Hispanics are the fastest growing ethnic population in the United States. By the year 2025 they will be the single largest minority group in the country.¹ Sometimes referred to as the "sleeping giant," Americans of Mexican descent are the dominant ethnic group within the Hispanic subculture. Sizable populations of Mexican Americans reside in California, Texas, Illinois, and Arizona. Research on Mexican American professionals is scarce since they have not been identified specifically as "Mexican Americans" in studies about career advancement. Instead, they have been lumped together with all Hispanics, or, in some instances, with all minorities.² It is questionable whether previous research accurately describes, explains, and predicts the careers of this growing subculture of the American workforce. It is important to know more about this growing group of professionals who will help to shape public policy for themselves and the nation in the next millennium.

Since Mexican Americans have not been identified specifically as Mexican Americans in studies about career advancement, little is known about this emerging group of municipal administrators. In this study, Anglo/Whites and Black administrators serve as comparison groups for Mexican American administrators. Other groups, such as Native Americans and Asians are not included because they comprise only 1.9 percent and 1.6 percent of the Phoenix population and 1.6 percent and 2.2 percent of the population of Tucson, respectively.³

Data collection was accomplished through self-administered questionnaires and interviews with City of Phoenix and City of Tucson administrators and officers of the City of Phoenix Hispanic Network, a Chapter of the International-City/County Management Association (ICMA). Self-administered questionnaires were distributed to Mexican American, Black, and Anglo/White administrators in three city-wide job categories: executives (upper-managers), professionals (middle-managers), and supervisors (lower-managers). In interviews for this study, some Mexican American administrators suggest that they perceive that Mexican Americans have been over-looked for new hires and promotions. These perceptions are consistent with the finding by Sisneros (1993) that Hispanics, as a group, remain underrepresented at all levels of public administration.⁴ Interview sessions revealed also that Mexican American managers perceive that they are over-looked in favor of other minority groups—even though they are the dominant minority group in both cities. Inter-minority competition and jobs is described by Alozie and Ramirez (1999)⁵ but that issue is beyond the scope of this study.

Several explanations have been advanced to explain why Mexican Americans lag behind Blacks and women, in municipal employment. Some researchers suggest that Hispanics are less well educated than Anglo/Whites and Blacks. Research shows that Hispanic youth have the highest incidence of school drop-out rates in the nation—twice the rate for Black students and three times as high as Anglo/White students.⁶ It is suggested that only 55 percent of all Hispanics complete high school, while 75 percent of Blacks and 82 percent of non-Hispanic Anglo/White students graduate from high school.⁷ Others suggest that Hispanics, as a group, receive little mentoring, have fewer opportunities to participate in on-the-job-training (OJT), and suffer from having acquired little workplace seniority.⁸ Access to quality education, seniority, OJT, and English-proficiency are the most frequently cited reasons used to explain differences in the career success of Hispanics and Anglo/Whites.⁹

Variables such as education, seniority, mentoring, and OJT have been shown to contribute to the career advancement of other groups. Earlier studies, such as that by Taussig & Joslyn (1932) and Warner & Abegglen (1955) found that education, seniority, mentoring, and OJT are among the variables contributing to the careers of thousands of American businessmen.¹⁰ More recent studies, Bayes & Col (1991), Hale & Kelly (1989) and Schwartz (1989 & 1992), examined whether these variables also influence the careers of women in the workforce.¹¹ In sum, these studies conclude that education, seniority, mentoring and OJT are some of the more critical factors contributing to career advancement.

The question addressed here is: Do these variables also influence the careers of Mexican American professionals when examined as a specific subset of the Hispanic population? This study questions findings that consistently suggest that Hispanics, including Americans of Mexican descent, lag behind other municipal administrators in terms of education, seniority, and OJT. Answers to this question cannot be determined from research that lumps all Hispanics together. Studies by a growing group of researchers on Hispanics, including Alozie & Ramirez (1998); DeFreitas (1991); Melendez, Rodriguez, and Figueroa, 1991; Bean & Tienda (1987); and Borjas & Tienda (1982), suggest that lumping all Hispanics together obscures differences in the various Hispanic ethnic groups.¹² This is especially true when recently arrived Mexican nationals, with little education, are included in the study.¹³ Thus, the primary purpose of this study is to determine if "white-collar" Mexican American professionals in municipal administration lag behind in terms of education, seniority, and OJT when compared with other professionals who also are employed in municipal administration.

Specific variables that should influence the careers of Mexican American managers employed in the City of Phoenix and the City of Tucson are examined in this paper. A critical difference of this study is that efforts were made to carefully separate data of the target group by ethnicity and nativity. The literature suggests that research combining all Hispanic subcultures, including Mexican nationals with little education and English-proficiency, risks masking important differences between Mexican Americans born in the United States and other Hispanic subcultures.¹⁴ It has been suggested that more ethnic specific studies are needed.¹⁵ Ethnic specific studies that illuminate the careers of Mexican Americans, may broaden current perceptions of Mexican Americans as professionals in municipal administration.

The focus of this study is the American-born Mexican manager working in municipal administration. Data were collected in two municipalities, the City of Phoenix and the City of Tucson, Arizona. Although only two cities are used in this study, there is no theoretical reason to believe that Mexican American professionals in these research sites differ from other Mexican Americans living in similar large metropolitan cities. Moreover, Mexican Americans are the dominant minority group in both cities. Mexican Americans comprise 26.4 percent of the population of Phoenix and 29 percent of the population of Tucson. Anglo/Whites are approximately 70 percent and Blacks are less than 5 percent of the population of Phoenix. Anglo/Whites are approximately 65 percent and Blacks are 4 percent of the population of Tucson.¹⁶ As noted above, Asians and Native Americans comprise small percentages of city populations, thus making statistical analyses questionable.

This study presents statistical analyses of data obtained from a survey. The questionnaire was circulated to administrators in three ethnic groups: Hispanic, Black, and Anglo/White. Because of the theoretical focus of the research question, Mexican Americans were separated from all others in the "Hispanic" group. Administrators were separated further by managerial level—executive, professional, and supervisory. Descriptive statistics are presented within ethnic categories. A review of career advancement and social and economic mobility literature suggests that differences can be expected in each substantive area by ethnicity and managerial level. The data are expected to show that Anglo/White administrators have more seniority than Mexican American or Black administrators across all managerial levels, thus supporting the proposition that seniority is related to managerial level and that Anglo/Whites will have occupied managerial positions for a longer period of time. Private sector research at AT&T and Bell Systems show that seniority is a key factor in job promotion for men (Bray, Campbell, & Grant, 1974; Howard & Bray, 1988).¹⁷ Other studies report similar results for women in the public sector (Hale & Kelly, 1989; Bayes & Col, 1991; Schwartz, 1992).¹⁸

Similarly, the data are expected to show that Anglo/Whites have more education than Mexican Americans and Blacks across managerial levels, thus supporting the proposition that education is related to managerial level and that Anglo/Whites have had greater access to educational opportunities and special "prepping," as suggested by Rudenstine (1996).¹⁹ Other research by Morales and Ong (1993) and Borjas and Tienda (1985) also suggests that education is one of the most important factors affecting job promotion.²⁰

Like seniority and education, the data also are expected to show that Anglo/Whites receive more OJT than either minority group. This pattern is expected to hold across all managerial levels, thus supporting the proposition that OJT is related to managerial level and that the opportunities for such training are higher for Anglo/Whites. Several studies demonstrate the importance of OJT to job promotion, including those by Forbes and Piercy (1991), and Tienda and Ding-Tzann Lii (1987).²¹

Returned questionnaires numbered 278 (72.02%) of the total 388 distributed. Of the responses, 277 were usable. This return rate exceeds the normal standard of 60 percent for sample surveys.²² Among the respondents, there are 103 Mexican Americans (37.1%), 107 Anglo/Whites (38.5%), 11 "other Hispanic" (4.0%), 44 Blacks (15.8%), 8 "other ethnicity" (2.9%), and 5 "no ethnicity" (1.8%). The focus of this study is Mexican American, Black, and Anglo/White administrators; therefore, all other categories were eliminated from analysis. Other groups such as Native Americans and Asians are included in the "other ethnicity" category and represent less than three percent of all respondents. Thus, there are insufficient numbers of Native American and Asian managers for meaningful analysis of these groups. Administrators who returned completed questionnaires include 58 executives (upper-managers), 121 professionals (mid-managers), and 98 supervisors (lower-level managers).

Questionnaires of respondents who claimed "no ethnicity" or "other ethnicity" were eliminated from the analysis leaving 254 useable questionnaires. Although the proportions of executives among both minority groups are smaller, as expected, there are sufficient cases for analysis across the combinations of ethnic group and managerial levels. Table 2 presents distributions by citizenship. Table 3 presents between-city distributions. For purposes of this analysis, descriptive statistics are grouped by ethnicity and managerial level.

Among Mexican American participants there are thirteen executives, fifty-six professionals, and thirty-four supervisors. Anglo/White participants include thirty-five executives, forty professionals, and thirty-two supervisors. Black participants include eight executives, fourteen professionals, and twenty-two supervisors. The minority participant pool reflects the fact that Mexican Americans are the dominant ethnic group in Phoenix and Tucson comprising twenty-six and thirty percent of city populations, respectively. Anglo/Whites make-up approximately seventy percent and Blacks are less than five percent of city populations.²³

Table 2 shows that ninety-one respondents who self-identify as Mexican American are U.S. born, eight are naturalized citizens and one is not a citizen. Since the focus of this study is primarily on U.S. born Mexican American administrators, naturalized citizens and respondents who are not citizens were excluded from the study. All Blacks are included in the study. Citizenship and nativity is not, generally speaking, a troublesome issue with Anglo/Whites and Blacks, thus it was decided not to control for nativity in these groups.

Table 3 shows the distribution of participants by City of employment and ethnicity. Overall the respondents are approximately evenly split between the two cities. That is, 124 people (49.0%) work in Phoenix and 129 (51%) work in Tucson. The ethnic distributions between the two cities are less balanced. Most of the Mexican Americans (70 administrators or 68%) are located in Tucson, versus 33 Mexican Americans participating from Phoenix. Conversely, most of the Blacks (35 administrators or 79.5%) who returned a questionnaire worked for Phoenix; only nine (20.5%) of the Black administrators who answered the questionnaire were located in Tucson. The Anglo/Whites are about evenly split between the two cities. The differences in numbers of minority participants reflect first, the differences in employee pools available for sampling and, second, the differences in the choices of minority administrators selected to complete the questionnaire. Since there is no theoretical reason to suspect between-city variation on key analyses variables, and because of the practical need to use two cities to acquire enough minority cases to analyze, the pool of respondents combines Tucson with Phoenix administrators.

The variables selected for this study of Mexican American professionals are seniority, education, and on-the-job-training (OJT). It was expected that Anglo/White administrators would have more seniority, education, and OJT than Mexican American and Black administrators at all three managerial levels.

Historically Anglo/Whites have dominated white-collar positions in the American workforce, thus it was expected that Anglo/White administrators would show greater seniority levels in administrative capacities than Mexican Americans and Blacks.²⁴ There are several ways to look at seniority, depending upon whether one is concerned with the individual’s work life or organization as a point of reference. As a preface to an organizational measure of seniority, two other measures of longevity are considered below.

Table 4 shows the mean age and years in the workforce for each of the ethnic groups in this study. With respect to age, Mexican American and Anglo/White respondents are virtually identical with 45.7 years and 45.9 years respectively. The Blacks in the study are somewhat younger than either of the other groups with a mean age of 41.2 years. A one-way analysis of variance indicates that these mean differences are statistically significant (F=4.07, p<.05). In spite of the age difference, study participants do not differ significantly (F=.13, p>.05) in terms of their mean number of years in the workforce. These figures tell us that although Mexican Americans and Anglo/Whites are somewhat older, all participants can be characterized as mature, having spent about half their lives in the work force.

Table 5 measures seniority as number of years the person has spent in city employment and cross-classifies that by managerial level and ethnic group. The reason for sorting by managerial level is to attempt to isolate any differences that may be related to differing years required to attain a particular position. While organizational longevity is not necessarily a criterion for promotion—municipal organizations currently hire for many different managerial levels from the outside—it was deemed appropriate to choose a conservative analytic route and control for this factor explicitly. Thus, the cells of Table 5 show the mean number of years with the city for different combinations of ethnic group and managerial level.

When one looks at the means within managerial levels and across ethnic groups, an interesting pattern is visible. Among executives, Anglo/Whites show greater seniority (a mean of 17.0 years) than either Mexican American (14.2 years) or Black (15.6 years). This finding matches the expectation that Anglo/Whites should show higher seniority levels. Thus, Anglo/White executives have spent nearly two years longer in the organization than Black executives and nearly three years longer than Mexican American executives. This indicates that minority administrators either move more quickly into these positions than Anglo/Whites, or that minorities are more often recruited into these positions from the outside, thereby being newer to the organization. When one looks at the mean number of years with the city for professional level managers, there is no statistical difference among ethnic groups. Similarly, the differences among supervisory level administrators are not statistically significant.

Table 5.

In summary, the initial expectation that Anglo/Whites would have higher seniority holds only for executive managers. Seniority levels for supervisors and professionals are virtually the same across the three ethnic groups. Consequently, when a two-way analysis of variance is performed with executive level and ethnicity as factors, the main effect for ethnicity (F=.117) and the interaction effect for ethnicity with managerial level (F=.352) are not statistically significant (p>.05). It is interesting to note that there is a significant difference by managerial level (F=4.97; p<.05) such that the mean number of years with the organization for professionals is much higher than is the case for either executives or supervisors. This finding suggests that professionals (mid-managers) spend longer times in the organization. One may speculate that professionals spend longer time in middle-management positions because organizations prefer to hire from the outside to acquire "new blood," and, thus, new ideas.

Historically Anglo/Whites have had greater access to educational opportunities and "prepping" than members of minority groups (Rudenstine, 1996).²⁵ Thus, it was expected that Anglo/White administrators would have greater levels of education than Mexican Americans and Blacks. Educational attainment varied widely among the study participants. Ultimately, education was measured as three ordinal categories, capturing people with less than a B.A. degree, those with a B.A. and possibly some graduate work, and those with a master’s or doctoral degree. Table 6 shows managerial level by education for each of the three ethnic groups.

A sub-table gamma statistic is shown for the relationship between educational level and managerial level for each ethnic group. Gamma is an ordinal level measure of association with a Proportional Reduction in Error interpretation.²⁶ Values, therefore, vary between –1.0 and +1.0, where unity indicates perfect predictability and 0.0 represents statistical independence. Other values of the statistic can be interpreted as the proportion of error in predicting managerial level that is reduced or accounted for when one takes into account the joint variability of managerial

level and education. One can "generally" interpret gamma as a measure of "correlation" between two ordinal scaled variables.²⁷

In Table 6 the order of values for managerial level is from highest (executive) to lowest (supervisor), so that the sub-table gamma coefficients are negative and thereby indicate that there is a positive relationship between education and managerial level. This holds true for all three ethnic groups, and represents a statistically significant relationship (p<.05) in each case: Mexican American’s, -.46; Anglo/Whites, -.52; Blacks, -.64. The expectation that Anglo/Whites would have more education than Mexican Americans and Blacks is supported to the extent that education is highly positively related to managerial level for each of the ethnic groups. The coefficient is largest for Blacks, but considerably smaller for Anglo/Whites and Mexican Americans. Thus, for all ethnic groups, people at executive levels have the highest education levels, with supervisors having the lowest educational levels. At this point, the difference in coefficient magnitude only indicates that the overall strength of the relationship between education and managerial level differs for the ethnic groups. The difference between Mexican Americans and Anglo/Whites is small, while the gamma for Blacks appears higher. The uneven marginal distributions and small cell sizes for the Black table suggest, however, that caution should be exercised in interpreting this value. Gamma is known to be attenuated by marginal distributions²⁸, and the value may be artificially high.

If we look at the ethnic group sub-tables in Table 6, it is clear that there are too few cases in the Black sub-table for meaningful comparisons. If we compare Mexican Americans with Anglo/Whites within managerial levels, there is no statistical difference in education levels of Anglo/Whites and Mexican Americans. This is an unexpected finding that may be partially attributable to separating data by ethnicity and nativity, thus eliminating Mexican nationals from analyses.

Table 6.

It also was expected that Anglo/White’s would have higher levels of education than either minority group, without regard to managerial level. Table 7 shows these data as a cross-tabulation. This table combines cases across managerial levels and, therefore, creates larger cell sizes. While this procedure obscures educational level differences between managerial levels, the basic pattern of relationships between ethnic group and educational level remains visible. Consequently, the proportions of each ethnic group with a bachelor’s degree remain almost identical, while there are more Anglo/Whites with a master’s degree and more Mexican Americans with less than a bachelor’s degree. These differences are small enough that they are not statistically significant (Chi-squared = 7.9, p>.05). Thus, the expectation that Anglo/Whites would have more education than all Mexican American and all Black managers in the study is not supported. Again, this is a surprising finding. It may be partially attributable to the sorting by ethnicity and nativity. In other words, Mexican Americans, Anglo/White and Black administrators do not differ significantly in terms of educational achievement, as expected.

Because of the historical dominance of Anglo/Whites in the workforce, it was expected that Anglo/Whites would have participated in more OJT. It also was expected that executive level managers would have acquired a greater number of OJT compared to mid- and lower-level managers. In measuring OJT experiences, participants were given a list of six specific areas of training offered regularly by the municipalities. The six areas are: leadership, time management, personnel management, strategic planning, budget management, and organizational skills. A seventh area of "other training" also was used and participants who reported "yes" to this category were asked to specify the type of training obtained. A simple index of OJT extensiveness was obtained by summing the number of OJT experiences reported by each participant.²⁹ The index varies between 0 (no experience) and 7 (seven experiences).

Since OJT opportunities are known to vary by managerial level, Table 8 shows the mean number of OJT experiences by ethnicity and managerial level. A two-way analysis of variance on these data indicates that the main effect for ethnicity is not statistically significant (F=1.6, p.>05). The two-way interaction of ethnicity and managerial level is similarly not statistically significant (F=1.33, p.>05). Thus, contrary to expectation, the mean number of on-the-job training experiences does not differ enough among ethnic groups to be statistically reliable. In fact, even the small mean differences that do exist among ethnic groups (but within managerial levels) do not favor Anglo/Whites as expected. This is a surprising finding that may be partially attributable to separating the data by ethnicity and nativity. There is, however, a significant main effect for managerial level (F=5.18, p<.05). Partly because of the very high mean for Black professionals, administrators in the category of professional show the highest levels of training, followed by executives, with supervisors showing the lowest levels of OJT. Overall, while the expectation of differences in OJT among supervisory levels was born out, the differences among ethnic groups was not. These findings are surprising given the historical dominance of Anglo/Whites in white-collar positions (Lafer, 1992).

The findings of this study are different from other studies. In this study, Mexican American managers do not appear to lag behind on important career variables such as seniority, education, and on-the-job training. These differences and what they may mean for the field are discussed below.

Separation of data by ethnicity and nativity distinguishes this study from previous research that includes all Hispanics. First, Mexican nationals were separated from Mexican Americans. Secondly, Mexican American professionals were separated from all other Spanish-surnamed ethnic groups. The need for separating data by ethnicity and nativity has been recommended consistently in larger studies about Hispanics in municipal employment.³¹ It is conceded generally that separation of data by ethnicity and nativity may yield a more accurate profile of each ethnic group on specific variables examined.³²

Separating the data by ethnicity and nativity in the current study did, in fact, result in very different findings. Other studies report that Hispanics, as a group, consistently lag behind in terms of seniority, education, and OJT. Findings from this study suggest that Mexican Americans, when separated from all other Hispanics, including Mexican nationals, do not appear to lag behind on these important career variables. Future studies that separate data by ethnicity and nativity may conclude also that Mexican American municipal administrators have invested in the important human capital known to influence careers such as seniority, education, and OJT. One may speculate that separating data on other American-born Hispanic subcultures may yield similar results. In other words, it may be found that all American- born Hispanic subcultures measure high on important career related variables such as education, seniority, and OJT, while foreign-born Hispanics do not. Stated differently, foreign-born Hispanic subcultures may not measure as high on these variables because they lack educational opportunities that preclude them from participating in professional management positions. Thus, the scores of foreign-born Hispanics, when averaged in with American-born Hispanics, may have a tendency to drive down the scores of all Hispanics in a study, giving the appearance that all Hispanic subcultures fall short in education, seniority, and OJT. This, of course, has been suggested throughout the literature.

Future studies should continue to separate data on the basis of nativity and ethnicity, thus highlighting important differences among members of Hispanic subcultures born in the United States. Also, since Mexican American administrators in this study do not appear to lag behind on the important variables that influence careers, it may be timely to examine further whether they are garnering their fair share of managerial positions. Additionally, one may ask also just how affirmative action policies and workplace discrimination may be affecting their careers. Affirmative action and workplace discrimination are variables that may help to further explain the career advancement, or lack of advancement, of Mexican Americans in public administration. These issues form the basis for future studies.