Recent work-life research has focused on the effect of employment characteristics on overall life satisfaction (Hughes, Galinsky, & Morris, 1995; Wallen, 2002) as well as the positive and negative effects of job demands on the family (Hughes & Galinsky, 1988). It has been reported that well-being in family relationships is compromised by an imbalance between work life and family life (Wallen, 2002). This imbalance, often reflected by feeling tired, emotionally drained, or in a bad mood, is associated with long work hours, highly demanding jobs, and lack of supervisor empathy for family demands (Burke, Weir, & DuWors, 1979; Hughes et al., 1995). This has been referred to as negative spillover from work to family. Spillover is defined as “processes whereby experiences in one role impact on other roles in the same individual” (Pleck, 1995, p. 21). In the case of work and family, spillover refers to the impact of work roles on family and vice versa.

Models employed to study work and family issues have explored demographic characteristics, as well as employment variables, as possible determinants of spillover. Demographic variables may include age, education, marital status, presence of children, occupation, and income, while employment characteristics may consist of supervisor support, pressure on the job, job autonomy, and workplace culture.

The significance of demographic and family variables on work and family conflict and spillover has been noted in previous research. The presence of children in the household, as well as the number of children, has been found to increase negative spillover (Elliot, 2003; Hughes et al., 1995; Maume & Houston, 2001). Along with the presence and number of children, worker’s age impacts negative spillover. According to Grzywacz, Almeida, and McDonald (2002), “the shape of the relationships between age and negative spillover between work and family was curvilinear” and “advancing age was associated with more positive spillover from work to family” (p. 31). Other demographic variables associated with negative spillover include educational attainment and occupation. Maume and Houston (2001) find that among men “workplace demands are highest among highly educated salaried workers” (p. 182). Similar findings were reported by Roxburgh (2002). Highly educated salaried workers are likely to be represented among professional and managerial occupations (Jacobs & Gerson, 1997).

Pleck (1995) recognizes that while individual and family variables, such as age, education, occupation, marital status, and presence of children are important as explanatory variables of work to family spillover, they should be seen as mediators of workplace variables. Previous research exploring the association between workplace characteristics and negative spillover from work to home has focused on variables such as work hours, work flexibility and autonomy, workplace pressure, and supervisor support. Exploring the impact of gender on conflict between work and family, Maume and Houston (2001) found that while controlling for other variables, “long work hours increased the reported incidence of work–family conflict for women” (p. 185). Conversely, “job autonomy and working in a supportive culture are both associated with reduced spillover for men and women” (p. 182).

While previous studies have made large contributions in the understanding of work-life issues such as those related to spillover, research gaps have been identified. For example, should we assume that the findings of these studies can be generalized to include other races/ethnic groups besides non-Latino White workers? While exploring the National Study of Daily Experiences, Grzywacz and colleagues (2002) found that African Americans have lower levels of negative spillover from work to family; is this the case for other ethnic groups? Furthermore, if workplace policies are going to be based on empirical evidence, there is a need to address the cultural diversity of the work force. As stated by Klein (1999), “we in the field of work–family have made progress in serving those who ‘made it,’ but we are underserving the traditionally underserved” (p. 117). The analysis of the experiences of other underrepresented groups in the United States, such as Latinos, needs to be addressed (Grzywacz et al., 2002).

Although Latinos are not a homogenous group, Latinos share similar work experiences and family characteristics that make them distinguishable from other ethnic and racial groups.1 Relative to non-Latino Whites in the United States, Latinos are younger, less educated, and are more likely to be unemployed, in poverty, and overrepresented in service, operator, and laborer occupations (Kazis & Miller, 2001; Therrien & Ramirez, 2000). More than one in five Latinos in the United States lives in poverty and over 40% work in the service sector, as operators, or laborers. Conversely, Latinos are grossly underrepresented in managerial and professional occupations and are less likely to work flextime (Sharpe, Hermsen, & Billings, 2002; Therrien & Ramirez, 2000). Due to their young age, their relatively high fertility, and immigration, Latinos are anticipated to continue growing and expected to constitute over a quarter of the U.S. population by the year 2025 (U.S. Census Bureau, 1997a). While the Latino population increases in size, our knowledge of the dynamics of work and family within this group is very limited.

Furthermore, household size and marital status of Latinos in the United States differ considerably from those of non-Latino Whites. Latinos have larger families compared to non-Latino Whites where nearly a third of Latino households consist of five or more people, relative to only 11.8% among non-Latino White households (Therrien & Ramirez, 2000). Moreover, a large percentage of households, especially among Puerto Ricans, are female-headed (Baca-Zinn & Eitzen, 2002; Lichter & Landale, 1995; Sanchez-Ayendez, 1998). Regardless of household size and composition, and amid negative economic characteristics, including high levels of poverty, family scholars and the popular media often point to the importance Latinos place on the family (Mindel, Habenstein, & Wright, 1998). While previous research by Dietz (1995) has found that younger generations of Mexican Americans in the United States are not providing the instrumental support needed by the elderly it does find that “more than one-third of the respondents baby-sit for their younger family members and more than one-half help their family members make decisions” (p. 353). The term familism is often associated with the Latino community. Familism, the idea that the family precedes the individual in importance, is often associated with resilient family networks which provide emotional and social support and persists with successive generations (Velez-Ibañez, 1996). According to Becerra (1998), “the importance of the familial unit continues as a major characteristic among Mexican Americans to this day” (p. 161). Similar statements about the persistence of familism were found by Suarez (1998) for Cubans and by Sanchez-Ayendez (1998) for Puerto Ricans. Familism and resilient family networks may be necessary among Latinos in order to cope with marginal employment and economic conditions (Dohan, 2003).

While general characteristics of Latino families and their work representation are known, the interface between these two spheres, family and work, has not been fully explored by researchers. Based on their socioeconomic characteristics, one would anticipate that Latinos encounter a greater challenge relative to non-Latino Whites in balancing work and family needs. However, large household size and strong family networks as part of the sociocultural system in their environment may moderate the impact of negative work roles on the Latino family (Deacon & Firebaugh, 1988) and may contribute to overall life satisfaction.

Theoretical Framework

In this study, the family ecosystems approach is utilized. The family ecosystems approach offers a versatile framework to study the complex interaction and possible conflict between work and family where cultural differences can be incorporated. According to the family ecosystems framework (Deacon & Firebaugh, 1988; Goldsmith, 2000; Paolucci, Hall, & Axinn, 2001; Rice & Tucker, 1986), a family forms a distinctive system separated from other systems by its own boundaries, but interacts with its environment, including the market/work system and the sociocultural system, through the relative permeability of these boundaries. Each one of these systems includes within its boundaries the members (family members, employers/employees) that interact with each other guided by their roles, rules, routines, and responsibilities as affected by the cultural values of the family and the workplace. The exchanges between systems “...may be in the form of goals, needs, or expectations one system holds or provides for another or as tangible support or resources” (Deacon & Firebaugh, 1988, p. 33). This interaction between the family system and other systems in its environment occurs at the point where the systems interface and it is here where information is exchanged and where problems between the systems may be generated. The elements of the family system include inputs, throughput, outputs, and feedback. Inputs are classified as demands and resources. When a system uses resources to meet demands, the changes in resources and the met demands then enter the environment as outputs from that system. Throughput is the transformation process from input to output. The information about the output that reenters the system is called feedback. Feedback can also happen within a given system.

An individual from any culture as a family member and a worker confronts the demands from both work and family systems (inputs). When there is a perceived imbalance between these demands and the resources available to respond (feedback), stress becomes evident and the overall life satisfaction (output) may be affected depending on the capacity of the members of the systems to adapt and to respond to these demands (throughput). Assuming permeability in the boundaries of the systems, it is also possible that the demands and resources from one system will spillover to the other system (e.g. output from the work system reentering the family system as input in the form of new demands or resources or vice versa), contributing to an increase (positive spillover) or a decrease (negative spillover) in the overall life satisfaction (output) of the individuals involved. One factor that may affect or influence the decisions and responses of a particular family system to the demands confronted could be their culture, which Paolucci et al. (2001) define as the “set of habitual and traditional ways of thinking and responding which characterizes a particular group at a given time” (p. 49). The particular culture of a group may influence or determine the throughput or transformation process. Furthermore,

[w]ithin self-regulating family systems, differences may occur in the throughput processes. Two system concepts apply to these differences. With varying initial circumstances or conditions, the throughput process of two individuals or families may lead them to similar conclusions. This phenomenon is called equifinality. And yet, relatively similar opportunities and beginning orientations can lead to different outcomes: multifinality. (Deacon & Firebaugh, 1988, p. 17)

The demographic, work conditions, and family characteristics that distinguish Latinos can influence the dynamics of work–family interaction (Baca-Zinn & Eitzen, 2002) including the effect of work on family well-being. Given their different work and family situations for Latinos and non-Latino White families (inputs) how do these groups compare in relation to life satisfaction and negative spillover (output)? Following the theory, it is reasonable to assume that the unobserved transformation process (throughput) includes the particular way in which each cultural group manages the demands from work and family. Evidence of equifinality thus merits further exploration of the throughput process in the management of work and family demands among Latinos. Due to data limitations, only the input and output parts of the family system will be examined in this study.

Empirical Model

The focus of this study is on the determinants of negative spillover from work to home/family as well as the relationship between those factors and life satisfaction, a proxy for quality of life. The dependent variables are outputs of the family system and the independent variables stand for inputs. Based on the systems framework and the findings of previous studies, the following empirical model will be followed in the statistical analyses: OLS=f (NS, FIV) NS=f (FIV, HWPW, SSUPPORT, JPRESSURE, AUTONOMY, WORKCULT)

Where:

  • OLS=overall life satisfaction (a proxy for quality of life).

  • NS=Negative Spillover.

  • FIV=A vector of family and individual related variables. These include age, gender, education, occupation, marital status, presence of children under 6 and family income.

  • HWPW=Hours worked per week in all jobs.

  • SSUPPORT=overall supervisor support.

  • JPRESSURE=perceived pressure on the job.

  • AUTONOMY=autonomy on the job.

  • WORKCULT=workplace culture (overall workplace culture).

More concretely, the following hypotheses were tested:

H1: There is a positive relationship between NS and overall life dissatisfaction after controlling for family variables such as family structure and composition.

H2: There is a positive relationship between hours worked per week and NS.

H3: There is a negative relationship between supervisor support and NS. The higher the overall supervisor support for the employee to manage the demands from both systems, the higher the level of resources available and the lower the negative spillover from work to home.

H4: There is a positive relationship between perceived pressure on the job and NS.

H5: There is a negative relationship between autonomy on the job and NS.

H6: There is a negative relationship between workplace culture and NS. The greater the workplace support, the lower the negative spillover.

Methodology

Data and Sample

The data for this research come from the 1997 National Study of the Changing Workforce (NSCW), conducted under the auspices of the Family and Work Institute (Bond, Galinsky, & Swanberg, 1998). The NSCW provides a nationally representative sample of United States workers. Due to their likely control over their schedule, individuals who categorized themselves as exclusively self-employed were deleted from the sample and only waged and salaried workers were investigated. The 1997 NSCW has a total sample of 2877 waged and salaried workers. Of all waged and salaried workers, 2226 are identified as non-Latino Whites and 192 are identified as Latinos (Hispanics in the NSCW). One of the limitations of these data is that the Latinos’ subgroup affiliation is not identified. Under these circumstances, the researchers assume similar characteristics among Latinos.2 It is also assumed that family dynamics remain relatively constant among the possible levels of assimilation of the respondents.

Measures

This study focuses on the determinants of overall life satisfaction and negative spillover. Several predictors of these two variables are explored, including individual and family characteristics as well as job characteristics (see Appendix).

Dependent Variables

The first dependent variable is overall life satisfaction. Respondents were prompted to answer the following question: “All things considered, how do you feel about your life these days? Would you say you feel very satisfied, somewhat satisfied, somewhat dissatisfied, or very dissatisfied?” The scale ranges from 1=very satisfied to 4=very dissatisfied. The second dependent variable is negative spillover from job to home. This variable is an index of five separate questions from the 1997 National Study of the Changing Workforce (Bond et al., 1998). The five questions used to measure negative spillover are presented in Table 1. The index (Cronbach alpha=.86) was created by taking the mean of the five items and it is coded as 1=low spillover to 5=high spillover (Bond et al., 1998). This index variable also functions as a predictor of overall life satisfaction (Table 2).

Table 1 Descriptive Characteristics of Waged and Salaried Latinos and Non-Latino Whites
Full size table
Table 2 Determinants of Overall Life Satisfaction Among Latinos and Non-Latino Whites
Full size table

Independent Variables

The independent variables may be grouped into individual or family-related variables and working condition variables. Individual-level variables include respondent’s age, gender, educational attainment, and occupation. Family-level predictive characteristics are measured via relational marital status, presence of children under age 6 in the household, and family income.

While family and individual-level measures will act as control variables, the focus of this study is on the impact of working conditions on negative spillover. Working conditions are reflected in hours worked per week in all jobs, overall level of supervisor support, pressure on the job, autonomy on the job, and workplace culture (see Appendix). Except for hours worked per week from all jobs, these variables are available as indexes in the NSCW (Bond et al., 1998). Overall supervisor support is measured using 4-point Likert scale categories. An index (Cronbach alpha=.88) was created by taking the mean of the four items. The index ranges from 1=low support to 4=high support. An index of job pressure was created by averaging three questions that employ a 4-point Likert scale (Cronbach alpha=.47). Autonomy on the job (Cronbach alpha=.67), was created by taking the mean of the three items. The index ranges from 1=low autonomy to 4=high autonomy. The final working condition measure, workplace support, is measured via a workplace culture index (Cronbach alpha=.74) that averages the level of agreement to four questions, such as “there is an unwritten rule at my place of employment that you can’t take care of family needs on company time.”

Analyses and Results

Characteristics of the Sample

Descriptive characteristics of Latino and non-Latino White workers are presented on Table 1. Latino and non-Latino White respondents report the same average overall life satisfaction of 1.8. This number indicates that respondents, on average, assess their level of satisfaction somewhere between very satisfied and somewhat satisfied. Latinos and non-Latino Whites also report an identical average negative spillover of 2.9, which lies midway between low and high spillover.

Individual and family characteristics of Latino and non-Latino White workers reveal some similarities and some differences. Marital status, the presence of children under age 6 in the household, and gender are similarly distributed between Latino and non-Latino White workers. Roughly two-thirds of both samples report residing with a spouse or partner and over one-sixth report having a child under age six in the household. Focusing on gender, we note slightly more males in the Latino sample relative to the non-Latino Whites. This is expected since labor force participation rates for Latino women in the United States are significantly lower than those of non-Latino White women (U.S. Census Bureau, 1997a). While these groups report similarities in marital status, presence of children, and gender, differences are evident in the remaining variables of age, educational attainment, occupation, and income. With respondent ages ranging from 18 to 91, the average age of a Latino and a non-Latino White worker is significantly different at 36.3 and 39.9, respectively. A significant difference in educational attainment between Latinos and non-Latino White waged and salaried workers is also evident. Roughly over a quarter of the Latino sample has less than a high school education, 60.4% report a high school degree or some college and 18.7% report a 4-year college degree or more. The figures for the non-Latino White sample are 5.3%, 62.2%, and 30.8%, respectively. It is important to note that the NCSW data on workers, specifically the Latino subsample, is not representative of the general population in the United States.3 A significant difference in occupation is also found between these two groups. While one-fifth of Latino workers identify themselves as executives, administrators, managers, or professionals, over a third of non-Latino Whites identify themselves as such.4 The general distribution in occupations is consistent with those in the general population. However, the NSCW Latino sample includes a greater percentage of managers and executives (20.6%) than is found in the general population, 14% (Therrien & Ramirez, 2000). Following the unequal distribution in occupations, the distribution of family income between the two groups shows similar disparity. Overall, Latino workers report an average yearly income nearly $12,000 below that of non-Latino Whites.5

Along with describing individual and family characteristics of the sample, working conditions are also included in Table 1. Not surprisingly, similarities and differences between the groups are encountered here. Latinos and non-Latino Whites report working a total of over 45 hours per week at all jobs and also report the same level of supervisor support. Job autonomy also produced identical results for Latinos and non-Latino Whites. Both groups indicate relatively high levels of independence on the job as measured via an autonomy index.

Conversely, the index of job pressure indicates that Latino workers experience significantly lower pressure on the job when compared to non-Latino Whites. Latinos also reported significantly lower levels of workplace support (as measured via workplace culture) than non-Latino Whites.

Regression Analyses

Regression equations (OLS) are used to estimate the empirical models. First, the relationship between overall life satisfaction and negative spillover controlling for family and individual characteristics is investigated (Table 2). Then, the relationship between negative spillover and working conditions is determined, also controlling for family and individual characteristics along with work related variables (Table 3).

Table 3 Determinants of Negative Spillover from Job to Home Among Latinos and Non-Latino Whites
Full size table

Negative spillover from work to home is undesirable in large part because it leads to lower levels of life satisfaction. Preliminary analysis demonstrated a significant correlation between negative spillover and overall life satisfaction for Latinos (r (191)=.42, p<.001) and non-Latino Whites (r (2225)=.33, p<.001). Lower spillover from the job to the home is associated with higher overall life satisfaction. In order to confirm the significance of negative spillover on overall life satisfaction while ensuring that this relationship is not working via individual or family characteristics, regression techniques (OLS) were followed where overall life satisfaction was examined holding individual and family characteristics constant. After controlling for individual and family characteristics (e.g., age, gender, education, occupation, marital status, having a child under age 6, income), negative spillover from work to home maintains its significant association with lower measures of overall life satisfaction for both Latino and non-Latino White workers (see Table 2). The regression equation is significant for both Latinos (F[8,164]=6.52, p<.001, R 2 =.21) and non-Latino Whites (F[8,2003]=45.86, p<.001, R 2 =.15).

While negative spillover is the most important predictor of overall life satisfaction among both Latinos and non-Latino Whites, the importance of gender, income, occupation, and marital status on overall life satisfaction differs between these two groups. Among Latinos, women are significantly more likely to express lower levels of life satisfaction relative to men. While gender does not significantly impact overall life satisfaction among non-Latino Whites, age, marital status, occupation, and income do. Younger and married or cohabiting non-Latino Whites show higher overall satisfaction than their older and single counterparts. Non-Latino White workers with higher income and who are in executive, administrative, managerial, or professional occupations have higher levels of overall life satisfaction compared to those with lower income and other occupations. Conversely, marital status, occupation, and income are not significantly associated with overall life satisfaction among Latinos.

Table 3 provides estimates of the association between individual, family, and working characteristics and negative spillover from job to home. The regression equation is significant for both Latinos (F[12,131]=5.74, p<.001, R 2=.29) and non-Latino Whites (F[12,1770]=54.41, p<.001, R 2=.27). Among Latino workers, the only significant individual or family characteristic associated with negative spillover is gender. Relative to Latino men, Latino women are significantly more likely to report negative spillover from work to home. Focusing on non-Latino Whites, several individual and family characteristics including age6, gender, having children under age 6, and marital status significantly impact negative spillover. Among non-Latino Whites, younger workers, women, and workers with children under age 6 are significantly more likely to experience negative work to family spillover relative to older workers, men, and workers with no children under age 6. Moreover, individuals residing within a marital or cohabiting union are significantly more likely to report negative spillover between work and home compared to single workers.

Working conditions, including hours worked per week for all jobs, supervisor support, job pressure, and autonomy of the job have a significant impact on negative spillover in both non-Latino White and Latino workers. However, the significance of workplace culture is exclusive to non-Latino Whites. This finding supports prior research by Maume and Houston (2001) who expected those working within a family supportive work culture to indicate lower levels of work to home spillover.

Discussion

This study focuses on the relationship between negative spillover (NS) from work to family and overall life satisfaction as well as the determinants of negative spillover among Latinos and non-Latino Whites in the United States. The family ecosystem approach provides the framework to illustrate the interaction between families and the different systems in their environment. It also helps in explaining how workers from two different ethnic groups confronting different life circumstances may manage their resources in such a way that their outcomes turn out to be similar.

Limitations

Prior to the conclusions and recommendations of this study, some notable limitations need to be addressed. First, the NSCW data rely on self-reported information with all its inherent limitations as recognized by Schwarz (1999). For example, these data appear to show an inflated overall life satisfaction which leads to a skewed distribution. However, the authors deem that including this variable in the analysis and its relationship to negative spillover contributes to the importance and rationale for this study. Second, the data only allow for the examination of the input and output parts of the family system and not the throughput or transformation process which could better explain how families manage their resources in different cultural environments. While the authors infer that sociocultural variables play a significant role in coping with work to family spillover, further analysis is necessary to test these assumptions. Lastly, the Latino sample in the data is relatively small and may not be representative of the Latino population of the United States as explained in the methodology section.

Conclusions and Recommendations

Despite these limitations, this study makes a contribution to the literature by focusing on Latinos, an underrepresented population in work-life research (Grzywacz et al., 2002; Klein, 1999). The data can be interpreted as evidence that Latino workers seem to follow different strategies (throughput) when managing work–family interaction compared to non-Latino White workers. The average Latino worker in the study has a lower income, lower educational attainment, and is more likely to be a blue collar worker when compared to the average non-Latino White. However, Latino and non-Latino White workers are equally satisfied with their lives and exhibit similar levels of negative spillover from work to family. These findings confirm the equifinality principle of the family ecosystems approach discussed earlier in this paper.

The data also show some differences in the determinants (inputs) of negative spillover (output) between the two groups considered in this study. For non-Latino White workers, individual and family variables are significant determinants of negative spillover after controlling for work-related variables. Among the non-Latino White sample, younger individuals, females, and individuals with a spouse and with young children in residence seem to experience higher levels of negative spillover compared to older individuals, males, singles, and those without young children at home, respectively. These findings are consistent with prior research (Elliot, 2003; Grzywacz et al., 2002; Hughes et al., 1995; Maume & Houston, 2001). In the Latino sample, however, the variables of age, marital status, and presence of young children do not significantly determine negative spillover. Although the data used for this study do not allow for the direct examination of the dynamic interaction among sociocultural variables and negative spillover, a possible explanation for these findings may be that familism is providing Latino workers with higher support from extended family members and the community (Becerra, 1998; Sanchez-Ayendez, 1998; Suarez, 1998; Velez-Ibañez,1996). In other words, the presence of children, age, and marital status are mitigated by community and family networks. This is consistent with Dohan’s (2003) research on two Mexican-American barrios in California. Dohan found that the limited financial resources of families in these communities forced them to rely on their extended kin for child-care, transportation, housing, and social connections. This additional support allows workers to cope with their demands from work more easily even if they do not have a partner or they have to take care of small children. As such, Latino family and community cohesiveness may serve as buffers against negative spillover from work to home and may contribute to the level of satisfaction in their lives. Individualism, on the other hand, is probably valued more highly by non-Latino White workers thereby contributing to lower levels of family support and thus higher levels of negative spillover, especially for younger females with a spouse and young children in residence.

Further, it is noteworthy to recognize the importance of gender for both groups. Latinas and non-Latina Whites have higher levels of negative spillover from work to home compared to the men in both groups. It seems that even employed women, regardless of their ethnicity, are expected to carry the responsibility for family services such as cooking, cleaning, and caring for other family members thus leading to the significance of gender as a determinant of negative spillover for both groups. It is recognized that familism among the Latino population may constitute an additional burden for the family members providing the support, usually females, which tend to report lower levels of overall life satisfaction when compared to Latino men.

When work-related variables are considered, total number of hours per week worked in all jobs, levels of supervisor support, job pressure, and job autonomy are significant predictors for both Latino and non-Latino White workers, as well as work culture on the job for non-Latino Whites. These variables are under the control of the employer and may be addressed through work policies and more comprehensive family-oriented public policy. According to Wallen (2002) “...work requirements and conditions may significantly affect workers and their families. There are a number of changes an employer can make to reduce role conflict and role overload in employees” (p. 28). According to the current analysis, these changes include, but are not exclusive to, decreasing weekly work hours, training supervisors to be more sympathetic, decreasing the amount of job anxiety, granting workers greater independence, and providing an overall work environment that recognizes and supports workers’ family responsibilities. These policy changes translate into, among other things, having realistic work expectations, acknowledging a job well done, being fair and accommodating to the workers non-employment responsibilities, providing ample time and independence for accomplishing work tasks, and having a workplace climate that encourages, and does not penalize, workers for trying to balance their work and life spheres.

For employers interested in the well being of a diverse workforce, this study shows evidence that diverse groups of workers can benefit from family-oriented policies which support the lives of employees outside their workplace. For practitioners and work-life researchers, this study notes that homogeneous non-Latino White samples cannot be generalized to include the experience of Latino workers. The workplace experiences of these two groups are significantly different, amplifying the need to focus on workplace policies that encompass the cultural diversity of the workforce and the needs of all workers. Further study is recommended to investigate Latino families’ resource management practices and the support they receive, as well as the demands placed on them from their extended families and their communities. It is also important to find out why under such adverse life circumstances, Latino levels of life satisfaction and negative spillover are no different from the non-Latino White workers and why with similarly unfavorable life conditions, African American and Latino spillover do not parallel (Grzywacz et al., 2002). In depth investigation, possibly using ethnographic techniques of families in predominantly Latino communities may provide some answers to these questions.

In order to pursue this research agenda, there is a need for data collection from representative national samples of the different Latino groups in the U.S. for quantitative, as well as qualitative, analysis. Future data should help address questions left unanswered by this study. Specifically, do family networks and the idea of familism facilitate the management of home and work for Latino families? Why are Latinas more adversely affected in their efforts to balance work and life relative to Latino men? Similarly, another issue to be addressed is why workplace culture, despite significantly poorer work environments, does not contribute to the variation in negative spillover from work to home for Latino workers.

Notes

  1. 1.

    As of the 2000 Census, the largest country-specific groups of Latinos in the US were comprised by 66% Mexicans, 9% Puerto Ricans, and 4% Cubans (Therrien & Ramirez, 2000). The remaining 21% of Latinos report Central- and South-American origins (14.5%) or other Hispanic origins (6.4%). Mexicans, Puerto Ricans, and Cubans have distinguishable migration characteristics. Some Mexicans are the descendants of original inhabitants of the Southwest before the territory became part of the US. Others have migrated for economic reasons and stay in the US as legal or illegal workers at different historical periods. Puerto Rico, a Spanish colony, became part of the US after the Spanish–American War of 1898. In 1917, Puerto Ricans became citizens of the US and have maintained a dynamic circular migration with the US ever since. Cuban migration to the US was motivated mainly by political reasons after Fidel Castro’s rise to power in 1961.

    Despite the migratory differences among these groups, they share similar socio-cultural systems including family values and cultural beliefs. They also share a common heritage in terms of language, Spanish, and religious ideology, Catholicism (Mindel et al., 1998). According to Becerra (1998), “studies show that Mexican families tend to live near relatives and close to friends, have frequent interaction with family members, and exchange a wide range of goods and services that include babysitting, temporary housing, personal advice, nursing during time of illness, and emotional support” (cited in Muller, et al., 1985, p. 161). Becerra (1998) concludes that “familial solidarity among Mexican Americans is not just a stereotypical ideal, but a real phenomenon” (p. 161). The same is also reported for Cubans and Puerto Ricans in Suarez (1998) and Sanchez-Ayendez (1998), respectively.

  2. 2.

    The only proxies available for acculturation are whether the respondent’s parent(s) were born-US citizens and whether English is the main language spoken in the home. Additional regression analyses among Latinos were conducted to determine if those respondents who answered “yes” to these questions demonstrate a significant difference in spillover relative to those who answered “no”. Although 41% of Latinos report having at least one non-US citizen parent at the time of their birth and 26% report a language other than English as the main language in the home, these variables are not significant predictors of negative spillover among Latinos.

  3. 3.

    While the 1997 Current Population Survey indicates that 54% of Latinos over the age of 25 in the US have a high school degree or above (U.S. Census, 1997b), the NSCW reports that 79% of Latinos over age 25 have more than a high school diploma. For non-Latino Whites the figures are 86.3% versus 94.8%, respectively. On the other end of the spectrum, 10.3% of the Latino population over the age of 25 report a college degree or greater, relative to 21.3% of the NSCW Latino population. For non-Latino Whites, the figures are 26.2% versus 33.1%, respectively (U.S. Census, 1997b). The authors acknowledge that the difference in educational attainment could be the result of the NSCW’s focus on waged and salaried workers, and not the general population.

  4. 4.

    When additional occupational categories are examined, we noted a greater concentration on Latinos in lower-skilled, lower-waged occupations. For instance, slightly over one third of the Latino sample is occupied in production, operation, or repair work, relative to 24% of the non-Latino White sample.

  5. 5.

    Reported family incomes exceeding $200,000 were classified as outliers and were deleted from the sample. This translated into four Latino and 45 non-Latino White individuals, or just over 2% in each sample.

  6. 6.

    Age square was tested for both Latinos and non-Latino Whites. The variable is significant for Non-Latino Whites but not for Latinos.