on earnings is even more pronounced for those have a higher level of English-speaking prof iciency. In the LC areas, respondents who speak "very well" earn approxim ately 14.3% more than the reference group, as indicated by the b coefficient of .134. However, in the HC model, those who speak very well earn nearly 25% more than their English-de ficient counterparts, as indicated by the b coefficient of .219. Only the Not well group is discounted from the analys is since its relationship with earnings does not prove statistically significant. These findings suggest that, in the case of the well and very well profic iency levels, English has a greater effect on earnings in the HC areas. In other words, in the labor ma rket in Florida, English is actually more important in areas with a high concentration of Spanish-speakers. Such findings run contrary to my research hypothesis and contradict McManuss (1990) study that found th at larger enclave size lowers

PAGE 52

52 returns to earnings and Chiswick and Millers (2002) study that conc luded that non-native English speakers have greater earnings opportuni ties in areas with a high concentration of Spanish-speakers. Many comments from interviews in Miam i-Dade support these findings. For example, Dan, a recruiter at a staffing agency in Nort h Miami, notes that, although most of his job applicants are bilingual: English is still very important in Miami. A lot of the clients that we deal with are not in the Latin or Spanish-speaking areas; theyre mo stly English-speaking, so they need to be able to communicate with thes e guys and let them know what needs to get done. If they cant communicate with them, they ar e going to send them off the job. Victoria, a recruiter for a hotel and food service employment ag ency that staffs companies in several counties in South Flor ida, seconds this opinion. Englis h is a must, she says. She explains that although all of the companies she staffs require English-speaking workers, her job applicants are mostly Spanish or Creole-speakers. She adds: Sometimes they dont even know a little bit of English; they ca nnot answer me. They have a companion with them to talk with me. I tell them, how can I communicate with them, when I will be calling them personally to send th em on a job? Usually what I give them are dishwashing jobs if theyre insistent. And sometimes jobs in the hotels, if the hotel manager or supervisor speaks Spanish. But if not, they will tell me, Please dont send this person, because we cant communicate Thos e jobs are usually the lowest paid, minimum wage. Steve, a recruiter at a staffi ng agency in Miami, likewise emphasizes the importance of English in the workplace. In addition to j ob experience, references, work ethic, and presentability, he cites English as one of the top qualifications that he seeks in applicants They absolutely have to have some proficiency in English He explains further that language proficiency indirectly influences pay because communication affects job performance, and job performance is what ultimately determines raises and bonuses.

PAGE 53

53 Indeed, relative to the other independent variables in the model, English language proficiency ranks among the most important dete rminants of earnings. To demonstrate this, I call attention to the standardized Beta coefficien ts, which offer a means of comparing the effect English ability to the effect of the other indepe ndent variables on earnings. Since these variables serve primarily as controls, I do not discuss their impact at lengt h. Rather, I mention briefly the magnitude of their effects in or der to give a general idea of the relative impact of English proficiency on earnings. Similar to findings from past studies (e g. Chiswick and Miller 2001, Tienda 1983, Borjas 1982), both models in Table 4-3 confirm the positive correlations between earnings and the standard human capital variab les: age, education, a nd work experience-squared1. However, in the HC model, the ability to speak English very well actually exerts a greater influence on earnings than educational attainment. In fact, wi th a Beta coefficient of .165, the ability to speak English very well functions as the fourth most important determinant of earnings, behind age (Beta=.731); work experience-squared (Beta= -.655); and hours worked (Beta=.177). Though its relative impact is lesser in the LC model, the ability to speak English very well still ranks as an important determinant of earnings. When compared to English-speaking abil ity, the other control variables exert a less powerful, but nonetheless significant, influence on earnings. For example, US nativity, US citizenship, and length of residence in the US are all positively asso ciated with earnings. Type of occupation also acts as a significant predictor of earnings. Workers in sales and in the service industry tend to earn less than t hose in professional positions. Such findings resonate with the aforementioned excerpt from the interview with Victoria, in which she indicates that workers 1 Collinearity exists between the age and work experience variables.

PAGE 54

54 with low levels of English proficiency are largely relegated to the lowest-paying jobs in the service industry. The models also show evidence of wage disc rimination along racial and ethnic lines. In both the HC and LC areas, White Hispanics tend to earn about 4.7% more than Black or other Hispanics. However, in relative terms, race is one of the least important determinants of earnings, as indicated by the Be ta values of .037 and .023 in th e second and fourth columns, respectively. Ethnic networks, however, do appear to have a strong impact on earnings. Indeed, the perceived effect of ethnic ne tworks emerged as a recurrent th eme in several interviews. The following quote from Raul, an administrative personnel manager for a local university, illustrates one possible rationale for the relationship between immigrant and/or ethnic networks and earnings and how language proficie ncy factors into that relati onship. When asked about his willingness to hire workers with limited English proficiency, his rationale recalls his own immigrant past: I look at myself: when I came to this country, someone gave me a chance and I was able to move up through the system, and I like to do th e same for people that are out there in the same position that I was thirty-some years ago. National originwhich serves as an opera tional definition of ethnic networks in particular stands out as a signi ficant determinant of earnings. The results from both regression models suggest that, among Hispanics, being Cuban is an added advantage in the labor market in Florida. The negative regression coefficients for the Hispanic origin variables indicate that, all other things being equal, Cubans earn more than other Hispanics in Florida. Compared to Puerto Ricans, Floridas second most populous Hispanic group, Cubans earn about 10% more in the HC areas and 14% more in the LC areas. Anna, a former career counselor and a current immigration lawyer at a legal services organization in Downtown Miami, observes:

PAGE 55

55 Sometimes I see Cubans come in; they hardly speak any English and they have only been here for a short time, but they already have great jobs, working at FIU, at hospitals, offices... Her comments hint at the ways in which national solidarity can function as a social capital trait that yields positive labor market outcomes. Together with the English-speaking ability variables, these additional independent variables form a statistical model that explains a lmost 36% of the variance in earnings in the LC areas (R2=.356) and 33% of the variance in earnings in the HC areas (R2=.331). Both the empirical evidence from these models and qualita tive support from the in terviews underscore the importance of English proficienc y in the labor market regardless of linguistic concentration. However, many interviewees, both in HC and LC areas, also highlighted the simultaneous importance of Spanish in the labor force. The next section presents the findings that from the models comparing the effects of bilingualism on earnings in each linguistic concentration. Research Question Two: Returns to Bilingualism The second set of regression analyses in corporates m onolingual English-speaking Hispanics in the sample in order to examine how the effect of bilingualism on earnings varies by linguistic concentration. This analysis uses the same regression equation used to test the first research hypothesis, but uses mono linguals as the reference group. The Very Well category in both tables refers to bilingual speakers, since bilingual respondents are defined as those respondents who 1) speak Spanish at home and 2) speak English very well. Table 4-4 presents the results, sorted by linguistic concentration. In this analysis, the b coefficient again serves as the main tool for comparison between the two concentrations. The LC model suggests that monolingual English speakers earn more than their bilingual counterparts, as indicated by the b coefficient of -.03 for the Very Well category in Column 1. Yet, findings in this case are

PAGE 56

56 not statistically signif icant, and thus cannot offer valid suggestions about the effect of bilingualism on earnings in LC areas in Florida. The HC model, however, shows strikingly diffe rent results. As predicted, bilingual language skills are positively associated with ea rnings in the HC area, as indicated by the b coefficient of .066 in the Very Well category in Column 3. All other factors being equal, bilingual Hispanics earn more than their monolingual English counterparts in the enclave. However, the positive relationship between bilin gualism and earnings only holds true for the fully bilingual speakers. Spanish-speakers who have intermediate English proficiency earn less than monolingual English Hispanic s, as shown by the negative b coefficient for the Well category in column 3. Comments from the qualitative interviews echo these empirical findings. For example, Kevin, a manager of an administrative and techni cal staffing agency in Hialeah, FL, notes that more often than not clients (employers) in th e South Florida area speci fically request bilingual English-Spanish speakers for positions. The applicant pool at his agency consists of monolingual English speakers, monolingual Spanis h-speakers, and bilingual English/Spanishspeakers. He explains that for jobs that requ ire little interaction w ith the public, monolingual language skills, Spanish-only or English-only, suffi ce. However, for most positions, bilingual skills are preferred since the businesses he staffs deal not only with bili ngual clients in South Florida, but also with monolingua l Spanish-speakers in Latin America and th e Caribbean. He adds: In Miami, bilingualism is pervasive. A good command of the English and Spanish languages is necessary. Generally, bilingual pos itions pay higher because there is more responsibility involved. Sometimes it creates issu es with our recruiting efforts we end up ignoring a lot of folks who do not speak Spanis h. There are those pe ople that are Englishonly who would not qualify. I had to learn Spanish myself when I came to Miami I use it everyday and it allows me to perform my job better. Yet, when you are limited to the

PAGE 57

57 minority language, (which I dont know if Spanish is anymore in Miami), but, in my experience, Spanish-only would generally pay lower than English-on ly. So, it goes both ways, and may be something that is unique to this area [Miami-Hialeah]. This excerpt highlights the labor market a dvantages bilingualism and disadvantages of monolingualism in the language enclave. Anecdotal evidence from a recent Miami Herald also points directly to the advantage that bilinguals have in the Miami-Dade labor market. The article quotes a Miami business executive about the language abilities one of his former Cuban-American employees Professionally, she was very good. But, she was almost incapable of writing Spanish. He eventually replaced her with a fully bilingual Puerto Rican secretary (Fernandez 2008). His experience shows the need for employees with high profic iency levels of both English and Spanish. Even Victoria, whose quote in the previous section stressed the importance of English in the labor market, and who late r expressed her belief that non-native speakers should speak English because they are in America, acknowledge s the growing importance of Spanish in the labor market in South Florida. She says, I know a little bit of Spanish. You have to at least learn the basics, because they [Spa nish-speakers] are everywhere. Thus, in this case, findings partially support the research hypothesis that predicted that bilinguals would earn more than monolinguals in th e enclave. Such findings run contrary to past studies (Pendakur and Pendakur 2002) that conclude that knowle dge of an unofficial, minority language, garners no additional earnings advantage in the labor market. However, because findings from the LC model were not significant, I cannot compare the effects of bilingualism on earnings in the two linguistic con centrations. Nevertheless, these findings, as well as those from the first research question, have important implicatio ns that I discuss further in the next chapter.

PAGE 58

58 Table 4-1. Mean and standard deviation of variab les used in Low Concentration model. Hispanic males, age 18-65. Variable Description Mean SD Wage & salary Income Total annual wage & 30717.2 30770.02 salary income in dollars Log of wage & Natural log of wage 10.101 0.631 salary income & salary income Age Age of respondent 36.2 10.828 Race Dummy variable 0.63 0.484 (ref=Black/Other) Educational attainment Years of school 9.42 3.454 Work experience Age-education-6 32.784 10.864 Work experience squared (Age-education-6) 2 1360.255 898.73 Occupation Dummy variables Management (ref=Management) 0.1831 0.387 Sales 0.1494 0.357 Service 0.1777 0.382 Weeks worked Total annual weeks worked 51.18 1.84 Hours worked Typical hours worked per week 45.12 8.507 Birthplace Dummy vari able 0.22 0.417 (ref=Abroad) Citizenship status Dummy variable 0.6 0.489 (ref=Non-US citizen) Years in the US Total years lived in the US 12.32 12.545 Linguistic isolation Dummy variable .72 0.447 (ref=Linguistically isolated) Hispanic origin Dummy variables Cuban (ref=Cuban) .134 0.34 Mexican .299 0.458 Puerto Rican .267 0.442 Other Hispanic .301 0.458 Speaks English Dummy variables None (ref=None) .08 0.271 Not well .177 0.382 Well .212 0.409 Very well .531 0.499 Source: 2000 US Census IPUMS 5% Sample: Florida N=8848; Ref=Reference category; coded as 0

PAGE 59

59 Table 4-2. Mean and standard deviation of variab les used in High Concentration model. Hispanic males, age 18-65. Variable Description Mean SD Wage & salary Income Total annual wage & 35507.16 37176.76 salary income in dollars Log of wage & Natural log of wage 10.221 0.662 salary income & salary income Age Age of respondent 39.42 11.437 Race Dummy variable 0.87 0.34 (ref=Black/Other) Educational attainment Years of school 10.49 3.072 Work experience Age-education-6 34.934 11.825 Work experience squared (Age-education-6) 2 1360.255 898.73 Occupation Dummy variables Management (ref=Management) 0.253 0.434 Sales 0.2361 0.425 Service 0.1261 0.33198 Weeks worked Total annual weeks worked 51.14 1.875 Hours worked Typical hours worked per week 44.68 8.143 Birthplace Dummy vari able 0.13 0.332 (ref=Abroad) Citizenship status Dummy variable 0.55 0.497 (ref=Non-US citizen) Years in the US Total years lived in the US 15.80 12.624 Linguistic isolation Dummy variable .73 0.446 (ref=Linguistically isolated) Hispanic origin Dummy variables Cuban (ref=Cuban) .5228 0.5 Mexican .033 0.179 Puerto Rican .0652 0.247 Other Hispanic .38 0.485 Speaks English Dummy variables None (ref=None) .0920 0.289 Not Well .1735 0.379 Well .223 0.416 Very Well .511 0.499 Source: 2000 US Census IPUMS 5% Sample: Florid a, N=7763; Ref=reference category; coded as 0

PAGE 60

60 Table 4-3. Returns to English-sp eaking ability. Log of income re gressed on English ability and other selected variables. Low concentration High concentration of Spanish-speakers of Spanish-speakers Independent variables b Beta b Beta Constant 6.904 6.957 Age .044* 0.762 .042* 0.731 Race (ref=Black/other) .048 0.037 0. 044** 0.023 Educational Attainment .021* 0.117 0.018* 0.082 Work Experience Work Experience Squared 0* -0.64 0* -0.656 Occupation Service -.158** -0.096 -.241* -0.121 Sales -.038* -0.022 -.098* -0.063 Weeks Worked .026* 0.077 0.022* 0.062 Hours Worked per Week .011* 0.153 .014* 0.178 Citizenship Status .060* 0.047 .091* 0.068 Birthplace (ref=Abroad) .1 00* 0.066 .135* 0.067 Years in the US .005* 0.108 .007* 0.14 Linguistic Isolation (ref=Not Isolated) 0.028* 0.02 0.042** 0.028 Hispanic Origin (ref=Cuban) Mexican -.133* -0.096 0.024 0.006 Puerto Rican -.153* -0.107 -.106* -.039 Other Hispanic -.60* -0.044 -.038* -0.028 English Ability (ref=Not at all) Not Well 0.012 0.007 -.039 -0.022 Well .052** 0.034 .076* 0.048 Very Well 0.134* 0.106 .219* 0.165 R2 0.356 0.331 N 8847 7763 Source: 2000 Census 5% IPUMS Sample: Florida; *Significant at .001 or less; **Significant at .05 or less. Sample includes Hispanic males, age 18-65, who reported speaking Spanish at home

PAGE 61

61 Table 4-4. Returns to bilingualism. Log of in come regressed on language ability and other selected variables. Low concentration High concentration of Spanish-speakers of Spanish-speakers Independent variables b Beta b Beta Constant 7.056 7.053 Age 0.036* .611 .031* 0.539 Race (ref=Black/other) .050 0.038 0. 042** 0.021 Educational Attainment 0.020* 0.108 0.030* 0.14 Work Experience Work Experience Squared 0* -0.588 0* -0.466 Occupation Service -.158* -0.094 -.240* -0.12 Sales -.03** -0.017 -.102* -0.065 Weeks Worked .028* 0.081 0.022* 0.063 Hours Worked per Week .011* 0.152 .015* 0.179 Citizenship Status .063* 0.047 .099* 0.074 Birthplace (ref=Abroad) .1 04* 0.075 .144* 0.075 Years in the US .005* 0.101 .007* 0.143 Linguistic Isolation (ref=Not Isolated) 0.028 0.019 0.041** 0.027 Hispanic Origin (ref=Cuban) Mexican -.118* -0.084 0.024 0.007 Puerto Rican -.160* -0.111 -.097* -.037 Other Hispanic -.063* -0.046 -.033* -0.024 English Ability (ref=English only) Not at all -.121* -0.048 -.138* -0.059 Not Well -.113* -0.63 -.181* -0.102 Well -.079* -0.048 -.072* -0.044 Very Well -.03 -.002 .066* 0.05 R2 0.36 0.332 N 10404 8084 Source: 2000 Census 5% IPUMS Sample: Florida; *Significant at .001 or le ss; ** Significant at .05 or less Sample includes Hispanic males, age 18-65, who reported speaking Spanish at home or speaking English only.

PAGE 62

62 CHAPTER 5 DISCUSSION Chapter 4 presented regression es timates that su ggest: first, that English is more important to earnings in the minority language enclave; and second, that bilinguals earn more than their monolingual English counterparts in the minority language enclave. The opinions expressed in the qualitative interviews also largely support th e empirical evidence. Together, these findings offer valuable insight into the relation ship between language, labor markets, and immigrant/minority incorporation. Consequentl y, they pose important theoretical and policy implications for these topics. This chapter discus ses those implications in the following order: implications for labor markets and incorporation theories; app lication of findings to current policy issues; and suggestions for further research. Theoretical Implications Language, Earnings, and Labo r Market Characteristics Returns to English language proficiency This study identifies the Super-PUMAS in Miami-Dade County as a Spanish language enclave. The rem aining Super-PUMAS in Florida qualify as areas with a low concentration of Spanish speakers. Given that the majority (62 %) of inhabitants in Miami-Dade County speaks Spanish at home, intuitive reason ing would suggest that English is less important in the labor market in this area. However, results reveal ot herwise. The regression estimates indicate that English proficiency has a greater impact on earnings in this ar ea. Such findings contradict McManuss (1990) and Chiswick and Millers (2 002) observed reduction in returns to English language proficiency in areas with a high minor ity language concentration, but support Dvila and Moras (2003) findings that English is more important to earnings in the labor market in minority language concentrations. Although these re sults run contrary to my research hypothesis

PAGE 63

63 and are seemingly counterintuitive, they are ne vertheless statistically significant, and thus provide insight into the relationship betw een language, earnings, and labor market characteristics. They suggest that the relationship between la nguage proficiency and earnings does indeed vary by context, but not always in the predicted ways The main question raised by such findings then becomes: why does English language proficiency have a greater effect on earnings in areas with a larg e Spanish language enclave? One possible explanation could be rooted in labor economics theory. As the supply of Spanish-speaking workers increases, the demand for their labor, and consequently, the value of their labor, decreases (Bloom and Grenier 1992 ). Similarly, as the population of Spanishspeakers increases in a given area, so does the competition among them in the labor market. In an applicant pool filled with Spanish-speaking ca ndidates with similar education and experience qualifications, high English literacy may be th e most important dis tinguishing factor among them. English language proficiency would then asse rt itself as a more mark etable trait in areas with a high concentration of Spanis h-speakers, and would thus be wo rth more within the enclave. In areas with a low concentration of Spanish-sp eakers, where there is a smaller pool of Spanishspeaker workers, the labor market cannot be as discriminating with regard to language skills. Bloom and Grenier (1992) find a similar effect in their study of earnings of Hispanic males in the US in 1970 and 1980. A second, related explanation could point to effects of the Cuban ethnic enclave economy in Miami. As previously mentioned, this study do es not utilize the ethnic enclave economy as a unit of analysis; rather, it focuses more broadl y on linguistic enclaves. However, given the prominence of the Cuban ethnic enclave economy in Miami, its presence may affect findings. Some research suggests that wh ile the ethnic enclave economy may have an insulating effect for

PAGE 64

64 new immigrants, it may eventually serve as a mo bility trap for immigrant employees (Sanders and Nee 1987; Booth 1998). Assuming that the et hnic enclave economy acts as a mobility trap and that it exists as a smaller unit within a larger linguistic or ethnic enclave, one could argue that English is a more valuable asset in the areas with a high minority language concentration because it serves as way to access better payi ng jobs outside the ethnic enclave economy. An alternate strategy for future analyses may be to control for the presence of the ethnic enclave economy within a larger ethnic or linguistic enclave. This measurement would require the simultaneous inclusion of language and place of work variables to determine the chances that the respondent works in the ethnic encl ave economy (Zhou and Logan 1989). A third possible explanation could be rela ted to an element that, for lack of a corresponding variable, is not considered by this studypublic opinion of immigrants/foreign language speakers. Past analyses have shown th at public opinion toward immigrants affects the incorporation process (Tienda 1983). Public se ntiment toward immigrants/foreign language speakers often varies according to the percentage of th e population that is fo reign-born in a given area (Stolzenberg 1990). De Jong and Tran (2001) find Miami residents to be largely receptive towards immigrants. Other studies, however, cont end that the large fore ign-born population in Miami is a source of contempt among many city residents. For example, Portes and Stepick (1993), through their use of competing discourses from the citys main ethnic groups, provide ample evidence of the deep-rooted tensions between Hispanics (particularly Cubans), Blacks, and Whites that have existed in Miami, at least in past decades. Given the high symbolic value attached to English in the US, prejudiced employers may use lack of English language skills as a basis for culture-based wage discrimination (P endakur and Pendakur 2002). Given the large foreign born population in Miami-Da de, it is possible that this t ype of discrimination occurs

PAGE 65

65 more frequently in this area than in areas with a smaller percentage of immigrants. Further research is needed to valida te this line of reasoning. Finally, methodological decisions may also explain results. Fi rst, the decision to include foreign and native-born Hispanic males in my samp le may impact results. Other similar studies (Dvila and Mora 2000; Hand 2006) run separate regression models for immigrants and natives and find that the earnings returns to English language proficiency differ for the two groups. I defend my decision to include both groups simultaneously since my primary focus is language, not birthplace. Additionally, OLS re gression analysis allows me to control for the effects of birthplace. Nevertheless, I ac knowledge that separate regressions for immigrants and native may alter results. Second, results may also be sensi tive to the linguistic conc entration measurement. As detailed in the literature review, past an alyses have invoked other methods to measure the linguistic enclave, such as Hisp anic origin variable (McManus 1990). Again, since language is the focal point of this study, I choose to define the enclave by the language spoken at home variable. It is possible that different definitions of the enclave will produce different results. Returns to bilingualism Findings from second set of regression m odels support the hypothesis that bilinguals earn more than their monolingual counterparts in th e enclave. The importance of bilingualism in Miami-Dade County also arose as a consistent th eme in the qualitative interviews. Since findings from the LC bilingualism model were not statis tically significant, I cannot draw a comparison between the two models regarding the effect s of bilingualism on earnings in each area. Nevertheless, the significant findings from the HC model on thei r own are unique and important for two main reasons. First, most prior research has not shown that knowl edge of an unofficial, minority language to be positively associated with earnings (Pendakur and Pendakur 2002).

PAGE 66

66 Second, they show that dual forms of linguistic capital are needed to maximize labor market outcomes in the minority language enclave. Immigrant/Minority Incorporation This study finds that English is positively associated with earnings of Hispanics both inside and outside the Spanish language enclave in F lorida. It makes clear that throughout the state of Florida there are economic incentives to learning English. This reinforces theories of majority language proficiency as a valuable human, social, and cultural capital trai t in the labor market; for non-native speakers, an investment in learning English will clearly reap monetary rewards. Additionally, as an aside, these findings may also serve to quell any nativist fears that English is losing its importance in areas with a large minority language popul ation, such as Miami (Portes and Rumbault 1996). The second set of findings does not fit as easil y into traditional incorporation theories. They show that although English ma tters in the enclave, Spanish doe s, too. In other words, there are also economic incentives to maintaining or acquiring Spanish language skills. Since these results contradict most previous studies of the economic value of b ilingualism (Fry and Lowell 2003), they may indicate a shift in the value of bilingualism in the US labor market. In addition, they highlight one way in which minority la nguage speakers have at once adapted to and transformed the linguistic landscape in Miami. These findings directly ec ho Portes and Stepicks (1993) discovery of the impact that immi grants had on Miami. They note: As sociologists, our principal focus was the adaptation of foreign-born minorities to their new environment. As time passed, however, it became clear that the environment itself was changing in ways that we could not have anticipated. The immigrants were transforming not only themselves, but also th e city around them (Por tes and Stepick 1993: xi).

PAGE 67

67 Policy Implications Two prim ary policy implications stem from th is research. First, findings offer helpful information about where English language trai ning programs would be most valuable for nonnative speakers of English. Adding to the urgency of this issue, a recent New York Times article underscored the growing need for government -funded English language training programs, particularly in areas with large immigrant populations. As th e foreign-born population increases in many states, the waitlists for admission into free or low-cost government funded English classes range from several months to two years (S antos 2007). In some cases, frustrated business owners have taken matters into their own hands, teaching Englis h to immigrant employees. As Tara Colton, author of the Center for an Urba n Future report, notes, The issue of English proficiency has become an issue of econom ic development (Santos 2007: 4). Secondly, there are important implications fo r bilingual education ini tiatives in Miami. Findings from this study show that Spanish langu age skills are important resources in the labor market in Miami. However, educational res earch links recent English-only initiatives and federal legislation such as the No Child Left Be hind to a decline in bili ngual language skills and a de-emphasis on bilingual education in Florida (SSTESOL 2005). A recent Miami Herald article also laments the perceive d decline of bilingualism in Miami. The article views the decline of Spanish language skills among Hispanics in Miam i as the loss of an asset. A telling quote from the article further highlights the importance of bilingualism in Miami. University of Miami linguist Andrew Lynch observes, Miami grew as a city along with the Spanish and bilingualism. Bilingualism was the foundation of Miami as a global city (Fernandez 2008). This study provides additional evidence of the need to place greater emphasis on bilingual and Spanish language education in Miami.

PAGE 68

68 Suggestions for Future Research The currency of the language and im migration de bates in the US and the inconsistencies in the research make this topic ripe for further expl oration. This particular study could be further expanded to explore a variety of angles. For example, th e inclusion of non-Hispanic monolingual English speakers into the sample wo uld allow for deeper comparative analysis of the economic value of bilingualism. Bilingual Hi spanics may earn more than their monolingual English Hispanic counterparts, but does the same pattern hold when comparing bilinguals to nonHispanics monolinguals? Larger scale modifications of this study could also prove interesting. For example, the application of these methods to a nationwide analysis would he lp draw conclusions about the between language, earnings, and enclaves in the general US labor market. While language and the labor market may interact in one way in Flor ida, their relationship may be different in other areas in the US. This is particularly important since Miami is often viewed as a unique case in urban sociology; it should not be considered a microcosm of the American city (Portes and Stepick 1993). Thus, a nationwid e analysis would be allow to s ee if such findings are unique to Florida or if the pattern repeat s itself in other states in th e US and on a national scale. In addition, longitudinal studies on the topic will offer furthe r insight into the ways in which the linguistic composition of the population affects the relationship between language and earnings. Since census language questions have not been uniform throughout the last century, there are certain complications involved conduc ting such a study (Stevens 1999). However, since 1980, the language questions have remained th e same and would thus be suitable for an analysis of how the effect of English on earnings has changed in accordance with changes in the immigrant population over the last several decades.

PAGE 69

69 Longitudinal studies should also be used monitor the future effect of bilingualism on earnings. The aforementioned Miami Herald article already cites anecdot al evidence of a decline in bilingualism in Miami (Fernandez 2008). Language attrition has indeed revealed itself to be a common pattern among children of immigrants in the US. As Portes and Rumbault (1996:11) note, The United States is unique in the rate at which other languages have been abandoned in favor of Englishin no other c ountry have foreign languages been extinguished with such speed. Thus, it is possible that Spanish language will lose prominence as second and thirdgeneration speakers become more acculturated and ab andon Spanish in favor of English. If this proves to be the case, the shift in linguistic preference will likely alter the effect of bilingualism on earnings in the future.

PAGE 70

70 CHAPTER 6 CONCLUSION English language proficiency functions as a va luable form of human, cultural, and social capital in the US labor market. Past research has shown English language proficiency to be a uniformly positive determinant of earnings in the US. While individual human, cultural, and social capital characteristics largely influence ear nings, context also plays an integral role in labor market outcomes. The primary theoretical proposition of this st udy assumed that labor market context specifically th e linguistic profile of the labor marketdetermines the value of language proficiency in the labor market. Despit e the linguistic diversity in the US and the rich data available from the decennial Census and other surveys, few recent studies address this topic. Using data from the 2000 US Census, this st udy provided empirical evidence of the ways in the effect of language proficiency on earnings varies according to the linguistic profile of the labor market. The central research hypothesis po sited that, among Latinos in Florida, English language proficiency would be le ss important to earnings in areas with a large percentage of Spanish-speakers than in areas with a small pe rcentage of Spanish-speakers. The corollary hypothesis argued that Spanish language skills woul d have a positive effect on earnings in areas with a large Spanish language en clave, such as Miami-Dade C ounty. However, results from the regression models suggest that E nglish language proficiency has a greater impact on earnings of Latinos in areas with a large pr oportion of Spanish-speakers. The second set of results show that while English language proficiency still plays a si gnificant role in areas with a large Spanish language enclave, Spanish language proficiency al so has a positive effect on earnings in these areas. Although lack of signifi cance in the low concentration m odel did not permit a comparison of the effects of bilingualism on earnings betw een the two linguistic contexts, the high

PAGE 71

71 concentration model shows that, a ll other factors being equal, bili ngual Hispanics earn more than monolingual English counterparts in these areas. These findings offer valuable insight into pr ocesses of labor mark et incorporation of immigrants and non-native speakers. Like pa st studies, my study reinforces the economic incentives for learning English. However, it differs from past studies in that it also offers evidence of the economic value of learning or ma intaining Spanish language skills in certain contexts. Consequently, these findings have im portant policy implications for English language training and bilingual ed ucation programs. English remains the most widely spoken language in the US and as such, clearly reaps the most monetary rewards in the labor market. Ho wever, as the Spanishspeaking population in the US continues to grow, it is likely that, as this study has shown, Span ish will also assert itself as a prominent force in the labor market. While the va lue of language proficiency clearly varies by labor market context, there is no consensus within the literature regarding the ways in which it varies. Thus, as the linguistic composition of the US continues to evolve, the study of the relationship between language, earnings, and labor market context remains fertile ground for additional research.

PAGE 72

72 LIST OF REFERENCES Becker, Gary. 1993. Human Capital:A Theoretical and Empirical Analysis, with Special Reference to Education Chicago: University of Chicago Press. Bloom, David E. and Gilles Gr enier. 1992. Earnings of the French Minority in Canada and the Spanish Minority in the United States. Pp. 373-409 in Immigration, Language, and Ethnicity: Canada and the United States, edited by Barry R. Chiswick. Washington D.C.: AEI Press. Booth, William. 1998. One Nation, Indivisible: Is it History? Washington Post. February 22. Retrieved March 1, 2008 (http://www.washingtonpost.com/wpsrv/ national/longterm/meltingpot/melt0222.tm). Borjas, George J. 1982. The Earnings of Hispanic Immigrants. Industrial and Labor Relations Review 35:343-353. Bourdieu, Pierre. 1986. The Forms of Capital. Pp. 241-58 in Handbook of Theory of Research for the Sociology of Education, edited by J.E. Richardson New York: Greenwood Press. ------. 1991. Language and Symbolic Power. Cambridge: Polity Press. Carliner, Geoffrey. 1981. Wage Differences by Language Group and the Market for Language Skills in Canada. The Journal of Human Resources 16:384-399. Carnevale, Anthony P., Richard Fry, and B. Lindsay Lowell. 2001. Understanding, Speaking, Reading, Writing, and Earnings in the Immigrant Labor Market. The American Economic Review 91:159-163. Chiswick, Barry R. 1978. The Effect of Amer icanization on the Earnings of Foreign-Born Men. Journal of Political Economy 86:897-921. ------. 1991. Speaking, Reading, and Earni ngs among Low-Skilled Immigrants. Journal of Labor Economics 9:149-170. Chiswick, Barry R. and Paul W. Miller. 1992. L anguage in the Immigrant Labor Market. Pp. 229-296 in Immigration, Language, and Ethnicity: Canada and the United States, edited by Barry R. Chiswick. Washington D.C.: AEI Press. ------. 1995. The Endogeneity between Language and Earnings: International Analyses. Journal of Labor Economics 13:246-288. ------. 2002. Immigrant Earnings: Language Skills Linguistic Concentrations, and the Business Cycle. Journal of Population Economics 15:31-57. Coleman, James. 1988. Social capital in the creation of human capital. American Journal of Sociology 94:95-120.

PAGE 73

73 Dvila, Alberto and Marie T. Mora. 2000. E nglish Skills, Earnings, and the Occupational Sorting of Mexican Americans along the U.S.-Mexico Border. International Migration Review 34:133-157. DeJong, Gordon F. and Quynh-Giang Tran. 2001. W arm Welcome, Cool Welcome: Mapping Receptivity Toward Immigrants in the U.S. Washington, DC: Population Reference Bureau. Retrieved March 1, 2008 ( http://www.prb.org/Articl es/2001/WarmW elcomeCoolWe lcomeMappingReceptivityTo w ardImmigrantsintheUS.aspx). Duany, Jorge. 1992. The Census Undercount, the Underground Economy, and Undocumented Migration: The Case of Dominicans in Santurce, Puerto Rico. Ethnographic Evaluation of the 1990 Decennial Census Report #17. Prepared under Joint Statistical Agreement 90-09 with the University of the Sacred Heart. Washington, D.C.: Bureau of the Census. Retrieved February 22 2008 ( http://www.census.gov/srd/papers/pdf/ev92-17.pdf ). Duany, Jorge and Flix Matos-Rodrguez. 2006. Pue rto Ricans in Orlando and Central Florida. Policy Report 1, No. 1. New York: Centro de E studios Puertorriqueos, Hunter College, CUNY. Retrieved March 1, 2008 ( http://www.orlando.org/clientuploads/ hsumm it/hsummit_prcentralflorida.pdf ). Dustmann, Christian and Arthur van Soest. 2001. Language Fluency and Earnings: Estimation with Misclassified Language Areas. The Review of Economics and Statistics 83:63-674. ------. 2002. Language and the Earnings of Immigrants. Industrial and Labor Relations Review 55:473-492. Fernandez, Enrique. 2008. The Spanish S poken today by bilingual Hispanics in South Florida has become watered down through generations. Miami Herald. March 1. Retrieved March 1 2008 ( http://www.miamiherald.com/news/br eaking_dade/v-pri nt/story/440517.htm l ). Fry, Richard and B. Lindsay Lowell. 2003. The Value of Bilingualism in the U.S. Labor Market. Industrial and Labor Relations Review 57:128-140. Gonzalez, Arturo. 2000. The Acquisition and Labor Market value of Four English Skills: New Evidence from NALS. Contemporary Economic Policy 18:259-269. Grieco, Elizabeth M. and Rachel C. Cassidy. 20 01. Overview of Race and Hispanic Origin. Washington, DC: US Census Bu reau, Retrieved February 22 2008 ( http://www.census.gov/ prod/2001pubs/cenbr01-1.pdf ). Grenier, Gilles. 1984. The Effects of Language Characteristics on the W ages of HispanicAmerican Males. Journal of Human Resources 19:35-52.

PAGE 74

74 Grin, Franois. 1994. The Economics of Language: Match or Mismatch? International Political Science Review 15:25-42. Guzmn, Betsy. 2001. The Hispanic Populatio n, Census 2000 Brief. Washington D.C.: U.S. Census Bureau. Retrieved March 1 2007 (http://www.census.gov/ prod/2001pubs/c2kbr01-3.pdf). Hand, Michael. 2006. Are enclaves amenities? An Empirical Investigation in the Southwest United States. Economics Bulletin 10:1-7. Hardy, Melissa. 1993. Regression with Dummy Variables. Sage Publications: Newbury Park. Light, Ivan, Georges Sabagh, Mehdi Bozorgme hr, and Claudia Der-Martirosian. 1994. Beyond the Ethnic Enclave Economy. Social Problems 41:65-80. Lovell, Peggy. 1989. Racial Inequality and the Brazilian Labor Market. Ph.D. dissertation, Department of Sociology, University of Florida, Gainesville, FL. McGroarty, Mary E. 1990. Bilingualism in the Workplace. Annals of the American Academy of Political and Social Science 511:159-179. McManus, Walter S. 1990. Labor Market Effects of Language Enclaves: Hispanic Men in the United States. The Journal of Human Resources 25:228-252. Mincer, Jacob. 1974. Schooling, Experience, and Earnings. New York: National Bureau of Economic Research. Mora, Marie T. 2003. An Overview of the Economics of Language in the U.S. Labor Market. Presentation Prepared for the American Economic Association Summer Minority Program. Denver: University of Colorado. Retrieved March 1 2007 ( http://www.econ.duke.edu/aeasp/seminar_files/sem inar2003_files/MORA2.pdf ). Nee, Victor and Jimy Sanders. 2001. Unde rstanding the Diversity of Immigrant Incorporation: A Fo rms-of-Capital Model. Ethnic and Racial Studies 24:386-411. Park, Jim Heum. 1999. The Earnings of Immigran ts in the United States: The Effect of English-Speaking Ability. American Journal of Ecnomics and Sociology 58: 43-56. Pew Hispanic Center. 2006. Hispanics at MidDecade." Washington DC: Pew Hispanic Center. Retrieved February 25 2008 ( http://pewhispanic.org/files/other/middecade/Table-12.pdf ). Portes, Alejandro. 1987. The Social Origins of the Cuban Enclave Econom y of Miami. Sociological Perspectives 30:340-372.

PAGE 75

75 ------. 1995. Economic Sociology and the So ciology of Immigration: A Conceptual Overview. Pp. 1-41 in The Economic Sociology of Immigration, edited by Alejandro Portes. New York: Russell Sage Foundation. ------. 1998. Social Capital: Its Origins a nd Applications in Modern Sociology. Annual Review of Sociology 24:1-24. ------. 2002. The Two Meanings of Social Capital. Sociological Forum 15:1-12. Portes, Alejandro and Richard Schauffler. 1996. Language and the Second Generation: Bilingualism Yesterday and Today. Pp. 8-29 in The New Second Generation, edited by Alejandro Portes. New York: Russel Sage Foundation. Portes, Alejandro and Alex Stepick. 1993. City on the Edge: The Transformation of Miami. Berkley: University of California Press. Ruggles, Steven, Matthew Sobek, Trent Alexan der, Catherine A. Fitch, RonaldGoeken, Patricia Kelly Hall, Miriam King, and Chad Ronnander. 2004. Integrated Public Use Microdata Series: Version 3.0 [Machine-readable database]. Minneapolis, MN: Minnesota Population Center [producer and distributor], http://usa.ipums.org/usa/ Sanders, Jimy M. and Victor Nee. 1987. Lim its of Ethnic Solidaryity in the Enclave Economy. American Sociological Review 52:745-773. Santiestevan, Stina. 1991. Use of the Span ish Language in the United Sates: Trends, Challenges, and Opportunities. ERIC Digest. Retrieved March 1, 2008 ( http://www.ericdigests.org/pre-9221/spanish.htm ). Santos, Fernanda. 2007. De mand for E nglish Lessons Outstrips Supply. New York Times. January 27. Retrieved February 22, 2007 ( http://www.nytimes.com/2007/02/27/educ ation/27esl.htm l?_r=1&oref=slogin ). Shin, Hyon B. and Rosalind Bruno. 2003. Language Use and English-Speaking Ability. Washington, DC: US Census Bu reau. Retrieved February 22 2008 ( http://www.census.gov/ prod/2003pubs/c2kbr-29.pdf ). Siegel, Paul, Elizabeth Martin and Rosali nd Bruno. 2001. Language Use and Linguistic Isolation: Historical Data and Methodological Issues. Pp. 167-190 in Statistical Policy Working Paper 32: 2000 Seminar on Integra ting Federal Statistical Information and Processes. Washington DC: Federa l Committee on Statistica l Methodology, Office of Management and Budget. Retrieved February 22 2008 ( http://www.census.gov/srd/www/abstract/ssm2007-02.html). Stevens, Gillian. 1999. A Century of U.S. Cens uses and the Language Characteristics of Imm igrants. Demography 36:387-397.

PAGE 76

76 Stolzenberg, Ross M. 1990. Ethnicity, Geography, and Occupational Achievement of Hispanic Men in the United States. American Sociological Review 55:143-154. Sunshine State TESOL of Florida. 2005. Pos ition Statement on Bilingual Education. Florida: SSTESOL. Retrieved March 1 2008 (http://www.clas.ufl.edu/users/rthompso/advocacybilingualed.html). Tienda, Marta. 1983. Market Characteristics and Hispanic Earnings: A Comparison of Natives and Immigrants. Social Problems 31:59-72. US Census Bureau. 1999. Findings on Questions on Race and Hispanic Origin Tested in the 1996 National Content Survey. Washington, DC: U.S. Census Bureau. Retrieved February 22 2008 ( http://www.census.gov/population/www/docu mentation/twps0016/report.html#intro ). ------. 2006. Facts for Features: Hispanic Her itage Month 2005. Washington, DC: U.S. Census Bureau. Retrieved March 1, 2008. ( http://www.census.gov/PressRelease/www/releases/a rchives/cb05ff-14-3.pdf ). ------. 2007. Am erica Speaks: A Demographic Prof ile of Foreign-Language Speakers for the United States: 2000. Washington, DC: U.S. Census Bureau. Retrieved March 1, 2008 ( http://www.census2010.gov/population/ www/socdem o/hh-fam/AmSpks.html ). ------. 2008. "State and County QuickFacts: Miam i-Dade County and Florida." Washington, DC: U.S. Census Bureau. Retrieved March 1, 2008 ( http://quickfacts.census. gov/qfd/states/12/12086.htm l ). Viglucci, Andres. 2001. Spanish is Spreading Northward in Florida. Miami Herald. November 22. Retrieved March 1 2008 (http://www.puertoricoherald.org/issues/2001/vol 5n49/SpanishNorth-en.html). Wilson, Kenneth L. and Alejandro Portes. 1980. Immigrant Enclaves: An Analysis of the Labor Market Experiences of Cubans in Miami. The American Journal of Sociology 86:295-319. Zhou, Min and John R. Logan. 1989. Returns of Hu man Capital in Ethnic Enclaves: New York Citys Chinatown. American Sociological Review 54:809-820.

PAGE 77

77 BIOGRAPHICAL SKETCH Molly Dondero was born in Philadelphia, Pennsylvania in 1981. She graduated from the Pennsylvania State University in 2004 with a B.A. in Spanish and English. Before attending the University of Florida, Molly worked as a Fulbri ght teaching assistant in the English Department at the Universidad Nacional de Villa Mara, in Crdoba, Argen tina. Upon graduating in May 2008 with an M.A. in Latin American Studies, Mo lly plans to pursue a Ph.D. in Sociology at the University of Texas at Austin.


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101106_AAAAGC INGEST_TIME 2010-11-07T01:52:03Z PACKAGE UFE0022196_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 964634 DFID F20101106_AABVPH ORIGIN DEPOSITOR PATH dondero_m_Page_33.jp2 GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
38bd979c788d7da8a86e578f3785df0d
SHA-1
507f37701a3541f605756495a23c450cc688634c
25271604 F20101106_AABVZC dondero_m_Page_10.tif
545688daad13ffc5880681b31c39ba64
1040c82ad892fa01372d15e8a4f72f65dca35557
42135 F20101106_AABWBC dondero_m_Page_33.pro
78e5cba3e6f74e9b78538fb373cafae1
a3454ba2954da2143cfaa59013e6080bfd363ff6
5967 F20101106_AABVUF dondero_m_Page_49thm.jpg
f5e3b009c99e4175594de75059117d4c
c85555b217825ad37b6c94f4e84049f2ae1e28dc
4807 F20101106_AABWGA dondero_m_Page_60thm.jpg
6030f6e3ebfde8d0a76153e17022e56e
05e75c6ce09027493ed51dfe76260e16fd83664b
27753 F20101106_AABVPI dondero_m_Page_72.QC.jpg
f41c0db9e5d863c65c287e178589499e
ca3af55463dd8101ddee0014cf954b5152c92cf6
F20101106_AABVZD dondero_m_Page_11.tif
202149935317f9b79ee80b515728aa07
4213d2e999db5fcf4a6a8ed884087fb5ed2ffb2f
51301 F20101106_AABWBD dondero_m_Page_37.pro
b3720cb930f15cf9bf63766736e9f750
ee4d6ca428bf5aea58a5f7ee22c98cb18b986b36
F20101106_AABVUG dondero_m_Page_68.tif
953fd816e80509093a522353be162575
1dc92a681f79713eacb1548b29eacdf9fbf9431d
16342 F20101106_AABWGB dondero_m_Page_61.QC.jpg
ce6b1627a6c2154ea53e1a8b0d869461
53f733e430a0221b26e1fa3b373bcc76b5c8171f
16593 F20101106_AABVPJ dondero_m_Page_47.jpg
dac84f6f9253bc1cc84f47dd6139e067
2fab03104896cca9cc9732702484ac65e707c55c
F20101106_AABVZE dondero_m_Page_12.tif
8cf74e13e71459e76644c9244a10157e
6f1d962bea86dd7983f486c363050286f13e0158
53798 F20101106_AABWBE dondero_m_Page_39.pro
81e01202c502f11c40b950551d414f3f
19faaf7715c486fcbb278e5981789fe090a962b4
F20101106_AABVUH dondero_m_Page_48.tif
ae4a557d1ef023a76dc0ac2b46514188
916f28a808e5c4874dad7d57345f8c5061261699
4568 F20101106_AABWGC dondero_m_Page_61thm.jpg
e604353a4aa8a91a2ab9081790c1bd72
caba58fff83e9a3e06c6cd6c31dfd19ab9d1b0c7
27175 F20101106_AABVPK dondero_m_Page_21.QC.jpg
d2b44402e8f6d1802eca15ea047e3257
6a098189d8f9a7a50df1c3a26758ef4100b4043b
F20101106_AABVZF dondero_m_Page_16.tif
f1d0c236febcc29d7280891a4da36c00
f11595938cea488fe6d37e91c7549f020c1439b6
53650 F20101106_AABWBF dondero_m_Page_40.pro
88d8923427e09dbde9d87d2c25dd968a
707234f99e1edcc2adfe2709e5191c7358afb265
53097 F20101106_AABVUI dondero_m_Page_36.pro
4467fe808596cc0b691e690971284e0c
5f9420807689170807edf69cd33f21f169b9dd3f
6249 F20101106_AABWGD dondero_m_Page_62thm.jpg
ffa9ad50ef851f9e708dc7fbe256b344
eda58c555e4d1a9c30288bd8f6222563d7d291c9
6633 F20101106_AABVPL dondero_m_Page_41thm.jpg
89082b5e892195f6a7a13169a171b589
7cdbce6634de2be26027887a19cf47ee4fdc488c
F20101106_AABVZG dondero_m_Page_17.tif
b2d9a29b61c7fcfa6a698384d7cc0bb8
e732a1aec14ee426fe54bbaf8192488f6f6441f1
53648 F20101106_AABWBG dondero_m_Page_41.pro
3e4b5f25ed8b83482431fc479c006e01
84ced52622f157a27d905ff5da3d9b8ac3556c49
2127 F20101106_AABVUJ dondero_m_Page_34.txt
6dbea4c0a33648953d684dc567ca4236
6337dab0060d5a8eccf4404cf0c36f4ce5f21942
6181 F20101106_AABWGE dondero_m_Page_66thm.jpg
cfd7c42d0e1ca9128784c7e6e7c3d061
fae5d05b2d6325263e54f961cd0d71825e23cb63
54628 F20101106_AABWBH dondero_m_Page_42.pro
319bbc53ad7906c0d232923307b06c29
5f6ae9c5ac434a04196275183ca5cc943a24e9fc
2109 F20101106_AABVPM dondero_m_Page_40.txt
610d81d759c99562aaa1bb8c16c436b3
f5f39dbaad447486336a6aec76f199c44a4a47be
F20101106_AABVZH dondero_m_Page_19.tif
0b019b11ada34c2ccec09ea43b923f74
52a8b83914f3b18ecedd023786a6d9aeb26acc51
26819 F20101106_AABWGF dondero_m_Page_67.QC.jpg
adcb46b6e232a6da7b1c9f8c23333490
c2f97a1334788101de42cd1fac3b43efb0670e90
50890 F20101106_AABWBI dondero_m_Page_45.pro
7c160e56c1ec5d494a30a694e6a6e1af
e70a7388784340d0afbe0461df89bbed5d1699ce
28672 F20101106_AABVPN dondero_m_Page_36.QC.jpg
b36bfd818bc53f0351d5df258c85a163
3bdb97511f748193d9baf56a72ff3ba622693f94
F20101106_AABVZI dondero_m_Page_22.tif
42b2fcc384fb781699db0b030656adbc
82aec5fa7a0fa82bd7c34d50d2e32afbe0e0bde2
95519 F20101106_AABVUK dondero_m_Page_24.jpg
f88ca1a362c6acb5118204c42e376d12
edcbfea16d94f091a760b27732a862d51123ccfd
26940 F20101106_AABWGG dondero_m_Page_68.QC.jpg
673525e8555a4edf56b4766266d74c72
bcbdd03fb891fe8c3e64cd37e6192f4c6ac8efcb
47349 F20101106_AABWBJ dondero_m_Page_49.pro
159061902b410c7dccaf36801e529521
481bb8265a772db23966073d444d7ad120fad8c5
1051975 F20101106_AABVPO dondero_m_Page_38.jp2
ebf442b920f98568f33e33751334e9f8
d5b63f3f336947fe48d7488efbe238081baadf76
F20101106_AABVZJ dondero_m_Page_23.tif
c82ba0bd23b7234e64c0a84437440cd0
dc735aadfca47c506c54aa308cb04feacabc73d2
F20101106_AABVUL dondero_m_Page_58.tif
5af6483db8efa97418ac591a1dae2ada
6bc9da1ae5f70074446b2debed9184bca68963c7
12755 F20101106_AABWGH dondero_m_Page_69.QC.jpg
ed3df32fc6548b221c63aef40e312eda
09621921da3dfb4b6e01606a1b4ba7f3f719c300
55980 F20101106_AABWBK dondero_m_Page_50.pro
1a83e23b32fbf63f8e30c3818b0ac7cc
fd0dafdb891188e074cfb24c011c95f32237b8df
7777 F20101106_AABVPP dondero_m_Page_01.pro
c488a6cc4c013f134e97af6ad45d7898
fbbe4a595c52b9b3642d844022609aada8e22c00
F20101106_AABVZK dondero_m_Page_25.tif
e5aaa93333e1f330110679803b8ed7d5
bb196c2bbe2915b1f007e7dd8ccc6df85b533166
2176 F20101106_AABVUM dondero_m_Page_61.txt
57a14984ae52a64e2891a0a92b23a5ac
37034b8b14c71df7256bad2d6e7cd7656e015346
3173 F20101106_AABWGI dondero_m_Page_69thm.jpg
f6dcbed55e38579a5796d63df0e0b558
0ee3ea4b1dccb686ffb2de269a52542bad9ccb94
53330 F20101106_AABWBL dondero_m_Page_54.pro
c299689868d000fd9950178dc3a92207
4e97dc68b24f46ffa5998e942de7a9fc161aa1e8
27482 F20101106_AABVPQ dondero_m_Page_40.QC.jpg
7a713eaeb91d43cd1c76c6b1001989b7
a4701ce61523dabb332b8b4603f81642da7ee08f
F20101106_AABVZL dondero_m_Page_26.tif
c3ebbffa2842b63cc8fc1fea57c9d3c9
3e7de8f1049355ac6bdfd2a93a7b960e070eb607
3786 F20101106_AABVUN dondero_m_Page_46thm.jpg
8df38a8b9ba66d161e20a67a0f2b7eb3
5aec2c0395efb050761e811f1c9c800cbcc9c836
27054 F20101106_AABWGJ dondero_m_Page_70.QC.jpg
69324b3090812ca49a6ddd9ab09f0eb5
b326de07306108be68c74589b341a9e95bdccf8a
54169 F20101106_AABWBM dondero_m_Page_55.pro
4162525d620b5db9b4edea5a1e9fb635
262c0f1ad85cf10b816dfb9d3760d5b02dcd6a3c
63748 F20101106_AABVPR dondero_m_Page_73.pro
f5b7ee0faf6a46a9b3178ca0f33ed7ee
c404d54c3c6ed9916d61737c69ab1edf7564efaa
F20101106_AABVZM dondero_m_Page_28.tif
116417f9b9e3e27703a86bf9e9e5ec5c
a811d869d74796366a2ae05bca8bb802be3a25bf
1051986 F20101106_AABVUO dondero_m_Page_55.jp2
bb07fba496720249a92b730e79b5137f
23bbd0b1617da7f6cb253fdd13d6a420a60de61e
6689 F20101106_AABWGK dondero_m_Page_72thm.jpg
ba99ff3a2e7275e97668d23a54a59ea7
bb45770723616d62ee961342881d983508517a06
37105 F20101106_AABWBN dondero_m_Page_58.pro
b81b7bcf3cad5d442ac8237f612f8c73
8b81b6c638edfcc2f40f9bc8fa1d2fd629b5116c
1051979 F20101106_AABVPS dondero_m_Page_74.jp2
6cf13c7131eba3cecdfc4047285c5740
5f1287fe5c4c4792951dbad7c4d92cfa40e7cdbe
F20101106_AABVZN dondero_m_Page_29.tif
febf708b6dfccc8bcbcca45d29348860
03989084e13b44547710a6604bfe7e18ae7fc3c9
28928 F20101106_AABVUP dondero_m_Page_50.QC.jpg
33daa5bc2c4fcbd47b259015a5fd5c22
4c612295374ac3c8457feaacf8c8ce0c3fd24532
30413 F20101106_AABWGL dondero_m_Page_73.QC.jpg
a27eb39b2733b413206c7df8328a9d68
b6f6ca2ab9a2292557189f20121089642027c79e
42619 F20101106_AABWBO dondero_m_Page_61.pro
5c7dcd3795ffbb793a8df25a7f6e410f
fd3b9cac1d46efc20b0f4484af08bc7409a15472
111693 F20101106_AABVPT dondero_m_Page_73.jpg
779053d75b200e62a4f8225ec2cc5cd6
ecab44e135754bbf979b192279c214bc8573e495
F20101106_AABVZO dondero_m_Page_32.tif
31bf2044ffe8b06250c2c8fc052d30ae
1a4b5c718eda63a51c204d7b378d7ace3f6fdfae
99078 F20101106_AABVUQ dondero_m_Page_74.jpg
056641d88dd3f45f6f35c4246f308a0f
6db45f25899f21a0ad2e9605f026fb2c09e1d235
7387 F20101106_AABWGM dondero_m_Page_73thm.jpg
30b977b748f8ba4cba286f53e9be33b7
17f91ea5aecca599b1a099f8dde2fc398b924e68
49534 F20101106_AABWBP dondero_m_Page_62.pro
40373186dbe0e0d0029a5cf6339f2a9b
b192bb1eec9fe5bb2529bf441576e66e1c335312
27817 F20101106_AABVPU dondero_m_Page_41.QC.jpg
c4ae4d1cef3d58f7748e1a27c4f43f6e
1424d330b32338ac0b9948ee51d5a7ccbfdae04a
F20101106_AABVUR dondero_m_Page_54.tif
eac4e187cc42ad91eac7f437ef9f89a1
f37a384a5ee0b1a71c68dd22e7b3fe36b7abfb77
28843 F20101106_AABWGN dondero_m_Page_75.QC.jpg
ffb069d2fc81e70a5420b6554efe9688
ed66adc875987b3e8382da9e7cee130b307b720d
51900 F20101106_AABWBQ dondero_m_Page_66.pro
5ef90f7df84174730ed7a7271ccbc5b4
2aa7b941650140fb1de1d23757927ffe79ced8be
2621 F20101106_AABVPV dondero_m_Page_75.txt
0cacf0cfc1a5be45914ff4b4f6e11a00
68c2769023028d3cac285a26f66eda84c79b5c71
F20101106_AABVZP dondero_m_Page_33.tif
498dcc4305570defba5f4653c221b217
63199c0f2e0f9dc7e0b7452e5f6ed59a0691ee04
29006 F20101106_AABVUS dondero_m_Page_64.QC.jpg
914f1f02b85271b35fdb435cafa11788
f5ce56a27d44eaec75b39bcb1fee3a79c0081964
25132 F20101106_AABWGO dondero_m_Page_76.QC.jpg
a251b0259b1950a3ee9eaa05edff76af
3da9a273534b1891e817a9de8c9ff369af2d492f
52644 F20101106_AABWBR dondero_m_Page_67.pro
433a5d71b86b299cfc96ee58543972cc
a157c4cc1a9d0f8f80bb8bc7fdd38e1af1f29101
23017 F20101106_AABVPW dondero_m_Page_69.pro
8c450500fc00752acdcf185344b621c8
93464d41744dd10161e80e812492d35a26d953d5
F20101106_AABVZQ dondero_m_Page_39.tif
f69a902bc99cd9275213a0a56f638985
6544483be0a6384ab081fde8a0612e946b67cc34
1051949 F20101106_AABVUT dondero_m_Page_20.jp2
412e35f0d0ec39feac9f51d595bde114
213c3039e29a958a33f12aede47b1a7242566889
6031 F20101106_AABWGP dondero_m_Page_76thm.jpg
de22ca8f060ab579be6147dc24197eaa
9e24863456be816552f65d2e7cb3a332f040eaaf
53179 F20101106_AABWBS dondero_m_Page_70.pro
3b675949716f6f3b59f173ba4c8b84bf
0a3a0345acc8ccd3d0884587d86ab5a5df9fa75f
27765 F20101106_AABVPX dondero_m_Page_74.QC.jpg
74ca2ecf841b6c0917af597d9a8d4edf
a79e12f98b5dca34381ab3f4674bffec75cfc318
F20101106_AABVZR dondero_m_Page_40.tif
4282e105a795e70ab2f164a590fb237c
04fd75c9f0f99beba72985de6fd7c5d81b9fa5d5
1051958 F20101106_AABVUU dondero_m_Page_56.jp2
8d7fe13c2b5775eb7e367fbf8c347bc9
60cc78fab123cef046c435b0c5e85600148cd9a0
8643 F20101106_AABWGQ dondero_m_Page_77.QC.jpg
06f1349de4edd1f1d55129c5c0d79e79
0715e396f4193fd04efe6ec1cdf86ef8dd4bd23c
63796 F20101106_AABWBT dondero_m_Page_75.pro
f2ba0ffcc8d70ce05f68d4f21468977b
a87729522fc51bf63a27552cc2da386819d60c9f
289 F20101106_AABVPY dondero_m_Page_08.txt
70d67cba1332ba1277fa3768f838819a
33241dd42501fb17ecf9092436772a0ba7884061
F20101106_AABVZS dondero_m_Page_41.tif
4a328a525ee404a8cff623146b042bba
ac7535930bd6f51378e90cb3319d81c6f77c9b81
7193 F20101106_AABVUV dondero_m_Page_75thm.jpg
ef1d11c5ac396fb41be2b60c9ab781d2
10e0173801717ce1676b5694cae6c44cdace0c32
325784 F20101106_AABVNA dondero_m_Page_77.jp2
e4793573b62928b43526e9b02814d4a5
f937610b8e98e9a18536ae49abc3687b1a394640
2082 F20101106_AABWGR dondero_m_Page_77thm.jpg
4e8fb0bbb524ccb36dcffc9f2fc456f5
295b75a5f38f414b5027158f8d18aa71baae1d2b
50509 F20101106_AABWBU dondero_m_Page_76.pro
c1fb3d7bba3ff060f2d6d80b13e56d20
312ef6a64ca002ab7a514c53455ac29e68b1df6a
1051962 F20101106_AABVPZ dondero_m_Page_10.jp2
0cafa39f16acbc515d9eb94304ace7cb
0fe7fecf0b784d2270573da0c3ce7cd263ad8052
F20101106_AABVZT dondero_m_Page_42.tif
11ac71772d981eaa4b9bc678286f61bd
84234a4533be7e9ea72a2ff05572ae7e572b0955
6802 F20101106_AABVUW dondero_m_Page_35thm.jpg
4b0f6bf205059f572c77e4098cfede3a
93c2176639086732ed889aa47fe53349f5b8e216
6780 F20101106_AABVNB dondero_m_Page_26thm.jpg
b79c2a581b15085f259882f69aa21ef4
0a04711d9453b3315dd680947f0a3c4648c7eb6f
90910 F20101106_AABWGS UFE0022196_00001.mets FULL
5314655d616a2b031e1ace3b9a16a3b0
2b514a12a41891baab96acbabceb89a485a819d7
13663 F20101106_AABWBV dondero_m_Page_77.pro
43eb8fca776de681a4529da831393765
6a8b9c349dc72c5dfa57532fa411184cf48765ff
F20101106_AABVZU dondero_m_Page_46.tif
9a8dd2d4383f5787124bf6a19f8243b8
f64c83a486511a24ca24fd6408099551f71e4d02
449 F20101106_AABVUX dondero_m_Page_48.txt
1dafe0dd1558a74c333b4c23b6179807
d7f62c3f57310b0b7f906c10f934942570892d8a
5470 F20101106_AABVNC dondero_m_Page_11.QC.jpg
c3d020bd3cd95a246ec73e229087a980
8901e6e0acee53af94efbda8f97324a20098f19f
F20101106_AABVZV dondero_m_Page_47.tif
06e31d255a4ac8dc4d97b22467b4e37c
959545896c629156170ad3002f3bd10b32ed4524
83353 F20101106_AABVSA dondero_m_Page_43.jpg
5e87bdd1a8f3534ed2d9e67b11b64654
810381293d27bb99e819349457e7e5c918060ef1
F20101106_AABVUY dondero_m_Page_43.tif
a7c01ef192010d69b132a060d8d2e555
21f6155f7de9ca3bc9ea757cd9055b5962cb5a5c
89 F20101106_AABWBW dondero_m_Page_02.txt
077cee001223c821e21e8fcc24f41119
6bb02f89790af8cb3f1a7b2816482ac6786020c1
F20101106_AABVZW dondero_m_Page_50.tif
682fcc56a7f6366e19eea0eaff37977f
9aa07bd9ed609be2951a3144b30cd0c09b809f50
583 F20101106_AABVSB dondero_m_Page_77.txt
655c17db3b1e71a713a28fe90ed0f441
e89cebbe7cad4facf98ecf644045c6c147a0359d
3775 F20101106_AABVUZ dondero_m_Page_05.txt
963fe80f0490d888ea54013145931b24
b4f71bd079acbc166a31ae7f9a2b1420991bc6ec
F20101106_AABVND dondero_m_Page_56.tif
ccb99446e1a793260a05a8711946e78e
175b39175fc11f560c304fe4338edd1b4aed4ef4
103 F20101106_AABWBX dondero_m_Page_03.txt
543e7258cbdee4c045aebed32db2f1e4
7ed6009a26c8a735e78261553910dd7b2d1db2e7
F20101106_AABVZX dondero_m_Page_53.tif
4d839fed952d50f7e80e7e965d821ff2
8c4ef5f4d1f187ba5c4fc9d54c9472f23d752ef1
6896 F20101106_AABVSC dondero_m_Page_24thm.jpg
c199e80499a75b979facb1c795fbcb76
cd196c3c286250d5e57b948527e1f99a7556ada9
53176 F20101106_AABVNE dondero_m_Page_27.pro
0b3c284564d42e0872db23431bd3619d
4b9bcd64740e284e6fea392dfcb688af6799f9fe
1834 F20101106_AABWBY dondero_m_Page_06.txt
0a530519feefeedfeea29ae55328fdb4
db4ad7b66c35e7609397452474c08139277021c0
F20101106_AABVZY dondero_m_Page_57.tif
e0bbc29d53c524978f07d7eb7c580393
8900cc02017d6502127cf768eefaebc4abe5a83b
6331 F20101106_AABVSD dondero_m_Page_52thm.jpg
259b0ead94ef3d3357123bacba89d730
081e4a89479e30980b49fdd3672a269d7aacee9e
9510 F20101106_AABVNF dondero_m_Page_32.pro
7aec6d95103b8771a3e2b87b867ffea6
508e1d154d24b47df68ef5d8c88280993be17a7b
90205 F20101106_AABVXA dondero_m_Page_55.jpg
9b18b57b7edfdae501286013f07fa4ff
af91ddfdcdc1fde21abd156140f59f01a766349a
674 F20101106_AABWBZ dondero_m_Page_07.txt
a948ac07ade02ccb261e4246f07aa470
5283451f9744e4df71f60a206c0b959711585a8e
F20101106_AABVZZ dondero_m_Page_59.tif
ba90fe5000f8e760bbc5afb3ea6b6aee
e97a7f33e2c3874038cb7666d0d511808b717566
81677 F20101106_AABVSE dondero_m_Page_38.jpg
a459553d0285efad197e184f41f98339
ae802c0f9f6ba52d286fbfd1fdca4fc64459f96e
6533 F20101106_AABVNG dondero_m_Page_67thm.jpg
7476f27c64b34c79aa7aecddaa26d98b
126e99c2ae0bd777c97bd818c4a8a3d7a76d0357
89355 F20101106_AABVXB dondero_m_Page_57.jpg
a25690eb6acbca40e85ee7450c317bc0
936d529211e16fa7e1db7bc78d4640ce31df9b4d
6167 F20101106_AABVSF dondero_m_Page_45thm.jpg
ea56a2df7fd3bedb29e30f58a03934e1
132ffa7bf90b5035bca51f7b5c257802f7c58100
27540 F20101106_AABVNH dondero_m_Page_55.QC.jpg
c2572e8f5f9a8e26a47dcd03d82442b3
7f0be6573e85040d939752a916157c5766913e57
53705 F20101106_AABVXC dondero_m_Page_58.jpg
a81baabe54f20d687949acb0704f313e
9da6d31ab2d6f7c4d7fbab49522ec0195543cc2f
1502 F20101106_AABWEA dondero_m_Page_11thm.jpg
e760c019c9e84209c50cd81e7ce8b643
108351de928813ec60c04470523292621dfc5dad
6723 F20101106_AABVSG dondero_m_Page_31thm.jpg
92a187ce030ad616d55ba9427e484a19
1a404e4649d40da320e4498ad7e4b89eb58016c7
F20101106_AABVNI dondero_m_Page_31.tif
31e749e97e7438a9af7575f6122f0391
e1841f5a6ecf32a53701a4108d00adc8c94aa072
55490 F20101106_AABVXD dondero_m_Page_59.jpg
b171d560ea4536391cc67174eb0a396e
e02ee379bee8dad6d168ab75b015357dced9716b
26179 F20101106_AABWEB dondero_m_Page_12.QC.jpg
203becc2a656e0e5d60cfd9a9d1f9993
446457c3efb95ccf5ded9d3ede43934b15c33891
1682 F20101106_AABVSH dondero_m_Page_16thm.jpg
29e105630f9c8fa491c2fce4c55e8fa4
15c96cc7e65b27855ef6cdf8c68f1b3855ce3dc8
2023 F20101106_AABVNJ dondero_m_Page_44.txt
673b95bf908527e516b5dd3bbf2dd010
17e334e4c1f3e1d72f69a009abf130ee72e4110c
83498 F20101106_AABVXE dondero_m_Page_62.jpg
f7e7b58ed057773b68d2e7069d733a47
f5745b9a1f570a60898a9eb93b44e3cb6aac28c7
6337 F20101106_AABWEC dondero_m_Page_12thm.jpg
a13f7a0d84ededac71925405eeb4f01c
500d9aa76ac91b6f4371b49b66df14e9caea3089
9022 F20101106_AABVNK dondero_m_Page_47.pro
e099ca0870f9da12f1e681bc3db638d4
c203c2d92308a21fc0491575c9820a42423207c8
88667 F20101106_AABVXF dondero_m_Page_63.jpg
43ae443a4b88aa3a694776d4bc0d8443
fad2435330896986b12b7e13efa5e7185c0ec685
6809 F20101106_AABWED dondero_m_Page_13thm.jpg
2cc9773791953c0b9efea8ee88238c5d
0a8d81e94919a195ebaf0f6dccb7be4540736d94
90201 F20101106_AABVSI dondero_m_Page_18.jpg
2ff0516f46243d77f31277b64525711b
794a42e03712d3ce5d77cf880cd6b14d4db394db
F20101106_AABVNL dondero_m_Page_13.jp2
6146c36f83ff4fdf00eea3b23a40043a
bda2585462334cbda3d0d5ceaf11bd8405eca4a8
89119 F20101106_AABVXG dondero_m_Page_65.jpg
7189edc01f0deff5a7c017cca9a900a8
528a7b259d05232621674f2fe905ed1ac18bfc0e
29399 F20101106_AABWEE dondero_m_Page_14.QC.jpg
b8b4fea0298399c5d44f578395bff085
2fe331031aa03ed4f5f9eb598a54da8f787ff42b
1403 F20101106_AABVSJ dondero_m_Page_71.txt
96009dca26d4cf9d540392b82e538ae0
6a85e91af0cbe89a53b099ab6483d835f905d21a
F20101106_AABVNM dondero_m_Page_45.tif
5bb08f1b899e9ce936bfcbd1b3829246
528284cceea1391c0fbef6d67e8636b7da4479ce
88134 F20101106_AABVXH dondero_m_Page_67.jpg
168661b2ff22cecf4a5590601e871cf4
749973a85a3216898c23b91681cfa7649b941733
6725 F20101106_AABWEF dondero_m_Page_14thm.jpg
737b0e2ec5aede82adda21e76a7cc2be
89e6a2c7aa6209ba20fb4cec6489f0d46c718896
88060 F20101106_AABVSK dondero_m_Page_53.jpg
aa6e37f898c244afdb5cfea8a05b7826
e79b2df064a3a78140ad81657aa6d8f564be88f5
2167 F20101106_AABVNN dondero_m_Page_55.txt
d7629b8e1cca848b096903858038dbfa
c0625b6e133af0cdb6703f14ee869d4b6372ed52
86784 F20101106_AABVXI dondero_m_Page_68.jpg
34302c520654a19410c71bdce617a7d6
18939071d0eb70f597e2e3caecd821217bb5f4a5
26154 F20101106_AABWEG dondero_m_Page_15.QC.jpg
7a8be67fa7da28d16215552fd7586b0e
1a82f83bf72f1a5fac3958559eaebce7b5f335b8
2113 F20101106_AABVSL dondero_m_Page_25.txt
9f8760e3f71c6a47337e62f151d4b5a4
203345898824fb8ee517f3c6c8217caf24e7e167
2114 F20101106_AABVNO dondero_m_Page_53.txt
e91026f5a1ecc54cd81c640846794074
f2d295bec836c825bac0b3a93db913f1bd80d7ff
99127 F20101106_AABVXJ dondero_m_Page_72.jpg
4ee723ac7aa1cc7cfd4265309dca331c
a008e28cb2f361501c72ce72de78b8a92e5d7e15
6359 F20101106_AABWEH dondero_m_Page_16.QC.jpg
b03db4c04e63d4cd99b1a42ead428f25
9fa34908744a964004750f4b6bf17a5a70dff445
F20101106_AABVSM dondero_m_Page_64.tif
1e0b4010e103a285afcba0ad6a26549d
0cd3b5d236e7a35c04272c1c20dc842c49578891
F20101106_AABVNP dondero_m_Page_25.jp2
b892b0947e48c9d2fc884b30152823be
e6692f128a8f3724228cd760b1245af3a8810e03
108397 F20101106_AABVXK dondero_m_Page_75.jpg
f5d0a1a27b71a23e625720f001049cb9
5396ba95f8d65c52fb07ab764b3e13e18ce15912
25830 F20101106_AABWEI dondero_m_Page_17.QC.jpg
1f8734c6ef4cf51291c0a82a9075f50c
526256a583e5a7f58fbbf00815404ddf8ce716d2
49500 F20101106_AABVSN dondero_m_Page_38.pro
e8beb633310d45c8893e408d7e02ac21
15dbafc20e6baa194d0a3316658299d324e9bc2e
2190 F20101106_AABVNQ dondero_m_Page_42.txt
5f8ba4d875d1bbf37b6928e26be5b6ac
14e9bc65411806f0f9ca0396474120e23708d617
91292 F20101106_AABVXL dondero_m_Page_76.jpg
ccc651e17936ccc8d50ca365fd6628d3
23296ef270d7099114f1c2c66330429f711eb3a8
6520 F20101106_AABWEJ dondero_m_Page_18thm.jpg
a475ccda6eb73fe25cf5f5ad1ad5f377
e8fa1a756afa93003238fd0d2765eddc99f67290
6559 F20101106_AABVSO dondero_m_Page_15thm.jpg
1aa3396328bad7fa721bf019719777bd
a84aede6910ce14ec59c605862b7d171ced4f510
51113 F20101106_AABVNR dondero_m_Page_44.pro
c952590c2214f0f9ec89f4073824b52a
e1a661af6d9baefea8bbaeec373818ef73a824a2
24950 F20101106_AABVXM dondero_m_Page_02.jp2
ec7d31fe6c859c2a8aedb1962e6590cf
84366a367abda103c9a9dad845b46814aa0db47c
27437 F20101106_AABWEK dondero_m_Page_19.QC.jpg
8a43b972cb4e40d4c90f4e4d5c18920b
0f288f807c0fc77f18d3d546b05b6d9968b4e7c4
52326 F20101106_AABVSP dondero_m_Page_68.pro
3da80d8dca72248b8073bd5d23f3e3a8
d559f96d5e6f5dc92fe59f8fa7c2ee0e5644dc5c
25612 F20101106_AABVNS dondero_m_Page_62.QC.jpg
195b6a0928e1140f3b95620d5d3e4557
525187171e325841bd1fb35fd1b00360bbb1e623
6913 F20101106_AABWEL dondero_m_Page_20thm.jpg
8c6d804952e7e77649667a81df2df826
31c1b6cb45710ff3c0ca4e457bb06c76c6fed92d
F20101106_AABVSQ dondero_m_Page_13.tif
c1ac23423a7b7a5dedaf8681d57c4e61
b84a5e6aeff4f423235c5e3311a0e8f62b9c96ed
27881 F20101106_AABVNT dondero_m_Page_26.QC.jpg
4ae31a2459ef0737262a4a3d83756e93
2a0513d4ef792c02d38e1948e860cbabd6da2dbc
735924 F20101106_AABVXN dondero_m_Page_04.jp2
13c8632c1eb3458129f7c7aa20dd09ef
2978deb7b281bcf41f56e3d87ad9dd327245f6fe
6609 F20101106_AABWEM dondero_m_Page_21thm.jpg
4b6c7f13f6c613e45112adcb353cc4ea
6eb60be6efb992177d27040949e0e6f64f9b1b37
88177 F20101106_AABVSR dondero_m_Page_70.jpg
907fb97d69ebdc2401bdbcbd49425d8e
853e6e1ed8b9d3da759fc702ddc80d8d2db64387
6413 F20101106_AABVNU dondero_m_Page_68thm.jpg
864f264c0474ebb6936a0be9ad5c7627
033438c0c354c9b4a907bf1d0b8ea6d3386f66d2
1051983 F20101106_AABVXO dondero_m_Page_06.jp2
45c575216e996323b9a5d149c5081154
e25afb152dab079388816deedca12e6f2c9ef15f
6868 F20101106_AABWEN dondero_m_Page_22thm.jpg
4889c24eeaa403bea2ce8d6ee8b48dc6
a3e03d5bec41caf1fb206bfbccbc4970e4993a8f
230751 F20101106_AABVSS dondero_m_Page_01.jp2
a7c7fc479146ebf80459f6c33577f27c
caa856ceabb6f3bbaa5434445f3f3f6c09fd4a68
85904 F20101106_AABVNV dondero_m_Page_66.jpg
3e0485679cf8ffec3ddd0fa83db14e65
1ebba0db48b687755b0a3fb178567af3e9aed8fc
114270 F20101106_AABVXP dondero_m_Page_08.jp2
e2b8c5720734323b9be41eb9940b7b4f
7fbb869d555bfc0859019e3c23834f5e91bae76f
6819 F20101106_AABWEO dondero_m_Page_23thm.jpg
6d5f07913878e3f23aa22de6654e3533
66d07f5e647b145ee7206e609986a6958c7aae01
21208 F20101106_AABVST dondero_m_Page_01.jpg
3339f4cb072dd22f8f1780a45258a2c0
25e7da4ccb49d4782d1f948dc9f9d6afb31d7e3a
3712 F20101106_AABVNW dondero_m_Page_09.QC.jpg
5d091b335efb772f951f0cffbb6d9c31
6d4e1e7826cc53c524431c307ed4e83c78d0dbe7
112926 F20101106_AABVXQ dondero_m_Page_09.jp2
08949ded5a8821139dd30dd598717597
1d5305295954bb95496febb599767cd3c46120d8
29888 F20101106_AABWEP dondero_m_Page_24.QC.jpg
4db878cf01155f51e00a652bf49c2ca2
0e95a26c0b954a6068d2ef4525fe386ff7502d5e
6777 F20101106_AABVSU dondero_m_Page_63thm.jpg
89a6a344ebfbbd8f53b8329fb1254dec
42a33e4ca8bc894d6c88aaf6b384fded6fb6dc7a
88855 F20101106_AABVNX dondero_m_Page_19.jpg
008bfe07a41d175373f45aab5185fe7b
f419d38aeb5e727cd385bdddb0c08bb677c3965e
192516 F20101106_AABVXR dondero_m_Page_11.jp2
598cd28a1807a7cc0f84ba5bc05fab65
8f4863bb495571429012b3d600d4316fc6bb8495
27305 F20101106_AABWEQ dondero_m_Page_25.QC.jpg
5833ef9407b7d4a40731e9af175f16f3
287682af01012a84657023637123c87972b1c6c2
29918 F20101106_AABVSV dondero_m_Page_29.QC.jpg
db6371d8a05d92f2b15fef5822b3ebf6
3a79ce50863e565172a86c29f6d6d2f270469aec
F20101106_AABVNY dondero_m_Page_56thm.jpg
bc22dc4255d7f90c1eba2434ae6315e3
7789cfd028cc862d8e4e17e4aefc4e93fffa22ef
F20101106_AABVXS dondero_m_Page_12.jp2
1625cfb8a748ec0c4105ff98469b619c
f759dff45f9f6b6a14d632fc3bd9a6b05bc17bae
28307 F20101106_AABWER dondero_m_Page_27.QC.jpg
7f5a96d95c77e0a5c2be332c9356edce
08b554d2ecc3297fa8f03c27e61ddb320e979bf9
4450 F20101106_AABVSW dondero_m_Page_71thm.jpg
f4c6629d1198f8e78940dc70a773062a
01884821f20016a78ccce1b3204b7af487c16647
F20101106_AABVNZ dondero_m_Page_36.tif
6ec2559c8f6a55d18a701845fac640c7
cc660b727df4c6577fd78a80d1169e9524f81491
1051948 F20101106_AABVXT dondero_m_Page_14.jp2
8bfdef437bc3c4542dd115c4cfbb7135
1d258dbd3b5633db50e12ebbab2c0d120ce90d4e
6738 F20101106_AABWES dondero_m_Page_27thm.jpg
6ae79d03f3c41816731433a108b7beb6
27e9cd6b4b6f2d9cd4d30576838d4aed4a008f03
1051974 F20101106_AABVSX dondero_m_Page_53.jp2
d4267051a9e40a048f38c32c075cc9d8
dbea74eb43c028c846128b8feb538e4841115a2d
240466 F20101106_AABVXU dondero_m_Page_16.jp2
10915116ec8cf219014fdc62f62cfcff
642e75b87493580338645a0774332d3d6bb3cf76
6885 F20101106_AABWET dondero_m_Page_29thm.jpg
aeab75a89f2604522bcb4a0e3bb1a976
01a73f3d848abe366b26d9b132fde7f8475511f0
6617 F20101106_AABVSY dondero_m_Page_40thm.jpg
4ee528461ec18476c6d5e155317db5ce
0329901492dd051178af3fe2c9a2c141a74b58ae
1051927 F20101106_AABVXV dondero_m_Page_23.jp2
af883a00204f18cfdaef2a12534a525a
c61d8ed919de5e6d4270e7f42e52dfb179ddf413
1640 F20101106_AABVQA dondero_m_Page_32thm.jpg
e5db82ed2c59eb1e868447dbacafae44
76b7d1d2ff8c550a46aa6d021303de4b7512edd0
28348 F20101106_AABWEU dondero_m_Page_30.QC.jpg
229f681aa04e269fc99ab4e6052c910e
1cf57d9383c425768a2eb15e71458bb31db0892c
1051981 F20101106_AABVXW dondero_m_Page_24.jp2
aacd45a6eb4593ddab44b8a615652f47
62f9df1a54b16872df4dc7e3773346b3b00ec137
56934 F20101106_AABVQB dondero_m_Page_74.pro
8016477ee3b3ad79898b6dec20e147ac
f02b429dd24b966480db04595677f6a8bdd111e0
F20101106_AABVSZ dondero_m_Page_44.tif
798d3209b29d5a5aba44471f90229981
2689d07037cf5eda35b9e852a602d77d0923c246
28267 F20101106_AABWEV dondero_m_Page_31.QC.jpg
569316ad96dc38505bc46dc72b30d46f
5fd5bdcf294d45a5136760f3cb86d8cdd19ac0ce
1051952 F20101106_AABVXX dondero_m_Page_28.jp2
2f7fd5ae4d60034286422d486e382de1
4a850a5d64628362f40fc63ad3dc82f97bcf12f2
1051922 F20101106_AABVQC dondero_m_Page_52.jp2
e7ab5210c1243f0641b5da105fac1d6d
e8621c2801076994a727326a74e82ddf16626ee5
22238 F20101106_AABWEW dondero_m_Page_33.QC.jpg
bbeccbeb0f27a388c9e12b405e2471e7
c8f15bc4062c85990b5f156bbd4d95ae7a6de9ea
920 F20101106_AABVVA dondero_m_Page_69.txt
646b12e930d318ecb4e8037247e7b65d
8f954a4911450ff07dafdc259018647db4f210ac
1051947 F20101106_AABVXY dondero_m_Page_29.jp2
f24a40645b79c1b7ba0aa2c50f458384
6a2c2362d22c974fa503d4d63483548f5d0aa77a
1051950 F20101106_AABVQD dondero_m_Page_31.jp2
e8d7b1d4a1f51185c1927e500460340d
936d4cb2ed043d0d5a2ed99a0977ed42de3e7393
28295 F20101106_AABWEX dondero_m_Page_34.QC.jpg
da73c32e4a76b3b56c482e36f0138b5f
793eb879b47bf015840575aee2fcbd88d77a7f40
87809 F20101106_AABVVB dondero_m_Page_40.jpg
2e05effefa7275f6a7637cbd9fa0b405
b07b2f072d8e9430fd1726c22ff4cf6532840311
F20101106_AABVXZ dondero_m_Page_30.jp2
128164945655b7a4757a0b22ae405005
48b84e0ac6f0951f195637b133ff5d7ab48a1f7a
53523 F20101106_AABVQE dondero_m_Page_53.pro
2ffdab3365113b0737fa47a0dc1d45c5
6737ce310e5365e337f68acd1999e1a22bcda1c2
F20101106_AABWEY dondero_m_Page_34thm.jpg
8df2d7d7dedad1c5b27426958430d9ce
e6d8b90bba9479b700ea2c10c7e962498a4a05ae
1051943 F20101106_AABVVC dondero_m_Page_43.jp2
500500926ae0ab4e88799e131ea76cf1
13b801471dddc4f6c9281c2b8f634bea8e640850
38637 F20101106_AABVQF dondero_m_Page_59.pro
559b508ddabc195f77572f800278e6dc
49f8103128818d8f61ea9a2533b9838be4a96f60
197 F20101106_AABWCA dondero_m_Page_09.txt
7907feae5ae97273615b3098322c791c
ee0ba4848f723a86095470e24c1ead2cdb25dfec
2054 F20101106_AABVVD dondero_m_Page_43.txt
778bbcfd68f8a13d2a8a83cabe655467
36d60747459fd75b214bc367a2e7895aaf700596
27648 F20101106_AABWEZ dondero_m_Page_37.QC.jpg
17f47e04642dd4e4ce6434327e1b4de8
98dda5cb3ed45fc31152045aa2f363722aaa4ba9
2101 F20101106_AABWCB dondero_m_Page_10.txt
619dd17d4060fa022e206d8a57307af9
2f107d3bfab4e141e1be72ad7a18a61f6fb0e2c8
6564 F20101106_AABVVE dondero_m_Page_19thm.jpg
4c8cf417ff812778e6856c3caba99b7b
24dfe9bcef3002ee4fc6e75b4aa800ee6a25b8e2
F20101106_AABVQG dondero_m_Page_37.tif
8e60fe40babcb0be46bed58b85303221
c5120e3b9207cad7f96195f3e704c7504799bd8c
330 F20101106_AABWCC dondero_m_Page_11.txt
72d6af5a5edcf2f9647f004895716110
e489f22ae8b6f1ffa3036b8e5ca8675ccb2fab02
25908 F20101106_AABVVF dondero_m_Page_66.QC.jpg
f1e95dd5b0c4d2d290a02658de7d3d85
9e03d0bb37e977c25923701e9dca9bb891a36532
2137 F20101106_AABVQH dondero_m_Page_35.txt
f05cb873fc0f514849d9d1b7182ccd5e
870aab6790d5cbe016f1842554248c6d0d4febb5
2178 F20101106_AABWCD dondero_m_Page_13.txt
c6fef7f4c3fb3abe4b7feb8a9d37a768
8dddb72123556bb88ac5be42bf641ec7bccbe1c5
94133 F20101106_AABVVG dondero_m_Page_56.jpg
425a537dd34ae5f3575aab1dec50c084
082791ab8e7729d364c6845a5b6cec0356ef6afd
89556 F20101106_AABVQI dondero_m_Page_26.jpg
797b5c02f09810177fbe29108496b580
e9b971cf7556e074a9e25621a4533eaff3c9333c
2080 F20101106_AABWCE dondero_m_Page_15.txt
79cb95017a032036ca682cb227ea5426
e67cdaf1035c1e7d9b9d87eaaf38cdd0eddea2d5
6807 F20101106_AABVVH dondero_m_Page_28thm.jpg
465ed0cda7ffaba6579023250fb40f73
62d28ed5f225d1475e9d742d56c0e8d90b805a78
6676 F20101106_AABVQJ dondero_m_Page_51thm.jpg
c01095754a5422ce5a710c75e92580d3
2bb3980adde3ae00228ba268e2a87903ff9dae72
428 F20101106_AABWCF dondero_m_Page_16.txt
2fe80a2797f8134cc42c0d3a516e8e8c
bd2006855ee4cc5059e8c0dc57d47fa00224ca95
F20101106_AABVVI dondero_m_Page_34.tif
b17b9fd581bd38e1e03315aa1366b105
ec62cc15cb7bf11087d290419b80227d4bc6e40c
90588 F20101106_AABVQK dondero_m_Page_52.jpg
bba11efeba67d8f22f6d0e0ef16b8fb1
49e43d7600f2e60eae1a94b4e679a73092017403
F20101106_AABWCG dondero_m_Page_17.txt
ab1ecb36f9c0b51e168564c17c0529db
1edce516c2a7961807e5957c3a68a9b8140ad557
F20101106_AABVVJ dondero_m_Page_62.jp2
d9f33e3d2c3d481fc3c305fcae72ebd9
f734bb95ec30fd2d2ffe55d323725cd6bd1ec917
F20101106_AABVQL dondero_m_Page_67.jp2
5701296ae10f602e81d2e254cc54c6d4
bd566bbb2ec092f52266f1bd1b8b729e020e1f97
2165 F20101106_AABWCH dondero_m_Page_18.txt
e67f2fcdae3bcd0ffbc2b8f69ab518aa
8a56b1d2ec274f9217624ba703ba7cd3225aca47
54886 F20101106_AABVVK dondero_m_Page_61.jpg
aac47d7126ba14a8fa636ee6854c68df
07381b7625491e07290a4df1a57812c2ea93691a
41175 F20101106_AABVQM dondero_m_Page_69.jpg
fb7427b8dc022def4cef860248504333
68eae3d7f595a9883403fe12708c001f5e221634
2099 F20101106_AABWCI dondero_m_Page_19.txt
5723c8f9981cdb8823c792194134a93c
a8b8be130bae634edb5d6f8b99287992d2cc03f3
F20101106_AABVQN dondero_m_Page_55thm.jpg
f40dcb986211fd95a65339729e1702b1
8cf134070a752324189ebf0ce3d2dc8045b260ed
2212 F20101106_AABWCJ dondero_m_Page_20.txt
2d6da6a9b929a6194343210669985c1c
e7f5a04e2b59d77b832f41f9923ecfc5b525b0d6
28692 F20101106_AABVVL dondero_m_Page_13.QC.jpg
7e0cd0a49ec6f8174477892c44a79c2b
8f8253db8c916795d53e1e191f4c4b57d1cd4e66
91991 F20101106_AABVQO dondero_m_Page_23.jpg
d0657f0f2dfd06dc08be9e04f93bfda7
a4d072b3e27f772ee685713bc42c5fbd250deb38
2102 F20101106_AABWCK dondero_m_Page_21.txt
b1fe9bacd7dfa665a2faf47148d11487
5a748b73906ab8281839add142748cb825d093c2
5939 F20101106_AABVVM dondero_m_Page_32.QC.jpg
75827b15e0a5e2a4d99378cdd74cbbb1
ce3e37700c76c3856cd64648a2a8f7570b011710
3562 F20101106_AABVQP dondero_m_Page_02.jpg
0683fb6b4830e525dd5b835a9ac232b9
93ae101abab4bb7f443260a952897ed12c483408
2159 F20101106_AABWCL dondero_m_Page_23.txt
e6c4eb7ecfe7f7fdfdb8e0754f86f0c0
cd769f33f66a60670a92cd4a3bd49d4d064054aa
F20101106_AABVVN dondero_m_Page_38.tif
ee062a5b1769fc3b0b8be5f01c5e9124
3e96637ac24bf5f907bf821724ffe758d6f721f7
6882 F20101106_AABVQQ dondero_m_Page_64thm.jpg
da4142ec77dcba6966adf7c68901b45d
9e8e8ff0acf9e6d5de44b19710aa6dcafa450f38
2299 F20101106_AABWCM dondero_m_Page_24.txt
b5670d0a1603e0915b9c6c7f211c4db6
073097278154580feb795ca1964950057bcb5d5a
53254 F20101106_AABVVO dondero_m_Page_57.pro
a06cff022b56e0b85f0ae1d99a6b4401
5a42e11d6c427dc3dd1dab0434dfffe9ebd4d171
93211 F20101106_AABVQR dondero_m_Page_22.jpg
d280d8aa573f8f513f1f8c23defeefc6
0cec3652cde1b060d31269557a84a735633ae47d
2131 F20101106_AABWCN dondero_m_Page_26.txt
4670bdbd71c8118f4ef8f0c6713d4194
5904cc1730b5809a76deea1c5b0e998f20853f15
88070 F20101106_AABVVP dondero_m_Page_51.jpg
2cd9c34020345d1321a7cb41c6eaddf2
0ec6b3fddb7cfc875565a1bc74e0eba5aa89c8da
53988 F20101106_AABVQS dondero_m_Page_34.pro
f036a625568afed01c139f7fb3b37cf7
574decb3b1382bbee7d97a1e3173b06eeecaf83c
55332 F20101106_AABVLV dondero_m_Page_30.pro
37212e7ba01b5b7807a07fde2a12cfa4
9dda328c1afbc4b085ac987a9afab22a78435290
2194 F20101106_AABWCO dondero_m_Page_28.txt
bfb600f62effdd947964d75a0857208e
f0fcfb47c09ef6a4588870004923096ef468939a
973 F20101106_AABVVQ dondero_m_Page_09thm.jpg
a791a2c08bb51bbebcd8d4c44bd60915
62a7d99c04c0f21ba0b24620571f551a238c5b54
F20101106_AABVQT dondero_m_Page_14.tif
0f7d01408f016c5f67eaa93112793752
7887d4e3d14015b4c24a0db1d42723057026b68a
54930 F20101106_AABVLW dondero_m_Page_23.pro
958314687291390216d222876157650d
c27ed07c27836726ff7b550cf5f8ec2ade47c4d4
2398 F20101106_AABWCP dondero_m_Page_29.txt
6fc48ca76d5cc70296e0126b8f321b09
6bd3801ace18511a8473354074da066de63e80f8
29874 F20101106_AABVVR dondero_m_Page_46.pro
09ee088f4461313b3ac32207e114d3b3
851d0dc530c0b955e4d9fb6ee94c0029d7b08bf5
703264 F20101106_AABVQU dondero_m_Page_60.jp2
c14602e32e84a589d9c3ba5eef2c1e36
9c3f6633776442d5800b286b6d7487b364c30f18
F20101106_AABVLX dondero_m_Page_51.tif
a5f02a636516601d3ac001ef2e79226b
1d498e13d52d894b5b496094c34faeda64830321
2174 F20101106_AABWCQ dondero_m_Page_30.txt
9ec9202030a5f76daab5d76881b8088c
62d53f298a1b8b099d44d3b7696a16ed895693ca
1750 F20101106_AABVVS dondero_m_Page_59.txt
061b4d14e7c350189dc7a1d28bf174c7
cde11cf7434f6f052e31801e99096add9b093c74
1797 F20101106_AABVQV dondero_m_Page_33.txt
8d5b153794c3754a1cf622d20d716ba4
ce655cf10e95109e95558ce02ed192613f8897e0
10539 F20101106_AABVLY dondero_m_Page_16.pro
e171641a6c2a27d39d2c0bebc0c510a4
6114c7dcbabb88925f8bec14cbcff08c96d18736
2191 F20101106_AABWCR dondero_m_Page_31.txt
f49dce2f9b08c75d69b336f67431c5ad
0db9b97af21ac8feae7e5ae01f3cadb4c964a269
1051951 F20101106_AABVVT dondero_m_Page_72.jp2
ccb9e8018dfbd7bdbe557de4881a7642
a4e03b75a57579c32d79f05d3612d85faf91b92b
51096 F20101106_AABVQW dondero_m_Page_43.pro
3316b5ebb70a6f85793c52eb961e6d59
c73cbdd8431f29212055779188fc8feffcd237d1
2091 F20101106_AABVLZ dondero_m_Page_27.txt
669042772cba8bdc72ac388aeba626b4
a80d10a76d6e56fd65dee37982cb4597dc299842
379 F20101106_AABWCS dondero_m_Page_32.txt
ad823e5f5f07450cb65b11241d0138b5
eb9453eda5cf95ff14662b224aaba9cb4ab46ba1
54473 F20101106_AABVVU dondero_m_Page_65.pro
e416bb657b2bfb82502d9b3b125ca9ef
42c8f02323b82ecc01f6ee69a6c1b4997883153e
38919 F20101106_AABVQX dondero_m_Page_60.pro
db11fd14a1da79601f8a8dc644bd2764
76768075f38c5f14583bc5dbaf49082b8ec1a864
2092 F20101106_AABWCT dondero_m_Page_36.txt
25abd28b7a169a30050ef1591c7f1ead
df73a362a8fe28d89f2b22e4d688000bcd84cf72
6784 F20101106_AABVOA dondero_m_Page_74thm.jpg
37e4a4e984f989a48b157bbfb9c1e00d
a9e0939e3755f96665c9a1f49e9cf9b9a44c8b2a
6726 F20101106_AABVQY dondero_m_Page_65thm.jpg
cb35c740254a9f97d24553603c178114
492f54b03f474982e3f1c78bb5252a78da4178a1
117377 F20101106_AABVVV UFE0022196_00001.xml
b3f736ae19d785b44f98820c3d3363d8
9fd3f021fb4b1985ef4f0ad7519af90a04167615
2020 F20101106_AABWCU dondero_m_Page_37.txt
b3b642b80296ede39d2c8ca7972b1310
b46092ca0b81b752fae33120700ba1be3863faf8
F20101106_AABVOB dondero_m_Page_20.tif
cc094106be7f4215e9e7b19b846eaf40
b27caf66e050489a500b911def89d5b191df6fa4
56384 F20101106_AABVQZ dondero_m_Page_64.pro
63d201dbda3767b829c7bc23ed2efa58
174ab5d0c09a4578ea13cc06f72501e22c9d14de
1953 F20101106_AABWCV dondero_m_Page_38.txt
fbe7c0dd7cf0a46a263e97d10384e10e
78e59d56fd713228ab25ff4cccfaae8d61c5945b
5482 F20101106_AABVOC dondero_m_Page_33thm.jpg
54196823c46b9bedebe9a54e0fc2c257
1607aed294d14eb1333365cb7d29940703f13142
2042 F20101106_AABWCW dondero_m_Page_45.txt
fc9db03c3da1fbd76c28e029aa40e35d
3d83e6820087d52d557343b6dbbbcf46e804e177
1051941 F20101106_AABVOD dondero_m_Page_39.jp2
66d5a2e1d9b83bed4a6ccf9f1419e210
d6ec53a082a85c5406c1f36e729d7380731cd375
684798 F20101106_AABVTA dondero_m_Page_46.jp2
08e799f51f7f5d48ad099047445eca30
3701a0f3d8fcec3e94c85ec639a0f1c2746603fb
56774 F20101106_AABVVY dondero_m_Page_04.jpg
9b29120446ffca3c2140dd8f99efa54f
5c4aba2b35676fb7e04efc64672a03fe755b468a
85796 F20101106_AABVTB dondero_m_Page_05.jpg
db4d60df31f01057e711c1ce5d55e817
23c3bb5f7e074af69298b5dabefbeb74556fa80d
45701 F20101106_AABVVZ dondero_m_Page_06.jpg
c2154810d450cfea7d89656aa791241a
4dbbfd43f98f8083b225c0f340a2fc669c130901
623 F20101106_AABWCX dondero_m_Page_47.txt
2f08a051551e017d9b61028bac005151
136969df0b23aa23ee43dadd715b677ec9315aa4
2059 F20101106_AABVOE dondero_m_Page_62.txt
7aa4b67a229894fe0f36e58d6605dd19
38c9ef24f0c3a11219ccff3995f6e89af2ed2528
1051908 F20101106_AABVTC dondero_m_Page_49.jp2
47cd8262d011c0c68321f642e2087948
6b0071c19ff526a9181ea7156e55e54081cfd246
2002 F20101106_AABWCY dondero_m_Page_49.txt
ad5fbb69404f0781afe87aaae6810ef1
ae50e3bddf1ead958efede930cecb68650bd1962
F20101106_AABVOF dondero_m_Page_27.tif
a6baee3f35579b714225306cfe7b2102
cce57d6d20a5023d7de055ef8e6a1f3bccce9d90
1051982 F20101106_AABVYA dondero_m_Page_34.jp2
e92ecb086f46dddbcc68e43c28e42b65
e137dab4b729572caeb0f677a42019b3ef754285
F20101106_AABWAA dondero_m_Page_60.tif
8bd1949e6938b16ec06ecff4ddec82a4
44994c91c946a6f4c5e9b5011c9a910655f9fb30
6154 F20101106_AABVTD dondero_m_Page_44thm.jpg
a44014b17fb304d423aaa0698f5467b1
8b1ff7922831583922bea7a5316bb5dd3a28e18e
2203 F20101106_AABWCZ dondero_m_Page_50.txt
606853767fd945a4511c84fba8701c50
5cffd3ee7ddce9b78c993367cb4e3cf69eeefca0
1051984 F20101106_AABVOG dondero_m_Page_18.jp2
16bcaa1f7f16500e984e52baa475dfa5
c1debff1e042397177288030abcfa3b3e32feadb
F20101106_AABVYB dondero_m_Page_35.jp2
e0c38bcec90eefb6ccc84a0f1404fbc1
591b55f1e96a79979c005814e0b086a76dedda69
F20101106_AABWAB dondero_m_Page_61.tif
0ab6736820a21b39c8096ac789709f9a
7354901656b842940e9010fd69b5500ebc842cb5
506 F20101106_AABVTE dondero_m_Page_03thm.jpg
13f80228d7c46736fc6fbd02b54eeb61
089252e447052e448b9912d392d24ddcac5edd76
F20101106_AABVOH dondero_m_Page_18.tif
c79dfe92a825e0aa534f0acb24adef21
8d681445d9140f2e3b988358b97715eadfb9745f
1051973 F20101106_AABVYC dondero_m_Page_40.jp2
947d75afaf4ea4882c3c7855460834b2
61e5e57d7c0739c38726f2a09dd71db8c0dbd92b
F20101106_AABWAC dondero_m_Page_63.tif
e62b3cbd00cf825badd1893ceedfe95b
ecf42d6cc2f0bbbc894882a77e6aaee3e4677129
59051 F20101106_AABVTF dondero_m_Page_71.jpg
bf36b1ecdb9e5bc9fdcece34472f275e
36f2df8b4a99dfd03d3eb0d965cdd5c8fc122ca0
6462 F20101106_AABWFA dondero_m_Page_37thm.jpg
53762a1e8ea835409546269ad1b9d381
800f663d97c6d1908dfc835a305837644d93d58e
F20101106_AABVYD dondero_m_Page_41.jp2
9ded10596ff05bafe78cc16c66e2bd4c
51199b3308f66044e469688e4930265c984d588a
F20101106_AABWAD dondero_m_Page_65.tif
32c5044b640c8f9de5f5e04cce973cef
23f3be2d520455a7e79c96fafe221e815b441809
724684 F20101106_AABVTG dondero_m_Page_61.jp2
aa122bfed94b1f65cb30ef449581653c
3b9b96cd57c009408a18813c6590ce0f0c0acb90
2110 F20101106_AABVOI dondero_m_Page_67.txt
ddbdc2e2ca9f33d376357effab4b6faa
f9e38017f6d579404705b577b4fddee3b4d8e450
25887 F20101106_AABWFB dondero_m_Page_38.QC.jpg
cc46d4f641c703be192af09b1445b189
44ae8f27bc9485ffac4798280d75cc63fb8a2b62
1051931 F20101106_AABVYE dondero_m_Page_42.jp2
17c15252ebb79ebce135cea4779fd093
3eaabc1e58d14352403726c6ac0ade648715a0ab
F20101106_AABWAE dondero_m_Page_69.tif
bfa2c56eee08fb58b5eeea32ba2c6dc5
c88def112ae1ff58a700b908bbacdf6f23d7530d
2628 F20101106_AABVTH dondero_m_Page_73.txt
a1a81cbaa6182bb09f9bec1028d3a9c9
fb29bc7fd312d59f1122cf77135a0304a4c6f9a1
4189 F20101106_AABVOJ dondero_m_Page_04thm.jpg
8e4253ab53aea675834f62e1dac42071
4452becbf1372e1c103b36332126bb61e8dd8fff
6206 F20101106_AABWFC dondero_m_Page_38thm.jpg
0235ae198b6cf0813f5d0322916336f4
289e717e8f61920a83931141594cdcee0d01fbc8
1051978 F20101106_AABVYF dondero_m_Page_44.jp2
82a19ec656cd5546c1de218db8b79dc2
75d31b71ea1fb113f3f1a03b25a727b65cb4814e
F20101106_AABWAF dondero_m_Page_70.tif
e21104828e8e3f455541a9c0504b55a4
d0a5ce0baf930ecabf7c420f77de9454f214e9a6
90805 F20101106_AABVTI dondero_m_Page_14.jpg
a5f114fc68e30f0a0cea40a81eb681a4
598c82478ee7a9ced4815a5480d3b8c5c419875f
6263 F20101106_AABVOK dondero_m_Page_17thm.jpg
e5540898e79b3bde7743fbfc1500a322
ac16a48360698a011f8ccf171be161741462be9f
27864 F20101106_AABWFD dondero_m_Page_39.QC.jpg
4fc760d137a7717f7c7d07a9011f8943
199277545b57a18bc727bfa51643678d6dd1fcb3
196931 F20101106_AABVYG dondero_m_Page_47.jp2
85ddebb4cbff35df11332ff05c6a9dcd
753d6c5728061ac63c6515f856096515adb26b4e
F20101106_AABWAG dondero_m_Page_71.tif
6b0ebb36b124434c82dba158de5e1340
dfacf5573d641e3c986b5bd4c0632e1099c153e0
53337 F20101106_AABVOL dondero_m_Page_51.pro
cfde9b2f94a05793436378aba89541b2
2b3c538a51b380120567382889bd4117c21b6bd0
6358 F20101106_AABWFE dondero_m_Page_39thm.jpg
ce5f4168b19c1e48ad4dd6455dfd460e
dc32b2394f02e51c0f0ba4365ac16aaca75ab55d
1051963 F20101106_AABVYH dondero_m_Page_48.jp2
fb6809ed42084fa3bde1ade18025f1a6
0cc2a0183bc17fb5b15fbce561d7c09bcdee4a08
F20101106_AABWAH dondero_m_Page_72.tif
49f75a33d5fe82d00acc21bc89e248cc
8d04a7f6dee378d8e86a95b572c18b26a4b9ba70
58296 F20101106_AABVTJ dondero_m_Page_24.pro
38b157cf3be64ba6d833d6d07088ee07
41d7b3a0a28a47312a2da4937cda0d43b3e70425
92731 F20101106_AABVOM dondero_m_Page_64.jpg
7a49e12f84ce63a9fa2a1cbb3ffdc4a8
3106d39a5bae0c5e59b497add409c7dde0d17f42
27873 F20101106_AABWFF dondero_m_Page_42.QC.jpg
bc3c660c1c5eae6e74bce507727a7f5e
c03b93c5425a23ecc2d24b8141212929dc6c7868
1051960 F20101106_AABVYI dondero_m_Page_50.jp2
ba25feea930dc291f1f516c6736f0fda
5358b0b43a9074454262d4b0da726d3ea2e272ce
F20101106_AABWAI dondero_m_Page_74.tif
0bb4d25f9b69f4e3d522cd9cce76736a
7587771f24e70a6763dd9c88e42f41d3c311391f
F20101106_AABVTK dondero_m_Page_66.tif
525f47ca454823da7d869af6f7b30188
6e63f3d2b81254ecf9cc37a99178f1087b12c6fc
F20101106_AABVON dondero_m_Page_54.jp2
61e00224f5efab9a4682a291e38081e6
9df5bad84fed5edf8a23b2dc00bd215020e4cff7
26095 F20101106_AABWFG dondero_m_Page_43.QC.jpg
757c21a83b0e7d4733aacd4a367a302c
63c2081bc68f37ce4d783a905065074d4ef253ed
1051976 F20101106_AABVYJ dondero_m_Page_51.jp2
470c0a2c2efe808a73d9efa9d10ad3b2
b6b5f621ea12af5b86bd92ed1242221f116a1066
F20101106_AABWAJ dondero_m_Page_75.tif
5f1d277d2ccace130bbe9917b79edf4a
0a4e9977227e2fad5c4b04e722895a2647609655
1191 F20101106_AABVTL dondero_m_Page_46.txt
ecfaedbfae8b7fc3cdf6c03e1cf27521
82220523f083bccac079f5b22d7a56d177ea1fb2
57968 F20101106_AABVOO dondero_m_Page_56.pro
fb480f76af1ab9c6c56e96f9387dc499
57325c7e127617ba91ff35967e5a731f83699b30
6307 F20101106_AABWFH dondero_m_Page_43thm.jpg
1fb6269e2f6e66f60f6ad2cfd374e5ea
782f55ffbf96d8ba21d7d63274780da437086820
F20101106_AABVYK dondero_m_Page_57.jp2
3b19dcc141670ff83410cb3fbbc1b9b2
f9e2e59c4b13c2b411151977be5bc0ba19491765
F20101106_AABWAK dondero_m_Page_76.tif
ad28064177bdfb5efa34faf9a1b7fded
378ef113daaca6d98672f3ab8cab910a80f9814e
F20101106_AABVTM dondero_m_Page_27.jp2
61307baecb941f2a809793c4bb9845bc
934b45c78630b09cfaf29a3c3f3b40524a368a04
54179 F20101106_AABVOP dondero_m_Page_63.pro
a16c45f2ad5f2445450af6da27efa475
9b28ed01051a96198ff81897ad80ebc47ca53600
16233 F20101106_AABWFI dondero_m_Page_46.QC.jpg
1a82e0c9af4fd6ff8997a26f0f71b939
ec8f3ee68c112da35e8737717ed152a874f3393d
701603 F20101106_AABVYL dondero_m_Page_58.jp2
3856d96e20e3464bdf83ca6684a114bd
5a7cd6af009718fbf9b23d092de26cdda9595f16
F20101106_AABWAL dondero_m_Page_77.tif
f0a5601a5e1625fc67a792a29c5b0b5f
51928577cf22f816b4ad42d0d7397cdbfc471fb1
84270 F20101106_AABVTN dondero_m_Page_15.jpg
65fcae5864eea1b3e9eab47ce1187897
60b17c8deccb7ebab04f33e54dbb4e2c33645798
2121 F20101106_AABVOQ dondero_m_Page_57.txt
5846585873e35f297f9ad84b612ef561
4727f4307d51c688da3a6505acc2bbc397a3af1b
5266 F20101106_AABWFJ dondero_m_Page_47.QC.jpg
9d5973680fc0322da07f484661c759ff
43f6623ac2230f9a4e689524f9af8374392b02ae
729047 F20101106_AABVYM dondero_m_Page_59.jp2
06a2a19acf034a2c8cdf99c887eae347
33056d7799c1e0b182231d5a69992d8f5c8b8c11
32530 F20101106_AABWAM dondero_m_Page_04.pro
042d3a084ba3265d40afad1c37eb1ffa
3e3e82704ef9e64e98cb8355d3e32f58600d6592
27301 F20101106_AABVTO dondero_m_Page_77.jpg
69354357d811d7a84eb175cc702c788c
2d78712ee191108b92effba1c0d27fe6a5f7e3b2
68579 F20101106_AABVOR dondero_m_Page_48.jpg
216d4445d01e0c1673717b980fde99f5
243128c565789c0d867234702caad879facee3b3
1635 F20101106_AABWFK dondero_m_Page_47thm.jpg
9269dc960bae8248414fac83d7bc9ab4
4a4cd9373956adec3dcc8c52b80333f029952f82
F20101106_AABVYN dondero_m_Page_64.jp2
f546dedc5cd8c021e9df77be6dbc70f2
108da1dc76fec60f97ddd6f4f97d92e1da55a2e4
91576 F20101106_AABWAN dondero_m_Page_05.pro
65a08e9b79a12ff3419ce9fb0281a494
fd4df5833d2c488f94032606a37444247dd6ea1c
53451 F20101106_AABVTP dondero_m_Page_19.pro
43c3cdf334a4415f66d5209eec13368a
a1a9a62f6cd0bf12ba1fe53faa2c78bb8c085f7f
58826 F20101106_AABVOS dondero_m_Page_72.pro
56735f4f5bdc89ff8602cec2fd91b039
571356087f9fe2c28e05312a4f4551add6ad12a7
21384 F20101106_AABWFL dondero_m_Page_48.QC.jpg
8b6e7ffcf696a0aa77de8ce6e0ee3ec1
d5a3ea9a8ea04a3255d9c8b2b82ba041485a8ae1
46028 F20101106_AABWAO dondero_m_Page_06.pro
c6ad5cd6a7d79fe5a463b338c9b3471f
cae77fffb2e17578d1b9f35454e3ec0c2a267dc0
1051971 F20101106_AABVTQ dondero_m_Page_15.jp2
ce478a4580e0c0f63a42266e36a063c1
de72a52ba1bdf4fa79ede4dfa4fe132588ae80ec
F20101106_AABVOT dondero_m_Page_24.tif
38d31359070fa4a17845ce91edb7b3bd
cb853cc906d43ad13a40ac70823ef5b1c3e6accc
6071 F20101106_AABWFM dondero_m_Page_48thm.jpg
d3950def70e7fd557682d17e4538461c
93b3eec086d6041ee5e814ae8a78832ad9869c34
1051939 F20101106_AABVYO dondero_m_Page_65.jp2
dd360d58592af573d9959878fc6719fe
561d11118832a47332202ff182e134971a133dd4
15918 F20101106_AABWAP dondero_m_Page_07.pro
9e6d9f7ff5d63b1e7c22c22e26e6462b
f491525399551ebf5fd3a186907eadb435b74957
809 F20101106_AABVTR dondero_m_Page_02.pro
12352dbf7a1aebdf32b3d0a495f16f88
d9b7d85be4133597a6b2edeeaa704635ae563ee9
2118 F20101106_AABVOU dondero_m_Page_14.txt
4b6da794b989b8bfd61cdf969421063c
0f87b10fb4bd2e2ecbb8fb5ea57d0c72862c9e4c
24321 F20101106_AABWFN dondero_m_Page_49.QC.jpg
d60bd56d2a0670f38453ea68daca75cf
7fdc473b20179de8c8694407e6f109e48fd0e0bb
1051914 F20101106_AABVYP dondero_m_Page_66.jp2
4e7ec0159f3cfd7577ee5496dd0b6485
f35ca78486f5d35af1cd3f27a78b9e3aee0f1782
5821 F20101106_AABWAQ dondero_m_Page_08.pro
bf2b7fed5ddf4ac3d81ac0b0f84c96a1
6a4e3060e1f91cbddcd5e686d3638da646c58f84
28717 F20101106_AABVTS dondero_m_Page_35.QC.jpg
54aebbf2557fa387a2e9a0017c63a79b
5d4bfbb103d4fdf889cd21cae53d544f080d229d
F20101106_AABVOV dondero_m_Page_67.tif
075b4869dd38343959a1ca3ea8eba96d
1a93be2920c876aa2727f8650bd85ee6bff358df
6724 F20101106_AABWFO dondero_m_Page_50thm.jpg
112aa45938e1016f6cdf8175979b9b1a
4631bb16afcfadfabdf1892cb35d1f76a2af0401
F20101106_AABVYQ dondero_m_Page_68.jp2
3361797a26ed2aceb68f192dd2964c75
a2f8586cc73bf23b26ef88dd4be38156454c3636
47757 F20101106_AABWAR dondero_m_Page_10.pro
f89585579e5ab2301c5373b8dfeda880
c242995fa6d062fe9b1e9406aab78cc7ea55f83b
28563 F20101106_AABVTT dondero_m_Page_18.QC.jpg
2331de967555284817045b2aa96dab2c
d5cf7fbfd2dad7209920813b5e96ca40dc9929b9
1051932 F20101106_AABVOW dondero_m_Page_45.jp2
64b19ddfd0be71d22a6e075e3d4bb136
d93efae9b333e7a22b74bbb3c9205cc18b0db7d4
27607 F20101106_AABWFP dondero_m_Page_51.QC.jpg
e457fcf3754343e3ab857751119ea4cf
1c2cd102fe04cb6c65b4e9b09dfdd72be0bc5ebb
8236 F20101106_AABWAS dondero_m_Page_11.pro
9d08feba98366905854c8068dc6429f4
09d1edf74a76b053462622cd61ec114127a18f43
55033 F20101106_AABVTU dondero_m_Page_60.jpg
a6132d7346809db59ad54068b3d1df1b
9dfed09884e9a6b81eb96e6c44049e414c1a8357
51864 F20101106_AABVOX dondero_m_Page_15.pro
0a108340c68e29f9c118fdfcde513a8b
9cd0ed1edf90e3ded6ffe4da73d95604287bd159
523308 F20101106_AABVYR dondero_m_Page_69.jp2
352bcb39ec6a71578efd30a40ff603c7
2228420e43673b92f43753695abfbe15d5bd6b2b
26512 F20101106_AABWFQ dondero_m_Page_52.QC.jpg
830ca5f91c96ed9c29b6d03e50d76973
5b88c1035c9d67fd87125d6164886874113930fd
51244 F20101106_AABWAT dondero_m_Page_12.pro
9ccf56ea1bd5df5c651c26397fe39aeb
69fb57bfbd69b65985a59d2f494841d8b9b6a0f5
F20101106_AABVTV dondero_m_Page_21.jp2
4e00cf2cfe84ee89d62b7a542bbf06aa
5483377ca05f2efa3be812aae3cc202ea1fadf62
6398 F20101106_AABVMA dondero_m_Page_70thm.jpg
ffb149b9638ba47fd77f55624f44c654
35454716c17fcb13a69c932c65ca6476a35e5145
10345 F20101106_AABVOY dondero_m_Page_48.pro
22994a059c317229f140887512d4a6e2
b4efc1ca75527d1b8193c90973a85affb2d207f3
F20101106_AABVYS dondero_m_Page_70.jp2
e55b1e15bcffd4322b4478e83bbddb82
c6f1d24e9c0819541bcac6e5d467319c66f9f362
26664 F20101106_AABWFR dondero_m_Page_53.QC.jpg
5fe695396fd9dd909deb7178595b5fd2
0634a5298cfea560fea30f6235e410c0f9c72acd
54543 F20101106_AABWAU dondero_m_Page_13.pro
9f69d8a5e9db7b3b05fd90017552a2a5
e796a8b65498c1d541d6646e1153275f088726b6
6569 F20101106_AABVTW dondero_m_Page_42thm.jpg
9b483681dfbfedc83a86ed127cfdf884
a336c614238b82bd32cb52e7efa060b7ee6a5d4d
F20101106_AABVMB dondero_m_Page_22.jp2
3be37bc1356e2743166ea10fb32f8a23
6b77352d4d30e43371602559ae3a1397e74c5c32
16210 F20101106_AABVOZ dondero_m_Page_58.QC.jpg
44c6aaf5580230a39ca202970af4980a
d0f013c57c2991630e5a61e740ac23da81a1a48d
788729 F20101106_AABVYT dondero_m_Page_71.jp2
7f838a58e384ffb706acb427aa6f823a
c14701a6463ee72269751df38e2f436b0822c1ef
6612 F20101106_AABWFS dondero_m_Page_53thm.jpg
97839cc7247d09c595870dc66b9d2af1
efd8a5917a06c519648a214793d79d1f54134874
56703 F20101106_AABVTX dondero_m_Page_52.pro
fcadd129fa9d311af92baf878449570f
ad01bdcf79d9c4ffc7698639339e1c11218b9d2d
F20101106_AABVYU dondero_m_Page_73.jp2
75644755bd7577285b666d7614b5464c
ca6e04580f23c5e26e8bc388aa3004b413a1c4a5
26488 F20101106_AABWFT dondero_m_Page_54.QC.jpg
8860ee86f857b52d7e4eb48ea7441adb
2a570276bf25a337186d310f56627a0160a7e729
53991 F20101106_AABWAV dondero_m_Page_14.pro
f7c7b442d794120559d05963280a3041
ed28cb5608fa7a1f4cd3b4057ad67ef7a7624a96
89513 F20101106_AABVRA dondero_m_Page_42.jpg
266398b7febf65271840757668bd11b9
7f7df7601a1677195db9e8dfe553e8aa6c2b2748
6524 F20101106_AABVTY dondero_m_Page_36thm.jpg
a86170847df6db1aadcd7b92de81bf23
0f337b71730626799c8a1a48f382bb9c6406a5ef
F20101106_AABVMC dondero_m_Page_21.tif
975750685c3a51a308d343186e5915d9
588eac81f8bf0bc1d02b4f6795a60bea4605826c
1051977 F20101106_AABVYV dondero_m_Page_76.jp2
08d42b02e0b13db6c32c1f3605af032a
747fe20c34e82fd7e4b82ddb4fbe15c32c8b264f
6357 F20101106_AABWFU dondero_m_Page_54thm.jpg
5a10e535abb5213da2f00158c6e6ee9d
52e6d8db36e84d2bdce4f7e4aca16207eb249e01
49080 F20101106_AABWAW dondero_m_Page_17.pro
3b69a7ab0a87b58a96d678f7e2e5c3f5
54eb5bc0bfb08e6c7bb0afb85dae5c678306b9f7
32498 F20101106_AABVRB dondero_m_Page_03.jp2
17b50585936df81d80c7562ee67ffbb3
285ea908edf6b4b1dd6f52082fc4a5eed6b9c6f4
2313 F20101106_AABVTZ dondero_m_Page_74.txt
a5dea031f2437aa98ede77e41ce5a48a
f8d15ba0db397f9264719b9b84187e51a728bc42
55125 F20101106_AABVMD dondero_m_Page_28.pro
a4c9fda8ed9bd1fa1cc09fdc74cffd5d
af725484abbc3e6ad222989f9392fc3bef0b9647
F20101106_AABVYW dondero_m_Page_01.tif
fc8078f7e44c07ad04e28b524813298f
78705e267e8fe2cfb769539f932458c92c88e50a
27679 F20101106_AABWFV dondero_m_Page_56.QC.jpg
70b2ce91e744ba24b22ecf9b339d4865
0a07ed52ffe549ec5ad697ea39ab4e8fbeb83143
55194 F20101106_AABWAX dondero_m_Page_18.pro
c1ea04badd408fcfbb0c8092b987ad48
e589a7903e01295299de2d5e130851500a260a22
F20101106_AABVRC dondero_m_Page_02.tif
28dc724bd7a7bcf1083c68a73bcd1d70
2eb0ca1355f04abd6ab60d6b2bec8e3376835599
F20101106_AABVME dondero_m_Page_55.tif
d692772cdba185819361f7be17f8b88b
400d21a20aa727cc0c0bf5085d95060899c8e70a
F20101106_AABVYX dondero_m_Page_04.tif
6b593cb4c5e3265a7a178de843c0f96a
db0fce5adc347554de6f155d1ee21e41bdeef9d8
26765 F20101106_AABWFW dondero_m_Page_57.QC.jpg
d97247d26901f4b739c69f290590d99e
312502b5fd11aae9856f5dd39c3e1516cf1eeb23
220148 F20101106_AABVRD dondero_m_Page_32.jp2
3a053f2eac14f68989882e76d6fed7cc
c027b07c41628f3421387d8ff641093f6e7c43f0
4262 F20101106_AABVMF dondero_m_Page_09.pro
798983b560f3192903bb69cfdc410a87
a04108e1f21a991fb351d30a687e4074be0a0aa3
25126 F20101106_AABVWA dondero_m_Page_07.jpg
10b6bf31bb53fc558411f6a4aa76e397
bed8003fe30f8426d6c0be5c58c1c68b7663f215
F20101106_AABVYY dondero_m_Page_05.tif
7e00f6cab6b32ff100153f768ac8a382
b05fc79a22865160f822d56ccaf13f4fb40167ed
56330 F20101106_AABWAY dondero_m_Page_20.pro
fe44513b2d527ad29c30032e7eed0b42
3ce8f6a7b0b6180a4847cf426f32b494745ee236
6639 F20101106_AABWFX dondero_m_Page_57thm.jpg
8c8eb298b1c69bad229038126de48621
f39331198ad7effddb1f5bea82204426495c9d5b
4593 F20101106_AABVRE dondero_m_Page_59thm.jpg
6cdcc1ccc6cac54f97d832cea309274e
9446e9520a6c72ed10a7364cba6e5c2d6e77a7bf
1051909 F20101106_AABVMG dondero_m_Page_17.jp2
2ce6b17e19c952c4f21d8eb8ce5f5e9d
9c253f8be900effeedb8695213098cdeddca5d2b
81771 F20101106_AABVWB dondero_m_Page_10.jpg
16820d0799b4ac0f8fdcd7d6e23a4ba4
2eb110afba15e1e7a95f404bf15b5fc88d16c69e
F20101106_AABVYZ dondero_m_Page_06.tif
217296ef682080fe851579111cd9f06a
5f98fd52ad1162758e21e0bd229cb34927255c55
57079 F20101106_AABWAZ dondero_m_Page_22.pro
6d6204cae400a1a3568d2c40bcd43101
c8de7c57d9fd4469a10f74bff5dd8ff07bf54832
4437 F20101106_AABWFY dondero_m_Page_58thm.jpg
3a59ae8a122d92300a850d86ec828a02
8dbbf8eff42c0a0cc3b1a0c3d568cc49c84dc3c2
28712 F20101106_AABVMH dondero_m_Page_28.QC.jpg
1c0b05592d57b44b7a58f2132516148f
89c5f532a81e96d5c1add0fc5ea4ecb4cd06496a
16716 F20101106_AABVWC dondero_m_Page_11.jpg
a58c74190440ae09bdca0a7f3365601c
faef405bf9e8f35f1477630855276630cbbadc1e
532751 F20101106_AABVRF dondero_m_Page_07.jp2
1a33d35230909a016a32b4972908dedb
accb1a97600ff2e58e8bb96960d0173385b25814
17157 F20101106_AABWFZ dondero_m_Page_60.QC.jpg
65c5ff591044da9a6f36880eb3c64399
e0c6b2fd7f4f84160c2bfc3ca4a6f9b5b9fbb3c3
2096 F20101106_AABWDA dondero_m_Page_51.txt
b69019826abdcca1b037f84dbd5840be
a1e95bcedcb0c8399b2b9112cc35d0c5577e07f1
35067 F20101106_AABVMI dondero_m_Page_71.pro
2d743a5b6ad3e0a0b1879fca64513758
94f120a84c806d4ba1eb54ba30970a1490afd55c
85971 F20101106_AABVWD dondero_m_Page_12.jpg
870da6278e5213946c46d3471218d9f2
84926ba92e4e294167402e7833cbd8cdf44dfd37
F20101106_AABVRG dondero_m_Page_52.tif
0c7dd458daa0b7d3723b4bed6f6613de
46b0c5afafdfebba47e1b14d006b09eb66599036
2298 F20101106_AABWDB dondero_m_Page_52.txt
1e32ad58829222cb2f9aace37b1cf8f1
00b795024cf210f99ccfc03a5d985c9097c2eb5f
53946 F20101106_AABVMJ dondero_m_Page_26.pro
924e39d9a2726b948602330a4c05bcd4
46636a39f1b0fb055bf311ba79fa4009018f0629
20438 F20101106_AABVWE dondero_m_Page_16.jpg
045bd71e9f23734769419d4b28afaa6e
f5ee17e2733b9540f1e9d3ccfb7b01bf51bc83a2
F20101106_AABWDC dondero_m_Page_54.txt
7756b6b7243fa7759a060d16d3fdb54f
f55e2288951cacd0837a1ef64e2965d094a3402e
2136 F20101106_AABVMK dondero_m_Page_12.txt
24b33824e3228960e34e4549b21622f0
b0efc027e378c20f2207f44cde9d5eb9c89f4c27
81724 F20101106_AABVWF dondero_m_Page_17.jpg
0f5e82b082473a892aecc283d7fbe51c
e3290c0518044b983973a4dd6d02323afedbee38
F20101106_AABVRH dondero_m_Page_49.tif
729b556bc2b81409d082f80fa4ab5e58
5d9f3554c7a28f78df2196be45d21b027af6ef90
2295 F20101106_AABWDD dondero_m_Page_56.txt
14cd9014030158a477be270bfe122a00
7bb14639b87ca60a64b13fc020a97632d787ce78
1603 F20101106_AABVML dondero_m_Page_58.txt
d240c67efd6183e7dc259308b10591ee
9969b5958d8fe4f577375f70524c867f29742ed5
92689 F20101106_AABVWG dondero_m_Page_20.jpg
33eb7024e5fe2cd34fa2b8f331f90939
a9b2e6b7b31d59f2582b3043f5e10078af4e17c9
F20101106_AABVRI dondero_m_Page_15.tif
c117bf4ce0c4e8738c2ae6c9f763dbeb
9b7861a7008ce66d11fd3b55bf5cbe20a9dc2257
1871 F20101106_AABWDE dondero_m_Page_60.txt
f52d08a3a6c13e5be9202f5c71962f11
445c336d21c53df6ff8baf7d8dc15bb2ea032c16
28779 F20101106_AABVMM dondero_m_Page_20.QC.jpg
2c6b7815f49448f3e1211cc5627f24ce
a3a9d1111ccad7d2281a9bfc27b42208a17e334f
88683 F20101106_AABVWH dondero_m_Page_25.jpg
f90229a03230f0d2b0250a81fb793ea9
0718cf54d602b9db05c36cc0a55037a23d920441
1051924 F20101106_AABVRJ dondero_m_Page_63.jp2
679ee01a192721c1d1721ba1c8da3d79
2108fdfc17cd3aea17967cd0ed07d1cf7f3329a5
2129 F20101106_AABWDF dondero_m_Page_63.txt
f1f2b24014b888e0fecac993bd071f4d
4ab50e3ad28399aec148873595001b2303b6c367
1171 F20101106_AABVMN dondero_m_Page_03.pro
086de1362c73671ab19ab0730ff36551
a8f735bc051fbfbf7472461f58848277309de37d
89562 F20101106_AABVWI dondero_m_Page_27.jpg
27522abc0d94a77a8f4c7cca3a26dc04
1a5219ad9d1475dc8c1141cbdd1f080134c26dea
28566 F20101106_AABVRK dondero_m_Page_23.QC.jpg
a7cdb191480a3d10ddca43ff45a00b4f
afc1c10c397eb09c8b65744beb365a8849b53d29
2207 F20101106_AABWDG dondero_m_Page_64.txt
7a24370aaef370acf6928f0733aa3719
6841d09387470b13f0e60d883038f13b2c2c7312
19068 F20101106_AABVMO dondero_m_Page_71.QC.jpg
110e289f07e4c70af003472da5b95116
e87409553d5597ff49d89c8ace6fac2df309d1e4
91031 F20101106_AABVWJ dondero_m_Page_28.jpg
18e00e1122020e25f6197aa5f01092fc
7db7dbfbd75bbc9dc4c17e684aba38e05b0a7139
F20101106_AABVRL dondero_m_Page_03.tif
329400f9b8dfe8bfb7744ff643faea38
c069c088dd490eeee3621f1744a923b82e34290a
2141 F20101106_AABWDH dondero_m_Page_65.txt
6af54380a7967cd1a845fdd0b5bb9421
b9dd4e2ab140fc069c9da43da87a32ef5ce23e0d
4243 F20101106_AABVMP dondero_m_Page_03.jpg
f79171c484bf54d4f65ce7e5b85b158e
175172e2a720a47bd13b2e341289392fc12cffb2
98735 F20101106_AABVWK dondero_m_Page_29.jpg
8e5d5b255cfa412601db288c31890914
febd67971faed711444fe3d5a1c9780d078ddfd6
2098 F20101106_AABVRM dondero_m_Page_68.txt
6babee5c19d73b8e2352f14cc586a7ca
40f83d89835e64377cc0548782f23d5248d169fd
2067 F20101106_AABWDI dondero_m_Page_66.txt
d8a2d9c35a6f73710912d403e3946913
66fbaf084f462018480f031387969411c952fa53
53483 F20101106_AABVMQ dondero_m_Page_21.pro
ed54f129bdf4e5603acbbadfdb030ee6
6433d6c3b96e6c81723eb7230737ee25d7b138b5
91071 F20101106_AABVWL dondero_m_Page_30.jpg
05afb8b0625ea7b8f4acae9e63844916
ec4fe77bd74e8857fa39d1e0b9011760897851bc
91438 F20101106_AABVRN dondero_m_Page_13.jpg
15dfbaa3fd3949e3eb6d012258fb14d5
34b1fb8a264696e8967283419aeeea09717f66c7
2171 F20101106_AABWDJ dondero_m_Page_70.txt
048dd329408d5e9192201a18165ea752
26a56e8dd730ddcec8268187523982a9cd3fd6f1
1051911 F20101106_AABVMR dondero_m_Page_37.jp2
b01266f926570cb65bc0bfa8f5981641
aa0dfc74630ad5f6d01d468eb11753200fcd80c9
1339 F20101106_AABVRO dondero_m_Page_04.txt
f4b121e4f7f37372c78b397db349f6e8
1fd10c70bd937e33638faf269458f5bba1c06341
2393 F20101106_AABWDK dondero_m_Page_72.txt
0448aaac3e642673df2f0bbba36a20a3
87c3543fb6cc8a9d20e383a736dce6ce49e5caba
F20101106_AABVMS dondero_m_Page_07.tif
648f8155d75e5e46ac159e7da9e014c3
dc50453c2693bd1a02db621df18d9820b508409d
90892 F20101106_AABVWM dondero_m_Page_31.jpg
87d3cffccf877be7ba33fd271b318f3e
9ca7aab1986574fa6048a5b0d65850a7daa71ee0
1523 F20101106_AABVRP dondero_m_Page_01thm.jpg
8d5c357b470527e215a09f9e60ccefcc
e899f628931580cc190bc618d56d6318bda36639
2093 F20101106_AABWDL dondero_m_Page_76.txt
bc60b579d5650299d89891d83eeaaa6e
da99cc0d8277269985dfa6cf852910dd2d2add38
F20101106_AABVMT dondero_m_Page_35.tif
0e0bfc0f2b55ea53e878c837ad0442f6
30ccddfa263c2fcda1599b6a2e7c8b7be7c2a0d4
18837 F20101106_AABVWN dondero_m_Page_32.jpg
7124c2b6583e747d92668dcb591404c2
c97cf22f9692d23369c69d2d44cb492c9a57e1e9
10837 F20101106_AABVRQ dondero_m_Page_09.jpg
f0920b51d889746838a1976f2b9ed949
7c1d44dc8c03b19ba48e4f701738ec54a65493d9
423178 F20101106_AABWDM dondero_m.pdf
3b31c2ec072bcfff7dfc2f7d2a3052dc
835cee8e008e0dee4b84e2919238d5e3a91229b5
26073 F20101106_AABVMU dondero_m_Page_45.QC.jpg
2ca6fd155373f5ab6d3492e37f4f12c4
7ba3890570719c9260f8068b7ebb1ec333eb38ea
71262 F20101106_AABVWO dondero_m_Page_33.jpg
cec0ba0bb083898d7dd717890c3b3667
dfa5d986450cc98c8a7faecd040425db1b2a8eec
29636 F20101106_AABVRR dondero_m_Page_22.QC.jpg
6d2aabab6f1bb11fdcfa18369357dc7d
02817d7e3bb154d7c9f4cab7877cff13c858c37c
6240 F20101106_AABWDN dondero_m_Page_01.QC.jpg
df7da3b671d8258297b14db78709760e
6f332c4075a828c0de7c6da97d7b15d97fc23e02
F20101106_AABVMV dondero_m_Page_73.tif
86854e1837dcf64ccae2f874fa5bc9a2
d0f70510a1f13132c0a0a2557cd9cace82b2d7f9
89283 F20101106_AABVWP dondero_m_Page_34.jpg
edfdd7a68fb0ae7b0db82e9bd3c42125
709ea440dab708ac0ea20d8330ff014d288b1a06
60296 F20101106_AABVRS dondero_m_Page_29.pro
9caea97557504a01a0e3512e247319df
86539fc7fded1920c6c12ae29a55f060ac7c72d5
1022 F20101106_AABWDO dondero_m_Page_02.QC.jpg
19df321ecd5e506acc8e8373f7ce6527
db49d986ccecaced22b3ac1bdeb96b2472aac8dd
26638 F20101106_AABVMW dondero_m_Page_44.QC.jpg
9083798640e7c0ec6896597016729e12
de6c948249f0ee5604c61fdb527d7bc4df48baa9
90138 F20101106_AABVWQ dondero_m_Page_35.jpg
1490d35073e09482d149cc43ea81212d
adefde74ac96978dd5e1c81b6508b0c4e4b60e7e
2268 F20101106_AABVRT dondero_m_Page_22.txt
8af42b38eadc497b4888b2366b3e52e8
39882690ae389d69e84a8252c00f96b6d99e2c16
459 F20101106_AABWDP dondero_m_Page_02thm.jpg
8d773573b76fd0784c14c3cf7ef64e47
b2b50f446d1a92167a21aa8f7b4fdb8f9cc96cd1
27585 F20101106_AABVMX dondero_m_Page_65.QC.jpg
2143f44ae4666af311ca05bb932d4c47
751b2f7a21810c39f9b6bb3f9d7e96f576be5d07
90414 F20101106_AABVWR dondero_m_Page_36.jpg
e62ad5a38ad66575732e371bc1380fde
9a9c9a88284cc2ab892b65f30a89a6fecaaa5be7
2602 F20101106_AABVRU dondero_m_Page_08.QC.jpg
46f7917932b9a1ff0a14dd5ab6621472
0c14e5a0cd4ee7d4653a493244d2492441314ef2
1201 F20101106_AABWDQ dondero_m_Page_03.QC.jpg
8c17beb539ba73f8dddc494c04e1b710
1bd3f8f89c9c74b9d442c517c5c722a298899109
F20101106_AABVMY dondero_m_Page_75.jp2
ca109f5a58417f3ee66d0683662de5e3
a5ca239c5543778e0dd4728185fdf626757785e1
86066 F20101106_AABVWS dondero_m_Page_37.jpg
4f3521cb476f1c94403dd08fb0dbc733
e23ae5c94c7b337e1fc4100cd21d05b6948e4096
6636 F20101106_AABVRV dondero_m_Page_25thm.jpg
827545c676a9a8febc398fbaa2b5f886
31141b8d7bff067ea3411028bdaac6383465d0c2
17702 F20101106_AABWDR dondero_m_Page_04.QC.jpg
56538039e4bca4bbfb447b3e74a1be6f
c7e90876d0e9fab1871450534136b610e98b05bd
6883 F20101106_AABVMZ dondero_m_Page_30thm.jpg
aa051732296b69aa0b486cad7895820e
e740bf2c97ad480dd6dd5a0c35cb010c6c2b1c2c
88398 F20101106_AABVWT dondero_m_Page_39.jpg
cb4765a7ac066449e88af20d87ad510f
aa1bc5b56991bc88857783dbefb23c4add228b9b
54404 F20101106_AABVRW dondero_m_Page_35.pro
72931390afbbdf25c70066a044f7ed5c
369ac759e63ce3f7a0ebebd90e331c7bbc5c39bc
18327 F20101106_AABWDS dondero_m_Page_05.QC.jpg
c44481edabf9415f85296e3640816cd0
3e58ee6b95b6a0aa74eb3dfbdfe1201340d475b8
86502 F20101106_AABVWU dondero_m_Page_41.jpg
7aa02e9d63b31d7e6c6e2d9724a3d552
eb99510e365bf4d5f36e7c5e6ebf884a7d6d4c4a
1033 F20101106_AABVRX dondero_m_Page_08thm.jpg
8060c40b3769492986fc26dfb3b2e69a
2141bcd31cdbf48002708a56e9e2b75a07fa07c4
4689 F20101106_AABWDT dondero_m_Page_05thm.jpg
6eae3e03c07e79777f2b88607ad4fd85
db108dda31bb1a0d4b242ad8f364bf7227fa918d
85186 F20101106_AABVWV dondero_m_Page_45.jpg
5b04de691b48af599852f5f6fd646607
a8de0a2ed9dc4fd7916fc83cd56fe58d2fe48c82
2123 F20101106_AABVPA dondero_m_Page_39.txt
57d21a30e38364c6072be3d0425e3957
2e68242893d11177bfc2615c2514a421e6927d05
87437 F20101106_AABVRY dondero_m_Page_21.jpg
f412f12bfe13a8ef11e2da5bba774086
e4aaafcb566577a8a99a9dc3a0ac57616195341e
10596 F20101106_AABWDU dondero_m_Page_06.QC.jpg
0382efd42801cd78f1227f3d8246cfea
c6d4646ed0a89530bffc66a7bccd70f185f814da
52011 F20101106_AABVWW dondero_m_Page_46.jpg
cb092a38e8e5c3c5ae15d78ce3f368b8
3c1371a832532f5f2e00ef8560185d4049dbc94c
F20101106_AABVPB dondero_m_Page_30.tif
b619b04018c40629cbf2942ac7c86390
96eb5f42b2340a58456c47cda58711f9d15020bb
F20101106_AABVRZ dondero_m_Page_26.jp2
15a763a1a9cb78d9496d679d60dbe1ae
32a93406d63947dd263baeed294c889978fb7469
2857 F20101106_AABWDV dondero_m_Page_06thm.jpg
5ca98aa0885a3ef3813e18f3337b6974
f93629d10951d33a58bf308850ac38ebfe0bcd72
78612 F20101106_AABVWX dondero_m_Page_49.jpg
9eaee57d532fd7b98c02421e1589be13
d880f638d7447b479a63313e5b349b9b2307ed2e
F20101106_AABVPC dondero_m_Page_62.tif
7341eab3c583b1ebe656710f2b3e785f
8e7b30dbfce8fd74e537a66ed4020009a071cc4d
7447 F20101106_AABWDW dondero_m_Page_07.QC.jpg
464f97c10d43a6254cf7e97d97131222
9df23bb35b4a2ebbe20c6660ce2bc87985b8bbcc
1051953 F20101106_AABVUA dondero_m_Page_36.jp2
ae41d475a45becf75d672dc0e4a5322b
aa7758bd60012c388d28027b47cb11044788a297
90950 F20101106_AABVWY dondero_m_Page_50.jpg
97df4951fb3e13aa71aa559cbb25bc64
d29586ed4e194a286eccfcff24699c513532627b
F20101106_AABVPD dondero_m_Page_05.jp2
c7be191c325dc2062794c3b304e42c59
d660dcec2e6426df9f74185fb786456121f3d10b
2066 F20101106_AABWDX dondero_m_Page_07thm.jpg
2413db920c5a844f1e2d834b1c3282e8
1d7964dd1ea64b2cad14460dbcf0f11b0f17d89e
86637 F20101106_AABVWZ dondero_m_Page_54.jpg
642a496d00009d1098c4894e9d887183
dcc9b570aa685fcbb3a8b536a234661efb322443
F20101106_AABVPE dondero_m_Page_41.txt
c3bbb162b5637d62a3d891effa05b303
f4acc380bd5db1e35dc7b418cae948bbfc61b6d6
27824 F20101106_AABVUB dondero_m_Page_63.QC.jpg
713507b97c85017321bba8fed33e19bc
c000b04b6ba2d37cd4ceab57a1366c964f42f249
84086 F20101106_AABVUC dondero_m_Page_44.jpg
70fb2c4e73abb43d3d56494e4960635e
8a0476840f7a2d508a69a96ff42e3d9a3b006b95
24962 F20101106_AABWDY dondero_m_Page_10.QC.jpg
750784f1ffeaf92489819868881e355d
79801ceec577f66c0a7962da655d642f38b551f3
F20101106_AABVPF dondero_m_Page_19.jp2
9311a55791e706d1d1e2c17a9dbad3a8
71b96159e17849c751c04e712fbb95e55489064e
F20101106_AABVZA dondero_m_Page_08.tif
31f60f4e47d1bfcdf9e61ac67027df59
2c08a8bfdb8886350cfb4891599ca3577078090e
53960 F20101106_AABWBA dondero_m_Page_25.pro
39d5d8422c91261e937c8018c035372d
3edae4b7fb4d348affcf36cfc0458ab872eb58ac
440 F20101106_AABVUD dondero_m_Page_01.txt
40ea13f5c2fd71e316a7db9415e68716
25814888e0b0d523d0bdc8ad63acc2f34bb2ce82
5814 F20101106_AABWDZ dondero_m_Page_10thm.jpg
be1d0e8d9efff77702f19b85e9edf99d
74e98d8cc10612951dc5c51bbcfe420d975b3e3b
8425 F20101106_AABVPG dondero_m_Page_08.jpg
8644a73e54478877d02933366502dc22
6b354ece41ec2e3bee8653cd685e34fb5cdebd77
F20101106_AABVZB dondero_m_Page_09.tif
6b5c6bbecfa90712080b2101e24950b6
3142ad10438089adc7e17ef5ec3e7fbb27df0c14
55710 F20101106_AABWBB dondero_m_Page_31.pro
c803994043307aebff331578a6871940
5ab51faaa88f58d09d71a523df719ab1e597b19c
17142 F20101106_AABVUE dondero_m_Page_59.QC.jpg
3289d9f2fb77065485b7ecd009e4a648
9b836bfa238fe04cd9481633b5a62284faa556ba