CALIFORNIA SOCIOLOGY JOURNAL, 2011, VOL. 3 (ISSUE 1: 12-32) Residential Segregation Jesus Campos Department of Sociology California State University, Los Angeles This study explores the influences of socioeconomic factors on residential integration. Data was collected through an internet questionnaire and respondents were gathered by snowball technique. Residential integration is measured by the percentage of White individuals living in the respondent's specific ZIP Code based off information from the U.S. Census Bureau. The author proposes that an increase in socioeconomic status will lead to an increase in residential integration. Despite having mixed results, a significant positive correlation was found between one's level of education and residential integration, supporting a part of the author's original hypothesis. INTRODUCTION 1Studying the residential patterns of the various peoples within the United States will almost certainly lead the reader to a well known set of facts: those of us living in this country are not evenly distributed and tend to cluster based on a number of demographic factors. Though this statement may come off as a common sense, , it nonetheless poses a difficult social challenge for the United States: How and/or why does a nation which pronounces itself as diverse and accepting continue to exhibit such a large degree of residential segregation. The implications here mean that U.S. residents may not be free (at least in a Jesus Campos is a graduate student in the Department of Sociology. His paper was originally submitted to SOC. 410 Advanced Statistics, taught by Dr. Hyojoung Kim in Winter 2011. structural sense) to settle down wherever they wish. In a sense, it would appear as if modern U.S. society has become tolerant of those of varying ethnicities/skin colors/outwardly appearances only in its public space (i.e. the workplace; schools; any place sufficiently distant from our own neighborhood), but has yet to conquer the unseen barrier that tends to keep our living spaces distinctly segregated from one another. Perhaps most critical, however, is the issue of whether or not such residential segregation, if continued, will perpetually prevent the populace of this country from achieving true and complete social unity. LITERATURE REVIEW In trying to explain the causes of residential segregation, two prominent theories have arisen. The first, known as RESIDENTIAL SEGREGATION by Jesus Campos 13 CALIFORNIA SOCIOLOGY JOURNAL, 2011, VOL. 3 (ISSUE 1: 12-32) spatial assimilation, assumes that economic factors in of themselves will always produce such uneven distribution, even if we were to presume no discrimination based on physical and/or identity traits. In essence, this theory proposes that, since the free- market model of the United States produces a variation in the costs of housing and living from one area to the next, it is a given consequence that those with more money will be the ones with access to the costlier housing, regardless of other factors. The second theory, known as place stratification, factors in the potential consequences of other non- economic variables, such as race or ethnicity. Even factors such as religion or nationality, not readily apparent to the naked eye, may serve to create an atmosphere of non-acceptance or outright discrimination, effectively keeping people away from certain residential areas. In this theory, economics tends to lose its importance, since it presumes that two individuals of different traits, even if in the same social class, would likely not settle down in the same residential area (Iceland and Wilkes 2006). The issue is more complex than this, however. As will be further discussed later, research has shown that residential segregation also varies between Whites and various individual race/ethnicities. For example, there appears to be less spatial segregation between Whites and Asians than between Whites and Blacks or even Whites and Latinos (Massey and Fong 1990). Depending on one’s point of view, this is to be attributed either to a greater ethnic acceptance by Whites for Asians than for other races (i.e. place stratification), or to economic differences (caused by educational and occupational differences) between these minorities that has led one minority to achieve a generically higher status (i.e. spatial assimilation). Unfortunately, there is no consensus on what exactly constitutes socioeconomic status. In their attempt to define socioeconomic status, Massey and Denton include a person’s objective level of both wealth and education, as well as the more subjective perceived status of the person by the rest of society (Massey and Denton 1985). These concepts can be further divided into various sub-dimensions. For example, wealth is often tied to income, and perceived status can be seen as a combination of such factors as the extent of social networks and type of occupation. Though we typically view these dimensions as achievable by anyone dedicated enough to pursue them, Massey and Denton point out that U.S. society has been plagued by the unequal achievement of these factors largely as a result of discrimination based on any number of traits, such as race, class, or gender. As such, they conclude that it becomes somewhat necessary to include such ascribed characteristics as race and gender when determining socioeconomic status, as such factors can have considerable impact on patterns of spatial integration. (Massey and Denton, 1985). Historically speaking, within the United States, the dominant WASP (White Anglo Saxon Protestant) group has constantly expected minority groups to eventually assimilate into the dominant group (Hirsch 1942; Woolston, 1945; Gordon; 1964). The RESIDENTIAL SEGREGATION by Jesus Campos 14 CALIFORNIA SOCIOLOGY JOURNAL, 2011, VOL. 3 (ISSUE 1: 12-32) desire to one day establish a homogenous national culture, albeit reflecting the traits of the dominant group, is considered to be the driving force behind such an overwhelming expectation. The residential aspect of the assimilation processes reflects the tendency of a minority group to integrate proximally (i.e. in terms of distance) with the dominant group. This becomes difficult, however, as several decades and generations of systematic ethnic and racial segregation has created a strong barrier to such spatial integration, creating somewhat of a paradox (the dominant group expects minorities to assimilate, yet still discriminate against and resist any encroaches by minority members). (Massey and Denton 1988) From the perspective of the minority individual, Fong and Wilkes identified several factors that minorities consider when determining ideal neighborhoods: health; crime rates; quality of local schools; the perceived socioeconomic status of the area; and, quite interestingly, level of interaction with the dominant group (Fong and Wilkes 1999). Although all factors are important, the basis of residential integration seems to imply that the perceived status (i.e. social class) of the area in question would be the main factor for where minorities choose to settle down and reside in, since this would have a direct effect on the degree of interaction and integration with the dominant Anglo population (Mullan and Massey, 1984). As such, Massey and Denton were actually able to record that increases in the social status of minorities (as a result of choice of neighborhood) often resulted in an increased likelihood of contact with the dominant group (Massey and Denton 1985; Beshers, Lauman, & Bradshaw, 1964). This leads Gordon to simply declare residential integration as a means for minorities to increase their socioeconomic status (Gordon 1961). This does not mean, however, that all minority group members choose to move into neighborhoods consisting mostly of dominant group members. Even within the segregated communities of the minority groups there exist hierarchies which allow local minority residents to achieve greater wealth and status relative to those around them. Some residents may prefer to buy homes within their segregated community than to rent homes outside of them in order to heighten their status within these minority hierarchies (Fong and Shibuya, 2000). Nevertheless, minorities are still aware that the areas outside of their somewhat isolated neighborhood hierarchies provide an opportunity for even greater social ascension, with the well-documented difference in the quality of schools between white and minority neighborhoods being a prime example (Massey and Denton 1988). In this regard, attempts by minorities to integrate into Anglo-dominant neighborhoods may be seen as nothing more than rational decisions made to improve their lives and that of their families. Though many studies have hinted at a positive connection between socioeconomic status and degree of residential segregation, race seems to produce variation among the patterns of residential integration (Fong 1965; Massey and Fong 1990; Massey and Bitterman 1985). According to Massey and Fong, Asians appear to be the RESIDENTIAL SEGREGATION by Jesus Campos 15 CALIFORNIA SOCIOLOGY JOURNAL, 2011, VOL. 3 (ISSUE 1: 12-32) quickest in residentially integrating into dominant Anglo communities, followed by Latinos. Blacks have the lowest degree of integration, though in no way does this mean they are incapable of reaching a high level of socioeconomic status (Massey and Fong 1990). Instead, the implication here may revolve around the level of acceptance by the Anglo community towards each minority group. Logan, Alba, and Leung also observed that dominant Anglo acceptance (in regards to residential integration) also vary by region (Logan, Alba, and Leung 1996). Consequently, though we may presume that higher socioeconomic status should lead to increased opportunities for residential mobility, it must be realized that the perceived status of any minority may be arbitrary to the dominant Anglo group; it is they who have the power to restrict residential integration to minority group members (with the White flight phenomenon serving as an ideal example). The question remains as to which of these two leading theories is more accurate? Does one trump the other, or is it possible that both theories can coexist side by side? If ascribed factors such as race and gender are to be included into socioeconomic status, then this would seemingly eliminate the need to measure residential segregation based on these factors alone (i.e. place stratification). However, the possibility exists that a measure of using socioeconomic status (including both ascribed and achieved features) may very well yield a different set of results than a measure using ascribed features alone. HYPOTHESIS I propose that an increase in socioeconomic status will lead to increased residential integration. My ultimate intention is to examine the proposed causes behind a person’s existing social class and their degree of residential integration. I’d like to determine if higher levels of a minority individual’s socioeconomic dimensions (i.e. education and wealth) are in fact correlated to their residential placement within society. Second, I would like to analyze the effect of secondary independent variables on residential integration: • Sex, an ascribed visible feature, is viewed by place stratification theory as having an impact on residential placement (though marital status may have a greater role than sex alone). I hypothesize that minority males are more likely than minority females to reside in areas that are predominantly white. • Race/ethnicity, being the most prominent ascribed feature, ultimately determines (according to place stratification) residential placement as a result of dominant group acceptance or rejection. I hypothesize that non-Whites respondents are more likely to reside in areas that are not predominantly White. • Religion, though not easily discernible, also has the potential to affect patterns of integration. (i.e. existence of predominantly Jewish neighborhoods; Catholic RESIDENTIAL SEGREGATION by Jesus Campos 16 CALIFORNIA SOCIOLOGY JOURNAL, 2011, VOL. 3 (ISSUE 1: 12-32) and Protestant divides). I hypothesize that non-Christians and non-Jews are more likely to find themselves in areas that are less predominantly white than those who are Christians or Jews. • Legal status, such as whether the respondent is a U.S. citizen, resident, or other perhaps even undocumented, may play a key role amongst certain groups, particularly Latinos and Asians, regarding their willingness to integrate into dominant Anglo communities. I hypothesize that as legal status increases (using an ordinal scale to be described later) so does residential integration into predominantly white neighborhoods. • Generation, though not applicable in all cases, may affect residential patterns. Some respondents of certain origin (i.e. Latinos and Asians) may be so far removed from their immigrant ancestors that they no longer see their racial or ethnic minority status as an impediment towards integration (i.e. they may feel sufficiently comfortable around dominant Anglo groups, or perhaps are fully assimilated and are now Anglicized). I hypothesize that for those respondents whose families come from immigrant backgrounds, the more generations the family has resided in the United States (using an ordinal measurement to be described later) the higher the percentage of Whites living in the respondent’s neighborhood. • Marital status may affect a person’s willingness to residentially integrate? For example, a single minority woman may become more willing to integrate into a dominant Anglo neighborhood once married. I hypothesize that married minority women are more likely to live in areas with a larger percentage of Whites than unmarried minority women. • Home ownership may potentially add to a person’s status. Furthermore, individuals who are seeking to purchase a home may tend to seek out neighborhoods based much more so on the demographic characteristics of the area than those who are seeking to rent or lease (presumably since renting or leasing is often viewed as temporary). • Job type may play a role in determining one’s socioeconomic status, as certain different societies attach greater prestige to some occupations than they do to others. SURVEY PROCEDURES All members of a focus group completed the survey in about five minutes. All were satisfied with the instructions within the survey, as well as with the overall appearance of the questionnaire. However, some members RESIDENTIAL SEGREGATION by Jesus Campos 17 CALIFORNIA SOCIOLOGY JOURNAL, 2011, VOL. 3 (ISSUE 1: 12-32) raised concerns about the response categories to some of the questions. For example, the questions measuring income and savings were originally open-ended; the respondent was instructed to write a numerical figure to the nearest thousand. However, the focus group felt that this could be rather difficult for the respondent to calculate in the course of a few minutes. As such, I decided to introduce an ordinal scale in intervals of ten thousand dollars. The focus group also raised concerns in the distinction of the response options between two questions: one which asked for race and the other for ethnicity. The race question originally allowed a respondent to choose between White (including White Hispanic); Black (including Black Hispanic); Asian (including Indian and Pacific Islander); and Native American. The group felt the options were too vague, and perhaps even insulting. In expanding the options, the resulting list was almost identical to the existing options already in place for the question regarding ethnicity. As such, I decided to combine race and ethnicity into a single question asking for the respondent’s race/ethnicity (though this has forced me to consolidate two separate secondary independent variables into one). Lastly, the group raised concerns about determining socioeconomic status. For example, the group raised the possibility of how home ownership can affect perceived status, persuading me to include such a question. The group also assisted me in expanding the options for the question regarding occupation, and also assisted in determining perceived various occupation options listed in the questionnaire. An online questionnaire, posted on the online survey site zoomerang.com, was utilized to measure the variables. Posting the questionnaire online allows the researcher to potentially reach a large number of people for a minimal amount of effort (as opposed to telephone surveys, which can be time-consuming). Furthermore, online surveys are comparable to mailed surveys in regards to minimizing any potential social desirability effects. Lastly, online survey sites allow a researcher to collect and analyze results rather quickly and efficiently. Zoomerang, the website used to create and post this questionnaire, gives users a fairly large number of options in creating a web-based survey (including survey templates, response types, etc). The website then creates a specific internet address for the survey, allowing respondents to reach the survey by either imputing the address or linking to it. Zoomerang considers a complete questionnaire as one in which respondents have clicked on the DONE button located at the bottom of the questionnaire page. Thus complete questionnaires can include those in which not every question was answered. Zoomerang also keeps track of what are referred to as views, which include both the number of surveys marked as complete (as previously defined) as well as those surveys that were opened but subsequently closed without clicking the DONE icon. Such ‘incomplete’ visits are not included in the tally of complete questionnaires. The site does not allow one to distinguish between those incomplete surveys in which a person simply viewed the survey and decided not to respond, and those in which a RESIDENTIAL SEGREGATION by Jesus Campos 18 CALIFORNIA SOCIOLOGY JOURNAL, 2011, VOL. 3 (ISSUE 1: 12-32) survey was completed but the respondent did not click DONE (for whatever reason) or accidently closed the page. It is also impossible to ascertain whether or not some respondents who simply ‘viewed’ the questionnaire the first time came back to complete one later on. Since the Basic survey package offered by Zoomerang was utilized, a maximum of 100 complete surveys were allowed. For the purposes of this survey, the site was instructed to close the survey once this limit was reached. This survey was completed by 100 people (the maximum allowed), and was viewed by 126. Thus, on 26 occasions the survey was opened but then closed without clicking the DONE icon. Though the questionnaire was posted online, the researcher is still responsible for finding those he or she wishes to survey and directing them to the questionnaire page online. As such, this particular survey was undertaken using the snowballing technique, in which roughly forty people, including friends, family, and classmates, were originally asked to participate in the survey, and were also asked to spread the word to others in their social network, with the only limitation being that an individual must be an adult and could only take the survey once. The 100 completed survey mark was reached roughly over two weeks after the survey was activated online. Since the majority of those originally asked to participate are non-Whites and tend to be of college age, this may arguably mirror itself in the final results at least in terms of age and race/ethnicity. MEASUREMENT OF VARIABLES Residential integration is to be the dependent variable. To determine the level of integration for each respondent, the percentage of Whites living within the respondent’s ZIP Code (given by the respondent in the questionnaire) will be obtained from the U.S. Census Bureau’s official website: http://factfinder.census.gov/home/saff/m ain.html?_lang=en. The data from the website is from the 2000 Census. The primary independent variable is socioeconomic status, to be measured amongst the following three separate components: 1) Income- Measured by asking respondents their level of income using an ordinal scale with ten thousand dollar intervals; 2) Savings- Measured using the same ordinal scale as for the question regarding income; 3) Level of Education- Measured using an ordinal scale consisting of the various educational degrees typically offered in the United States. In addition to these three measures of socioeconomic status, I will ask the participants to also list the following secondary independent variables: 1) sex- obtained from a simple dichotomous male/female option; http://factfinder.census.gov/home/saff/main.html?_lang=en http://factfinder.census.gov/home/saff/main.html?_lang=en RESIDENTIAL SEGREGATION by Jesus Campos 19 CALIFORNIA SOCIOLOGY JOURNAL, 2011, VOL. 3 (ISSUE 1: 12-32) Residential Integration Socioeconomic Status (achieved traits only) Sex Race/Ethnicity Religion Legal Status Generation Marital Status Home Ownership Job Type (both achieved and ascribed traits) 2) race/ethnicity- obtained from a nominal-level list consisting of the more common race/ethnic backgrounds found within the United States; 3) religion- obtained using another nominal list consisting of various religious backgrounds; 4) legal status- determined by an ordinal list of possible options in descending order; 5) generation- obtained by defining first generation, and then asking if respondent to mark which generation they belong to; 6) marital status- obtained from a list consisting of various relationship possibilities; 7) home ownership- obtained from a dichotomous list of either rent/lease or own; 8) job type- obtained by asking respondents to choose their occupation from a nominal list; a separate question, Employment, will ask respondents if they are temporarily employed or directly employed. Though ascribed factors such as race and sex can be construed to some as being a determinant of socioeconomic status, I will keep them separate for the purpose of this piece. Thus, since I will define socioeconomic status using only achieved factors, the distinction between the spatial assimilation and place stratification theories will be intact (the ascribed secondary independent variables can be used to test the latter). The following diagram is fairly representative of the correlations I intend to do: DATA CLEANING In total, 14 distinct variables were measured throughout the course of the questionnaire. However, many of these variables were later recoded, and some were eliminated altogether. All nominal sets of response categories consisting of more than two possible options were recoded into various dichotomous variables. However, original response categories that did not illicit a single response were dropped entirely from any future analysis. Table 1 in Appendix II is a summary of the cleaning and recoding process and includes all the new dichotomous variables (with the exception of the reference groups, though they too have been recoded). Table 1 also summarizes the mean and standard deviation of all listed variables. The nominal Marital Status variable originally consisted of three separate response categories. However, since one of the original response RESIDENTIAL SEGREGATION by Jesus Campos 20 CALIFORNIA SOCIOLOGY JOURNAL, 2011, VOL. 3 (ISSUE 1: 12-32) categories did not illicit a single response, the remaining two categories have automatically become a dichotomous variable. The values for these two categories were simply changed to 0 and 1 to match all other dichotomous variables. Similarly, the variable Generation, which originally consisted of four different response options, did not receive a single response for the category 3rd Generation. This response category, therefore, was completely eliminated (with the value numbers changed to remain consecutive), and the remaining categories recoded into dichotomous variables. A variable labeled Job Type Status was to be included into the final analyses. This variable was to be determined by a designation of Lower Status and Higher Status based upon the respondent’s occupation as given for the variable Job Type. However, since the variable Job Type contained 18 missing answers, the idea was abandoned. The actual occupation options for the variable Job Type will be used in analysis instead (they will be recoded into dichotomous variables). Furthermore, a variable entitled Employment Type was simply eliminated, as its analytical justification was difficult to establish. UNIVARIATE ANALYSIS The dependent variable, consisting simply of the percentage of Whites living within the respondent’s ZIP Code, yielded a mean value of 39.36%, with a standard deviation of 24.57 %. Level of Education produced a mode response of a Bachelor’s Degree; Level of Income produced a mode response of $30,000-$40,000; and Less than $10,000 was the mode response for Level of Savings. Table 2 in Appendix II summarizes the response frequencies for the dependent and key independent variables. The average age of all respondents was 27.8 years with a standard deviation of 5.6 years. Of the 100 total respondents, 47 are females and 53 are males. Most of the respondents (44) are Roman Catholic and most (64) are Latino/Hispanic. Furthermore, 64 of the 100 respondents are not married and 80 either rent or lease (but do not own) their housing. Another vast majority of respondents (77) were born in the United States, and 64 respondents come from immigrant families and are the first generation to be born in this country. In terms of occupation types, the Administrative option had the most responses with 28. BIVARIATE ANALYSIS (Refer to table in Appendix III for all Pearson’s values obtained from the bivariate analyses) In comparing the key dependent variable (utilizing White Percentage of ZIP Code) with the variable Level of Education (the first of three key independent variables), a Pearson’s value of -.561 (significant at the .01 level) is obtained. This is not what was expected, as an interpretation of this value means that as the level of education a respondent has obtained increases, the average White percentage of the respondent’s neighborhood (i.e. the area encompassed by their ZIP Code) RESIDENTIAL SEGREGATION by Jesus Campos 21 CALIFORNIA SOCIOLOGY JOURNAL, 2011, VOL. 3 (ISSUE 1: 12-32) decreases. The relationship between the White percentage of the respondent’s neighborhood and their level of income did not produce a statistically significant value. However, there does appear to be some association between White percentage of ZIP Code and Level of Savings. A Pearson’s value of .237 (significant at the .05 level) was obtained in the bivariate analysis of the two variables. This implies a positive association, meaning that the average percent of Whites living in the respondent’s neighborhood increases alongside one’s amount of savings. Other factors that may potentially influence the dependent (i.e. the control variables) include one’s race/ethnicity, religion, legal status, job type, immigrant generation (if applicable) marital status, home ownership, sex, and age. The Pearson’s values obtained from comparing the various non-White race/ethnicities with the White percentage of the neighborhood are somewhat mixed. There is a significant value (at the .01 significance level) between Latino/Hispanic and White Percentage of ZIP Code of -.275, meaning that the average White percentage of the neighborhood decreases when moving from non- Latino/Hispanic to Latino/Hispanic. However, the Pearson’s values for the remaining race/ethnicities are somewhat inconclusive, with none being significant even at the .05 level. There are only two religions (out of six) with Pearson’s values that are statistically significant: Atheist/Agnostic with a value of -.384 (at the .01 significance level), and Other Religion with a value of .220 (at the .05 level). Thus, there is a strong negative association between those that are Atheist or Agnostic and the average percent of Whites living in their neighborhoods, whereas there is a potential positive association between those who responded to belonging to an unlisted religion and the dependant variable. The Pearson’s value measuring the relationship between Legal Status (in regards to citizenship) and White percentage of ZIP Code is not statistically significant (its value is - .147). This implies a lack of relationship between the two variables. Of the seven different occupations to receive a response for the Job Type variable, only two had any sort of statistically significant association with the dependant variable, though in opposite directions. These two are Administrative and Education, with Pearson’s values of .565 and -.585, respectively (with both values significant at the .01 level). This implies that those working in the administrative field are more likely to live in areas of higher concentrations of Whites, on average, than those working in the education field. Though retail, food service, and public services jobs yielded figures denoting negative associations with the dependant variable, while business and health/medicinal fields yielded positive association values, none of these occupation types yielded results with any acceptable statistical significance levels, so no such associations can be established. The bivariate analyses between the dependant variable and the three dichotomous variables constituting the [immigrant] Generation variable yielded results that can be said to be expected in RESIDENTIAL SEGREGATION by Jesus Campos 22 CALIFORNIA SOCIOLOGY JOURNAL, 2011, VOL. 3 (ISSUE 1: 12-32) regards to the relationship between the two factors. Though the Pearson’s value for those who responded to belonging to 1st Generation yielded a statistically insignificant value, the analyses for 2nd Generation and Beyond 2nd Generation produced values (significant at the .01 level) of -.338 and .345, respectively. An interpretation implies that those who belong to the 2nd immigrant generation (i.e. the first generation of their family to be born in the United States) are likely to live in areas with a lesser concentration of Whites than those whose families have been in this country longer, though the exact generation the respondent belongs to cannot be determined. The bivariate analysis between the dependent variable and Marital Status did not produce a statistically significant Pearson’s value, nor did the analysis with the variable Age. However, the relationship between the dependent variable and Home Ownership did produce a positive association of .415 (significant at the .01 level). This means that those who own their own housing live, on average, in neighborhoods with a greater percentage of Whites than those who rent or lease. Similarly, the association between the dependent variable and Sex also produces a Pearson’s value that is statistically significant (at the .01 level) value. The value, .353, implies a positive association between the respondent’s sex (when moving from female to male) and the dependent variable. OLS REGRESSION ANALYSIS Though the various Pearson’s values allow us to establish a correlation between two variables, such an analysis does not allow us to measure the effect that all possible key and control variables have on the dependent variable as a whole. Thus, a linear regression must be utilized to determine the overall influence of all independent variables on the dependent variables; in essence, a linear regression allows one to measure the strength of the model as a whole. For this particular model, an OLS regression analysis will be utilized since the single dependent variable is of continuous value (a percentage figure measuring the percentage of Whites living in the respondent’s ZIP Code). In theory, a linear regression analysis of the research model would include the single dependent variable plus the twenty-nine key and control independent variables (most of which are recoded dichotomous variables). However, SPSS excluded six of the independent variables (though not key ones) during the regression analysis. The excluded variables are White; Latino/Hispanic; Roman Catholic; Deist/Spiritualist; Administrative Job; and Beyond 2nd Generation. The reason for these exclusions may be related to an issue of perfect multicollinearity with the variables in question. Several attempts were made to solve this issue by recoding some of the existing dichotomous variables pertaining to Race/Ethnicity and Religion in order to compress them into fewer variables. No such recode attempts produced a model in which no independent variables were excluded by SPSS. The model which ultimately bore the fewest excluded variables (four) was one in which four separate religious variables (Buddhist/Tao/Shinto; Atheist/Agnostic; Deist/Spiritualist; and RESIDENTIAL SEGREGATION by Jesus Campos 23 CALIFORNIA SOCIOLOGY JOURNAL, 2011, VOL. 3 (ISSUE 1: 12-32) Other Religion) were merged into a single category known simply as Other Religion, and two racial/ethnic variables (Multi-Racial/Ethnic and Other Race/Ethnicity) were merged into a single category known as Other Race. The four excluded variables under this new model are Latino/Hispanic; Roman Catholic; Administrative Job; and Beyond 2nd Generation. This new model is summarized in Appendix IV as Table 1. Despite the changes between them, the initial and newer models yielded similar regression values, with F-test values of 11.23 and 11.91, respectively, and Adjusted R-squared values of .77 and .76, respectively. Since this second model yielded fewest excluded variables, it shall be used to analyze all other independent variables. As stated, the linear regression analyses for the research model yielded a statistically significant (at the .01 level) F-test score of 11.91, meaning that at least one of the independent variables has a significant effect on the dependent variable). The Adjusted R-squared value of .76 means that the model can explain 76% of the variance in the dependent variable. In regards to specific independent variables, three of the variables (Income; Other Race; and 2nd Generation) produced values significant at the .05 level. The variable 2nd Generation produced a negative relationship, meaning those who are in the first generation of their family to be born in this country decreases the average White percentage of one’s neighborhood (a somewhat expected outcome). Another expected outcome is the positive relationship between one’s level of education and the dependent variable, meaning that, on average, as the respondent’s level of education rises, so does the percentage of Whites living in their neighborhood (i.e. ZIP Code). The positive relationship between the dependent variable and the variable Other Race may be somewhat confusing. Though this variable excludes Whites, Latino/Hispanics, and Asians, it is difficult to specify as to what race this significant association can be attributed to. CONCLUSION Overall, the regression model may be interpreted as having produced somewhat ambiguous results. The model is strong in that in can explain much of the variance in the dependent variable; however, the fact remains that only three of the over two dozen independent and control variables yielded any sort of statistically significant influence (with only one of the three key variables having any sort of significant influence). Furthermore, variables that were largely expected to produce a significant positive relationship (such as Education, Savings, and White) failed to do so. Similarly, variables that were expected to produce negative relationships (most notably Latino and Black) also failed to do so (with Latino even being completely excluded by SPSS from the model). Obviously, such outcomes may be a result of issues relating to sample size, or perhaps more likely to issues relating to the use of availability sampling. Nevertheless, in taking into account the absence of such largely expected outcomes, it may perhaps be inaccurate to conclude that the overall model is indeed a strong one. RESIDENTIAL SEGREGATION by Jesus Campos 24 CALIFORNIA SOCIOLOGY JOURNAL, 2011, VOL. 3 (ISSUE 1: 12-32) REFERENCES Beshers, James M.; Laumann, Edward O.; Bradshaw, Benjamin S. Ethnic Congregation-Segregation, Assimilation, and Stratification. Social Forces, Vol. 42, No. 4. (May, 1964), pp. 482-489. Fong, Stanley L. M. Assimilation of Chinese in America: Changes in Orientation and Social Perception. The American Journal of Sociology, Vol. 71, No. 3 (Nov., 1965), pp. 265-273. Fong, Eric; Shibuya, Kumiko. Suburbanization and Home Ownership: The Spatial Assimilation Process in U.S. Metropolitan Areas. Sociological Perspectives, Vol. 43, No. 1 (Spring, 2000), pp. 137-157. Fong, Eric; Wilkes, Rima. The Spatial Assimilation Model Reexamined: An Assessment by Canadian Data. International Migration Review, Vol. 33, No. 3 (Autumn, 1999), pp. 594-620. Gordon, Milton. Assimilation in American Life: The Role of Race, Religion, and National Origins. New York: Oxford University Press, 1964. Gordon, Milton. Assimilation in America: Theory and Reality. Daedalus, Vol. 90 Iss. 2 (Spring 1961), pp. 263- 285. Hirsch, Walter. Assimilation as Concept and Process. Social Forces, Vol. 21, No. 1 (Oct, 1942), pp. 35-39. Iceland, John; Wilkes, Rima. Does Socioeconomic Status Matter? Race, Class, and Residential Segregation. Social Problems, Vol. 53, Issue 2 (May, 2006), pp. 248-273. Logan, John R.; Alba, Richard D.; Leung, Shu-Yin. Minority Access to White Suburbs: A Multiregional Comparison. Social Forces, Vol. 74, No. 3 (Mar., 1996), pp. 851-881. Massey, Douglas S.; Bitterman, Brooks. Explaining the Paradox of Puerto Rican Segregation. Social Forces, Vol. 64, No. 2 (Dec., 1985), pp. 306-331. Massey, Douglas S.; Denton, Nancy A. The Dimensions of Residential Segregation. Social Forces, Vol. 67, No. 2 (Dec., 1988), pp. 281-315. Massey, Douglas S.; Mullan, Brendan P. Processes of Hispanic and Black Spatial Assimilation. The American Journal of Sociology, Vol. 89, No. 4 (Jan., 1984), pp. 836-873. Massey, Douglas S.; Eric Fong. Segregation and Neighborhood Quality: Blacks, Hispanics, and Asians in the San Francisco Metropolitan Area. Social Forces Vol. 69, No. 1 (Sep., 1990), pp. 15-32. Massey, Douglas S.; Denton, Nancy A. Spatial Assimilation as a Socioeconomic Outcome. American Sociological Review, Vol. 50, No. 1 (Feb., 1985), pp. 94-106. Woolston, Howard. The Process of Assimilation. Social Forces, Vol. 23, No. 4 (May, 1945) pp. 416-424. http://www.jstor.org/search/BasicResults?Search=Search&Query=aa:%22Stanley%20L.%20M.%20Fong%22&hp=25&si=1 http://www.jstor.org/browse/00029602 http://www.jstor.org/browse/00029602 http://www.jstor.org/browse/00029602/dm992565 http://www.jstor.org/browse/07311214 http://www.jstor.org/browse/07311214 http://www.jstor.org/browse/07311214/ap050069 http://www.jstor.org/search/BasicResults?Search=Search&Query=aa:%22Eric%20Fong;%20Rima%20Wilkes%22&hp=25&si=1 http://www.jstor.org/view/01979183/di009803/00p0518p/0?currentResult=01979183%2bdi009803%2b00p0518p%2b0%2c67408005&searchUrl=http%3A%2F%2Fwww.jstor.org%2Fsearch%2FAdvancedResults%3Fhp%3D25%26si%3D1%26All%3Dfong%2Bwilkes%26Exact%3D%26One%3D%26None%3D%26sd%3D%26ed%3D%26jt%3D http://www.jstor.org/view/01979183/di009803/00p0518p/0?currentResult=01979183%2bdi009803%2b00p0518p%2b0%2c67408005&searchUrl=http%3A%2F%2Fwww.jstor.org%2Fsearch%2FAdvancedResults%3Fhp%3D25%26si%3D1%26All%3Dfong%2Bwilkes%26Exact%3D%26One%3D%26None%3D%26sd%3D%26ed%3D%26jt%3D http://www.jstor.org/view/01979183/di009803/00p0518p/0?currentResult=01979183%2bdi009803%2b00p0518p%2b0%2c67408005&searchUrl=http%3A%2F%2Fwww.jstor.org%2Fsearch%2FAdvancedResults%3Fhp%3D25%26si%3D1%26All%3Dfong%2Bwilkes%26Exact%3D%26One%3D%26None%3D%26sd%3D%26ed%3D%26jt%3D http://www.jstor.org/browse/01979183 http://www.jstor.org/browse/01979183/di009803 http://www.jstor.org/search/BasicResults?Search=Search&Query=aa:%22John%20R.%20Logan;%20Richard%20D.%20Alba;%20Shu-Yin%20Leung%22&hp=25&si=1 http://www.jstor.org/view/00377732/di010952/01p0005n/0?currentResult=00377732%2bdi010952%2b01p0005n%2b0%2cFF2A02CC&searchUrl=http%3A%2F%2Fwww.jstor.org%2Fsearch%2FAdvancedResults%3Fhp%3D25%26si%3D1%26All%3Dalba%2Blogan%26Exact%3D%26One%3D%26None%3D%26sd%3D%26ed%3D%26jt%3D http://www.jstor.org/view/00377732/di010952/01p0005n/0?currentResult=00377732%2bdi010952%2b01p0005n%2b0%2cFF2A02CC&searchUrl=http%3A%2F%2Fwww.jstor.org%2Fsearch%2FAdvancedResults%3Fhp%3D25%26si%3D1%26All%3Dalba%2Blogan%26Exact%3D%26One%3D%26None%3D%26sd%3D%26ed%3D%26jt%3D http://www.jstor.org/view/00377732/di010952/01p0005n/0?currentResult=00377732%2bdi010952%2b01p0005n%2b0%2cFF2A02CC&searchUrl=http%3A%2F%2Fwww.jstor.org%2Fsearch%2FAdvancedResults%3Fhp%3D25%26si%3D1%26All%3Dalba%2Blogan%26Exact%3D%26One%3D%26None%3D%26sd%3D%26ed%3D%26jt%3D http://www.jstor.org/browse/00377732 http://www.jstor.org/browse/00377732/di010952 http://www.jstor.org/browse/00377732/di010952 http://www.jstor.org/browse/00377732 http://www.jstor.org/browse/00377732/di010923 http://www.jstor.org/browse/00377732/di010923 http://www.jstor.org/search/BasicResults?Search=Search&Query=aa:%22Douglas%20S.%20Massey;%20Brendan%20P.%20Mullan%22&hp=25&si=1 http://www.jstor.org/view/00029602/dm992676/99p03734/0?currentResult=00029602%2bdm992676%2b99p03734%2b0%2c00&searchUrl=http%3A%2F%2Fwww.jstor.org%2Fsearch%2FAdvancedResults%3Fhp%3D25%26si%3D1%26All%3DMassey%2Bmullan%26Exact%3D%26One%3D%26None%3D%26au%3Don%26sd%3D%26ed%3D%26jt%3D%26dc%3DSociology http://www.jstor.org/view/00029602/dm992676/99p03734/0?currentResult=00029602%2bdm992676%2b99p03734%2b0%2c00&searchUrl=http%3A%2F%2Fwww.jstor.org%2Fsearch%2FAdvancedResults%3Fhp%3D25%26si%3D1%26All%3DMassey%2Bmullan%26Exact%3D%26One%3D%26None%3D%26au%3Don%26sd%3D%26ed%3D%26jt%3D%26dc%3DSociology http://www.jstor.org/browse/00029602 http://www.jstor.org/browse/00029602 http://www.jstor.org/browse/00029602/dm992676 http://www.jstor.org/search/BasicResults?Search=Search&Query=aa:%22Douglas%20S.%20Massey;%20Eric%20Fong%22&hp=25&si=1 http://www.jstor.org/browse/00377732 http://www.jstor.org/browse/00377732/di010930 http://www.jstor.org/view/00031224/di974363/97p0080r/0?currentResult=00031224%2bdi974363%2b97p0080r%2b0%2c00&searchUrl=http%3A%2F%2Fwww.jstor.org%2Fsearch%2FAdvancedResults%3Fhp%3D25%26si%3D1%26All%3DMassey%2BDenton%26Exact%3D%26One%3D%26None%3D%26au%3Don%26sd%3D%26ed%3D%26jt%3D%26dc%3DSociology http://www.jstor.org/view/00031224/di974363/97p0080r/0?currentResult=00031224%2bdi974363%2b97p0080r%2b0%2c00&searchUrl=http%3A%2F%2Fwww.jstor.org%2Fsearch%2FAdvancedResults%3Fhp%3D25%26si%3D1%26All%3DMassey%2BDenton%26Exact%3D%26One%3D%26None%3D%26au%3Don%26sd%3D%26ed%3D%26jt%3D%26dc%3DSociology http://www.jstor.org/browse/00031224 http://www.jstor.org/browse/00031224 http://www.jstor.org/browse/00031224/di974363 RESIDENTIAL SEGREGATION by Jesus Campos 25 CALIFORNIA SOCIOLOGY JOURNAL, 2011, VOL. 3 (ISSUE 1: 12-32) APPENDIX I This survey questionnaire is a part of a study attempting to measure the connection between socioeconomic status and residential patterns. The questionnaire consists of 14 questions, and should take roughly 5 minutes to complete. Please note that the survey is intended to be anonymous – do not write your name or any other identifying information on open-ended portions of the questionnaire. Your participation is greatly appreciated. Thank you. 1) Age: ______ 2) Sex: [ ] Male [ ] Female 3) What ZIP Code do you currently reside in? _________________________ 4) What is your current citizenship/residency status? [ ] Natural-born citizen [ ] Naturalized Citizen [ ] Legal Resident [ ] Temporary Resident/Visitor [ ] Undocumented Status 5) Marital Status [ ] Single [ ] Divorced/Widowed [ ] Married (including common law) 6) What is the highest level of education you have completed so far? [ ] Less than high school [ ] High school diploma/GED [ ] Associate’s Degree [ ] Bachelor’s Degree [ ] Master’s Degree [ ] Doctorate or equivalent RESIDENTIAL SEGREGATION by Jesus Campos 26 CALIFORNIA SOCIOLOGY JOURNAL, 2011, VOL. 3 (ISSUE 1: 12-32) 7) What is your approximate annual income (including that not reported to the government) to the nearest thousand? Please include wages from employment, money received from tenants, government subsidies (welfare), etc. Do not include income from bank and/or government loans that must be repaid. [ ] Below $10,000 [ ] $10,000-$20,000 [ ] $20,000-$30,000 [ ] $30,000-$40,000 [ ] $40,000-$50,000 [ ] $50,000-$60,000 [ ] Above $60,000 8) To the best of your knowledge, how much money do you have in savings to the nearest thousand? This includes money stored in a bank account, trust fund, mutual fund, stock options, real estate, etc. [ ] Below $10,000 [ ] $10,000-$20,000 [ ] $20,000-$30,000 [ ] $30,000-$40,000 [ ] $40,000-$50,000 [ ] $50,000-$60,000 [ ] Above $60,000 9) What is your religious affiliation/preference? [ ] Protestant [ ] Jewish [ ] Muslim [ ] Roman Catholic [ ] Hindu/Sikh [ ] Buddhist/Tao/Shinto [ ] Atheist/Agnostic [ ] Deist/Spiritualist [ ] Other 10) Do you own or rent/lease your current housing (whether it be a house, apartment, or otherwise)? Rent includes any money you pay your parents for housing if you still reside with them. [ ] Own [ ] Rent/Lease 11) If you or either side of your family immigrated to the U.S. from abroad, what generation do you belong to (using the definition of first generation as those who originally immigrated here)? Please leave blank if the question is not applicable to you. [ ] 1st Generation [ ] 2nd Generation RESIDENTIAL SEGREGATION by Jesus Campos 27 CALIFORNIA SOCIOLOGY JOURNAL, 2011, VOL. 3 (ISSUE 1: 12-32) [ ] 3rd Generation [ ] Beyond 3rd Generation 12) What is your current employment status? Temporary employment is employment through agencies/staffing services or seasonal employment. [ ] Unemployed [ ] Temporary employment [ ] Direct employment 13) How would you classify the field of your current employment (if you have more than one job, please choose the job which provides you with the majority of your income)? [ ] Retail [ ] Business [ ] Food Services [ ] Public Service [ ] Administrative [ ] Education [ ] Manufacturing [ ] General Clerical [ ] Health/Medicine [ ] Sales (including Insurance) [ ] Repair/Mechanic [ ] Other (Please specify): ________________________ 14) Which race/ethnic group do you identify yourself as belonging to? [ ] Black [ ] White (non-Hispanic) [ ] Asian [ ] Latino/Hispanic [ ] Native American or Alaskan Native [ ] Middle Eastern [ ] Indian [ ] Pacific Islander [ ] Other [ ] Multi-Ethnic RESIDENTIAL SEGREGATION by Jesus Campos 28 CALIFORNIA SOCIOLOGY JOURNAL, 2011, VOL. 3 (ISSUE 1: 12-32) Variable Categories Frequency Mean S.D. D V Percentage of Whites in ZIP Code N/A N/A 39.36 24.57 IN D E P E N D E N T V A R IA B L E S Education Level 1=Less than high school 2=High School Diploma/GED 3=Associate’s Degree 4=Bachelor’s Degree 5=Master’s Degree 6=Doctorate or equivalent 0 10 10 63 17 0 3.87 .81 Income 1=Less than $10,000 2=$10,000 to $20,000 3=$20,000 to $30,000 4=$30,000 to $40,000 5=$40,000 to $50,000 6=$50,000 to $60,000 7=More than $60,000 21 11 11 28 24 5 0 3.43 1.68 Savings 1=Less than $10,000 2=$10,000 to $20,000 3=$20,000 to $30,000 4=$30,000 to $40,000 5=$40,000 to $50,000 6=$50,000 to $60,000 7=More than $60,000 93 4 2 1 0 0 0 1.11 .45 C O N T R O L V A R IA B L E S Race/ Ethnicity#1 Black/Non-Black Asian/Non-Asian Latino or Hispanic/Non-Latino or Hispanic Other Race-Ethnicity/Non-Other Race-Ethnicity Multi-Ethnic/Non-Multi-Ethnic 5/95 7/93 64/36 5/95 5/95 .05 .07 .64 .05 .05 .23 .26 .48 .22 .22 Religion#2 Roman Catholic/Non-Roman Catholic Buddhist, Tao or Shinto/Non-Buddhist, Tao, or Shinto Atheist or Agnostic/Non-Atheist or Agnostic Deist or Spiritualist/Non-Deist or Spiritualist Other/Non-Other 44/56 5/95 16/84 3/97 30/70 .44 .05 .16 .03 .30 .45 .22 .37 .17 .46 Legal Status 1=Undocumented 2=Temporary Visitor 3=Permanent Resident 4=Naturalized Citizen 5=Natural-born Citizen 0 0 3 20 77 4.74 .51 Job Type#3 Retail Job/Non-Retail Job Business Job/Non-Business Job Food Services Job/Non-Food Services Job Public Service Job/Non-Public Service Job Administrative Job/Non-Administrative Job Health & Medicine Job/Non-Health & Medicine Job 7/75 3/79 3/79 16/66 28/54 6/76 .09 .04 .04 .20 .34 .07 .28 .19 .19 .40 .48 .26 Generation#4 1st Generation/Non-1st Generation 2nd Generation/Non-2nd Generation 12/83 64/31 .13 .67 .33 .47 Marital Status 0=Single 1=Married 64 32 1.33 .47 Home Ownership 0=Rent/Lease 1=Own 80 20 1.20 .40 Sex 0=Female 1=Male 47 53 .53 .50 Age 27.80 5.60 APPENDIX II TABLE 1: RESIDENTIAL SEGREGATION by Jesus Campos 29 CALIFORNIA SOCIOLOGY JOURNAL, 2011, VOL. 3 (ISSUE 1: 12-32) White Percentage of Neighborhood (N=100) Percent White 10 10 63 17 0 10 20 30 40 50 60 70 High School Diploma/GED Associate's Degree Bachelor's Degree Master's Degree Level of Educaton (N=100) 21 11 11 28 24 5 0 5 10 15 20 25 30 Income (N=100) 93 4 2 1 0 10 20 30 40 50 60 70 80 90 100 Savings (N=100) TABLE 2: #Denotes recoded variable in X/Y format, where X has value of 1 and Y has value of 0. 1 Reference group is White (non-Hispanic). 2Reference group is Protestant. 3 Reference group is Education. 4 Reference group is Beyond 2nd Generation. R E SI D E N T IA L S E G R E G A T IO N b y Je su s C am po s 30 C A L IF O R N IA S O C IO L O G Y J O U R N A L , 2 01 1, V O L . 3 (I SS U E 1 : 1 2- 32 ) 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 . W hi te P er ce nt ag e of Z IP C od e -.5 61 ** .0 70 . 23 7* . 37 5* * -.2 75 ** -.0 58 .0 23 .1 20 -.0 66 .1 13 -.0 30 .1 95 -. 38 4* * -. 01 9 2 . L ev el o f E du ca ti on .0 07 -. 48 9* * -. 36 3* * .2 14 * . 04 4 .0 37 .0 37 -.0 20 .0 23 -.0 82 .0 37 . 30 6* * . 10 1 3 . L ev el o f I nc om e . 23 9* . 19 5 .3 65 ** .3 64 ** -.0 55 -.3 46 ** -.2 88 ** -.0 34 .2 59 ** .2 35 * -. 15 7 -. 26 5* * 4 . L ev el o f S av in gs . 54 8* * -.2 83 ** -.0 68 -.0 57 -.0 57 -.0 57 -.0 35 -.2 19 * -.0 57 -. 04 7 -. 04 3 5 . W hi te -.5 38 ** -.1 11 -.0 93 -.0 93 -.0 93 -.0 58 -.3 58 ** .5 69 ** . 06 0 -. 07 1 6 . L at in o/ H is pa ni c -.3 66 ** -.3 06 ** -.3 06 ** -.3 06 ** .1 07 .6 65 ** -.3 06 ** . 04 3 -. 23 4* 7 . A si an -.0 63 -.0 63 -.0 63 -.0 39 -.2 43 * -.0 63 -. 12 0 -. 04 8 8 . B la ck -.0 53 -.0 53 -.0 33 -.2 03 * -.0 53 -. 10 0 -. 04 0 9 . O th er R ac e/ Et hn ic it y -.0 53 -.0 33 -.2 03 * -.0 53 -.1 00 -.0 40 10 . M ul ti -R ac ia l/ Et hn ic -.0 33 -.2 03 * -.0 53 . 76 7* * -. 15 0 11 . P ro te st an t -.1 27 -.0 33 -. 06 2 -. 02 5 12 . R om an C at ho lic -.2 03 * -. 38 7* * -. 15 6 13 . B ud dh is t/ Ta o/ Sh in to -. 10 0 -. 04 0 14 . A th ei st /A gn os ti c -. 07 7 15 . D ei st /S pi ri tu al is t 16 . O th er R el ig io n 17 . L eg al S ta tu s 18 . R et ai l J ob 19 . B us in es s Jo b 20 . F oo d Se rv ic es Jo b 21 . P ub lic S er vi ce Jo b 22 . A dm in is tr at iv e Jo b 23 . E du ca ti on Jo b 24 . H ea lt h & M ed ic in e Jo b 25 . 1 st G en er at io n 26 . 2 nd G en er at io n 27 . B ey on d 2n d G en er at io n 28 . M ar it al S ta tu s 29 . H om e O w ne rs hi p 30 . A ge 31 . S ex 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 A P P E N D IX I II *C or re la tio n is s ig ni fic an t a t . 05 le ve l. ** Co rr el at io n is s ig ni fic an t a t . 01 le ve l. #C an no t b e co m pu te d be ca us e at le as t o ne v ar ia bl e is c on st an t. R E SI D E N T IA L S E G R E G A T IO N b y Je su s C am po s 31 C A L IF O R N IA S O C IO L O G Y J O U R N A L , 2 01 1, V O L . 3 (I SS U E 1 : 1 2- 32 ) 1 .2 20 * -.1 47 -.0 24 .0 99 -.0 15 -. 15 2 .5 65 ** -.5 85 ** .1 15 .0 62 -.3 38 ** .3 45 ** -.0 46 .4 15 ** -.0 83 .3 53 ** 2 -.2 19 * -.0 83 -.2 89 ** .0 35 . 10 8 . 08 8 -.5 36 ** .6 18 ** .0 50 .0 63 .3 46 ** -.4 57 ** -.0 71 -.2 91 ** -.0 77 -.4 20 ** 3 -.1 57 .0 11 -.3 36 ** -.3 50 ** -. 35 0* * -. 20 0 .4 40 ** .1 98 .0 48 -.4 07 ** .2 41 * .0 55 .4 20 ** .2 28 * -.0 02 .3 67 ** 4 .3 29 ** .1 28 -. 08 4 -.0 54 -. 05 4 -. 13 5 .3 29 ** -.0 92 -.0 77 -.0 97 -.3 16 ** .4 50 ** .3 09 ** .4 38 ** -.1 72 .6 80 ** 5 .1 13 .0 37 -. 13 9 .4 29 ** -. 08 8 -.2 23 * .4 25 ** -.2 49 * -.1 27 -.1 58 -.5 97 ** .8 31 ** .1 93 .4 47 ** -.1 97 * .9 06 ** 6 -.5 55 ** -.0 15 .0 43 -.2 63 * -. 26 3* .1 71 -.3 28 ** .3 46 ** .1 10 -.0 73 .4 25 ** -.4 38 ** -.0 16 -.3 54 ** .3 37 ** -.3 48 ** 7 .4 19 ** .1 42 -.0 34 -.0 22 -. 02 2 -.0 55 -.0 80 -.0 61 .3 95 ** -.1 07 .1 96 -.1 41 -.1 98 -.1 37 .2 58 ** -.1 39 8 .3 50 ** .0 27 -.0 78 -.0 50 -. 05 0 -.1 25 .3 54 ** -.1 40 -.0 72 # # # .3 31 ** .4 59 ** -.2 44 * -.1 57 9 .3 50 ** -.3 38 ** -.0 69 -.0 44 -. 04 4 .3 17 ** -.1 63 .0 10 -.0 64 .6 20 ** -.3 39 ** -.1 18 -.1 66 -.1 15 -.2 24 * -.1 99 ** 10 .1 19 .2 87 ** -.0 50 -.7 65 ** -. 12 5 -.1 83 -.1 40 -.0 72 .0 90 -.1 18 -.1 64 -.1 18 -.1 66 -.1 15 -.2 44 * -.1 57 11 -.0 94 -.4 95 ** -.0 48 -.0 31 -. 03 1 -.0 78 -.1 14 .2 88 ** -.0 44 .3 86 ** -.2 11 * -.0 73 .2 06 * -.0 71 -.1 52 .0 18 12 -.5 80 ** .0 18 .1 31 -.1 95 -. 19 5 .2 46 * -.1 03 -.1 45 .1 87 -.0 99 .2 41 * -.2 01 .1 92 -.1 41 .3 91 ** -.2 31 * 13 -.1 50 -.1 55 -.0 78 -.0 50 -. 05 0 -.1 25 .3 54 ** -.1 40 -.0 72 -.0 90 -.3 39 ** .4 71 ** -.0 73 -.1 15 -.2 44 * .4 62 ** 14 -.2 86 ** .2 26 * .0 93 .4 29 ** -. 08 8 -.2 23 * -.3 27 ** .4 42 ** -.1 27 -.1 71 .1 33 -.0 14 -.3 16 ** -.0 14 -.1 36 .0 10 15 -.1 15 .0 91 -.0 60 -.0 38 1. 00 0* * -.0 96 -.1 40 -.1 07 -.0 55 -.0 69 .1 26 -.0 90 -.1 27 -.0 88 -.1 87 -.1 10 16 -.0 09 -.1 56 -.1 00 -. 10 0 .0 52 .3 30 ** -.2 10 -.0 28 .2 05 * -.1 96 .0 60 .0 48 .2 73 ** -.0 83 .0 58 17 .1 74 .1 11 . 11 1 -.4 10 ** .0 26 .0 43 .1 60 -.5 53 ** .3 03 ** .1 04 -.1 94 .0 10 .2 70 ** .0 83 18 -.0 60 -. 06 0 -.1 50 -.2 20 * -.1 68 -.0 86 -.1 15 -.2 31 * .3 43 ** -.2 62 * -.1 74 .1 62 -.1 08 19 -. 03 8 -.0 96 -.1 40 -.1 07 -.0 55 -.0 73 -.2 66 * .3 52 ** -.1 67 .3 43 ** .2 15 .1 84 20 -.0 96 -.1 40 -.1 07 -.0 55 -.0 73 .1 52 -.1 15 -.1 67 -.1 11 -.1 77 -.1 33 21 -.3 55 ** -.2 70 * -.1 38 .4 11 ** -.0 12 -.2 93 ** .2 22 -.2 80 * .0 48 -.3 13 ** 22 -.3 95 ** -.2 02 -.2 37 * -.1 56 .3 50 ** .1 78 .5 49 ** -.0 33 .4 56 ** 23 -.1 54 .0 73 .2 45 * -.3 28 ** .0 73 -.3 12 ** -.0 33 .4 56 ** 24 -.1 06 .2 20 -.1 66 -.2 41 * -.0 51 -.1 61 -.0 58 25 -.5 46 ** -.1 90 -.1 11 -.1 65 -.1 72 -.2 43 * 26 -.7 18 ** .0 01 -.2 53 * .1 94 -.5 01 ** 27 .1 00 .4 33 ** -.0 85 .7 89 ** 28 .2 90 ** -.0 74 .2 69 ** 29 -.1 30 .4 22 ** 30 -.2 20 * 31 *C or re la tio n is s ig ni fic an t a t . 05 le ve l. ** Co rr el at io n is s ig ni fic an t a t . 01 le ve l. #C an no t b e co m pu te d be ca us e at le as t o ne v ar ia bl e is c on st an t. RESIDENTIAL SEGREGATION by Jesus Campos 32 CALIFORNIA SOCIOLOGY JOURNAL, 2011, VOL. 3 (ISSUE 1: 12-32) APPENDIX IV Table 1: Linear Regression Independent Variables# Coefficients Standard Error Intercept 73.30 56.30 Education -14.02 5.26 Income 16.64 5.33* Savings -3.72 5.91 White -71.80 29.76 Asian -34.66 20.27 Other Race 37.26 11.56* Protestant 32.49 19.43 Other Religion 4.77 11.51 Legal Status 1.12 5.80 Retail Job -34.20 16.44 Business Job 70.62 32.23 Food Services Job 11.76 21.58 Public Service Job -8.55 13.47 Education Job -27.62 16.54 Health & Medicine Job 18.17 11.22 1st Generation -19.79 24.53 2nd Generation -57.07 17.32* Marital Status -3.60 8.93 Home Ownership (Rent→Own) .87 10.26 Age .37 1.33 Sex (Female→Male) -1.32 5.87 F-test 11.91 Adjusted R-squared .76 # Variables excluded by SPSS: Latino/Hispanic; Roman Catholic; Administrative Job; Beyond 2nd Generation *p<0.05