Job satisfaction has been the subject of over 11,000 studies (Judge, Parker, Colbert, Heller, & Ilies, 2002). Most of this research has been done in the United States of America (USA), with some exceptions (e.g., Campbell & Campbell, 2003; Huang & Van de Vliert, 2003; Liu, Borg, & Spector, 2004; Thomas & Au, 2002). Tsui, Nifadkar, and Ou (2007), in their review of cross-national organizational behavior research published in the most recent decade, found that of 93 studies in top journals in the field, only eight focused on job attitudes and nine on job behaviors; the authors note that, given the escalating interest in cross-national research to meet the needs of increasingly global organizations, more research is needed.

Job satisfaction is an aggregate multidimensional construct, which means that it is an aggregate or algebraic function of interrelated attributes or dimensions, typically called facets (Law, Wong, & Mobley, 1998; Locke, 1969, 1976). Conceptually, global job satisfaction is equivalent to the sum of individual satisfaction ratings with job elements (Locke, 1976). Thus, correct identification of the “parts” of job satisfaction is important to accurate specification and measurement of the “whole” and to understanding the relationship of job satisfaction to outcome variables like retention (Rothausen, 1994; Scarpello & Campbell, 1983). However, only limited research has identified facets beyond those in the most commonly used measures, which were developed in the USA.

In this study, we bring together these two under-researched streams within job satisfaction research—cross-national and facet identification—using concepts from the cross-national, cross-cultural, job satisfaction, work-life, and turnover literatures. We examine the impact of nationality on the relationship of facet satisfactions to global job satisfaction and intention to quit, and explore additional facets that have rarely been measured. This research has implications for the complete measurement of the multidimensional construct of job satisfaction in cross-national research as well as for cross-cultural management practices.

Overall job satisfaction has been measured with globally worded items or as an average of satisfaction with separate facets of the job. Scholars use one or the other of these types of measures depending on their particular research purpose and goals. Although there is empirical evidence to suggest that both global and sum-of-the-facets measures tap the same underlying construct (Judge et al., 2002), job satisfaction theory (Law et al., 1998; Locke, 1969, 1976) and other research (Rice, Gentile, & McFarlin, 1991; Rothausen, 1994) demonstrate that facet satisfactions relate to global satisfaction and other outcomes differently.

The two most used measures of job satisfaction—the Minnesota Satisfaction Questionnaire or MSQ (Weiss, Dawis, England & Lofquist, 1967), which measures twenty facets including achievement, advancement, compensation, job security, and responsibility; and the Job Descriptive Index or JDI (Smith, Kendall & Hulin, 1969), which measures satisfaction with the work itself, co-workers, supervision, pay and opportunities for promotion—were both developed in the USA. Although Liu et al. (2004) found that a German job satisfaction instrument based on facets from the JDI and the MSQ was robust in terms of measurement equivalence across eighteen countries representing four cultural groups, there were cultural distance effects; results from the Far East showed lower equivalence to Western Europe than to English-speaking and Latin American nations.

One way to classify facets, or characteristics, of the job is to adopt the theoretical framework of Herzberg (e.g., Herzberg, Mausner, & Snyderman, 1959) and examine satisfaction with intrinsic facets (e.g., challenge, achievement, autonomy, the work itself) versus extrinsic facets (e.g., relationships with supervisors and co-workers, pay, job security, working conditions). Research has shown cross-cultural differences in the importance of intrinsic versus extrinsic facets. For people in more individualist and lower power distance countries, intrinsic facet satisfactions are strongly related to overall job satisfaction; however, intrinsic job satisfaction may “fail to work” in collectivist countries, especially in those with poor social welfare or security systems, where extrinsic facets may be more important (Diener & Diener, 1995; Huang & Van de Vliert, 2003; Kanungo, 1990; Lu, Gilmour, Kao, & Huang, 2006). In both the MSQ and the JDI, intrinsic facets are central to job satisfaction. Although the measures include extrinsic facets, because they were developed in the USA, they may not have explored as full a range of extrinsic facets as may be important in other cultures.

Some researchers in the USA have examined the effects of adding extrinsic facets that are not measured by the MSQ or the JDI. For example, Highhouse and Becker (1993) found that adding employee-chosen facets better explained global job satisfaction, and also noted the same effect adding benefit satisfaction. Scarpello and Campbell (1983) found that adding facets like flexibility explained additional variance in global satisfaction. Taking this further, Rothausen (1994) found that the degree of additional variance explained by added facets may be contingent on work-life variables related to level of responsibility for family members. National context may also shape the manner in which employees experience job satisfaction (Lu et al., 2006). We posit that adding additional extrinsic facets will impact overall job satisfaction differently in high individualist-low power distance and high collectivist-high power distance countries.

In order to provide a conservative test of these differences, we studied countries that, although they are in two different regions of the world—one from the East, one from the West, nonetheless have an unusually close cultural connection—the USA and the Philippines. The Philippines is an Eastern country with a long history of Western colonization, most recently by the USA (Enriquez, 1989). The Philippines was acquired by the USA in 1898 and held for 48 years (Gupta & Kleiner, 2001). It has a largely English speaking population that has adopted many Western style educational and business practices (de Leon, 1987; Galang, 2004; Jocano, 1999; Teehankee, 2004). Both countries are also very heterogeneous and both are low in uncertainty avoidance and high in masculinity (Hofstede, 1991; Teehankee, 2004). These similarities may mute differences between the two nations (Liu et al., 2004), thus making our results a conservative test of differences by nation. Selmer and de Leon (2001: 127) argue that “Americanization has penetrated deeper into (Filipino) society than into any other Asian country.”

These factors lead to similarities in language and ways of doing some things in organizations and society that may be misleading in business management (Galang, 1999, 2004; Teehankee, 2004); the similarities are misleading because underlying the apparent diversity and Anglicization of Filipinos are strong common beliefs and values regarding institutional behavior and other aspects of life that derive from core indigenous culture (Enriquez, 1989; Jocano, 1999). Core values in Filipino culture include the concepts of kapwa, which reflects the unity of the self with others and the importance of relating to others in a shared identity, including a strong focus on identity and relationship with family and kin in a hierarchical way (de Leon, 1987; Enriquez, 1978; Jocano, 1999; Selmer & de Leon, 2001); damdamin, which reflects an emphasis on feelings and emotions, including a high level of emotional sensitivity (Jocano, 1999; Selmer & de Leon, 2001); and dangal, which reflects a moral standard around dignity and integrity which in turn also includes the relational notion of utang na loob or a reciprocal binding when one person helps another (Selmer & de Leon, 2001).

This summary is of course cursory, and there is a danger in lifting such token concepts out of the rich cultural network from which they originate (Enriquez, 1978). However, many Filipino and Western authors who study the Philippines agree that although not the same, these core values translate, in the language of general cultural values research (e.g., Hofstede, 1991), to an emphasis on collectivism and higher power distance (e.g., de Leon, 1987; Galang, 1999, 2004; Jocano, 1999; Selmer & de Leon, 2001; Teehankee, 2004). This puts the Philippines in direct contrast to the USA on these important dimensions; the USA has one of the highest scores on individualism, and has relatively low power distance, in Hofstede’s (1991) research (Teehankee, 2004).

In addition to providing a conservative test of differences in how job satisfaction facets relate to outcomes, there are other factors that suggest comparing the Philippines and the USA in terms of job attitudes is of value. Because of the continuing high level of economic interaction between these two countries (Audea, Teo, & Crawford, 2005) and their strong historical links (Gupta & Kleiner, 2001), findings can have immediate applied implications for cross-cultural management and HRM practices between the two.Footnote 1 In addition, attitude and behavioral intention research on the Philippines is underrepresented (Tsui et al., 2007).

In cross-cultural organizational behavior research, job satisfaction is an important area of study not only in and of itself, but also because of its relationships to other job attitudes and behaviors important to both individual employees and to organizations, such as commitment, absenteeism, turnover, organizational citizenship behaviors, and life satisfaction (Judge et al., 2002). Voluntary turnover is costly to organizations when good performers leave; thus, it has been the subject of intensive research and theorizing by academics and practitioners. Scholars differentiate between process and content turnover research (Maertz & Campion, 2004; Steel, Griffeth, & Hom, 2002). Process addresses issues of how people end up staying in or leaving organizations. Content addresses what makes them stay or go. Influential process theories include the work of Mobley and colleagues (e.g., Mobley, 1977; Mobley, Griffeth, Hand, & Meglino, 1979), and more recently, Lee, Mitchell, and colleagues (e.g., Lee & Mitchell, 1994; Lee et al., 1999). Most of the research done on retention and turnover uses Mobley and colleagues’ theories as a basis (Lee & Mitchell, 1994). In this research, job dissatisfaction has been shown to lead to thoughts of leaving the organization or intention to quit, which leads to identification of alternatives and utility analyses of moving versus staying. When combined with having other more attractive alternatives, this leads to turnover.

The content, or what makes people think about leaving, is dissatisfaction with the job in general or with different facets of the job (Lee & Mitchell, 1994). Recent research reported by Steel et al. (2002) and Nalbantian and Szostak (2004) reveals that levels of satisfaction with job facets not measured in the MSQ and the JDI may be important to retention, including opportunity to learn new skills, work hours, incentives, and supervisor stability.

Judge et al. (2002) call for further research on the relationship of job satisfaction to withdrawal attitudes and behaviors in international contexts, noting that cultural norms and values may affect the relationship between satisfaction and withdrawal behaviors. Satisfaction with different facets of the job may relate differently not only to global satisfaction, but also to intention to quit for different groups of workers with different norms or values (Campbell & Campbell, 2003; Rothausen, 1994; Thomas & Au, 2002). In this study, we examine differences in how facet satisfactions relate not only to global job satisfaction, but also to intention to quit in the Philippines and the USA.

The Philippines can be classified as a vertical collectivist country (high power distance and collectivism) and the USA as a horizontal individualist country (low power distance and collectivism; Thomas & Au, 2002). Horizontal individualists value freedom of choice and individual rewards, whereas vertical collectivists value harmony and conformism over confrontation (Koch & Koch, 2007). Research has shown that intrinsic facets are generally more important in vertical individualist countries because individual achievement and experience is more important in such cultures than are group relations, identity, or respect for hierarchical place (Huang & Van de Vliert, 2003; Thomas & Au, 2002). Because the JDI and the MSQ were developed in the USA, it is likely that intrinsic facets are emphasized in both measures at the expense of facets that may be valued more in a vertical collectivist nation.

In addition, the MSQ and the JDI were developed in the 1960s, at that time when the USA had not only a horizontal individualistic culture, but one in which one spouse, generally the man, worked and provided financial care for the family while the other spouse primarily cared for the home and administered the emotional and physical care of dependents (Williams, 2000). This may have exacerbated the individualistically oriented nature of the measures at the expense of facets valued by those with values toward, and obligations for, the physical and emotional care of others.

In a study of married Hong Kong employees, Aryee, Fields, and Luk (1999) show that the work-family interface is shaped by the collective values of that society. In comparison with the USA, where people see themselves as unique and look after themselves over social group goals, in the Philippines people self-define through belongingness to, and identity with, social groups, especially family and kin (Enriquez, 1989; Markus & Kitayama, 1991; Selmer & de Leon, 2001; Triandis, 1995). Recent GLOBE study researchers also found these differences for both countries (Gelfand, Bhawuk, Nishii, & Bechtold, 2004). Concern for family is one aspect of collectivism in general (Spector et al., 2004), and is especially important to Filipinos (Enriquez, 1989; Jocano, 1999). Filipinos display practices related to altruism, friendliness, generosity, kindness, and caring, and they value personal one-to-one contact more than do Americans (Kabasakal & Bodur, 2004; Selmer & de Leon, 2001). Thus we expect that job satisfaction facets that allow employees to personally meet health, physical, and emotional care needs of family and kin, not just pay for someone else to do this, will be more strongly related to global job satisfaction and intention to quit in the Philippines than in the USA. Two such aspects of the job not measured by the MSQ and the JDI are benefits and flexibility.

Although not measured in the JDI or the MSQ, Scarpello and Campbell (1983) and Rothausen (1994) found support for a satisfaction with flexibility facet, and Rothausen also found that flexibility was more important to workers with more family responsibility. Because benefits help workers care for the health of family members, we expect that satisfaction with benefits will also be more important in societies with higher collectivism.

  1. Hypothesis 1

    Satisfaction with extrinsic facets not traditionally measured will explain a larger proportion of the variance in and be more strongly related to global job satisfaction and intention to quit in the Philippines than in the USA.

Although Huang and Van de Vliert (2003) found that extrinsic facets were equally important to job satisfaction across countries, the finding was counter to their hypothesis. The above arguments as well as the studies reviewed by Huang and Van de Vliert, suggest that patterns found in previous research on the impact of individualism-collectivism on job attitudes will be found in this comparison of the USA and the Philippines; that is, that intrinsic facets will be more related to overall global job satisfaction in the USA than in the Philippines, and that extrinsic facets already measured by the JDI and MSQ will be more related to overall global job satisfaction in the Philippines than in the USA. In addition, we expect that after accounting for overall global job satisfaction, intrinsic facets will be related to intention to quit more in the USA than in the Philippines, and that extrinsic facets will be related to intention to quit more in the Philippines than in the USA.

  1. Hypothesis 2

    Satisfaction with intrinsic facets will explain a larger proportion of the variance in and be more strongly related to global job satisfaction and intention to quit in the USA than in the Philippines.

  2. Hypothesis 3

    Satisfaction with traditionally measured extrinsic facets will explain a larger proportion of the variance in and be more strongly related to global job satisfaction and intention to quit in the Philippines than in the USA.

Job satisfaction is an important construct partially because it impacts the retention of good employees (Mobley, 1977). Above, we hypothesize that satisfaction with different facets of the job will predict global job satisfaction and intention to quit in the USA and the Philippines. The relationship between global job satisfaction and intention to quit may also vary by nation. For example, in a study that compared Hong Kong to New Zealand, Thomas and Au (2002) found horizontal individualists were more likely to respond to low satisfaction with voice and exit, whereas vertical collectivists were more likely to respond with loyalty and neglect. In line with this, Luthans, Zhu, and Avolio (2006) found that job satisfaction mediated the relationship between general attitudes and intention to quit in a USA sample, but not in an Asian sample made up of people from Indonesia, Malaysia, and Thailand.

This suggests that workers in the USA will be more likely to respond to lower job global satisfaction with thoughts of turnover than will workers in the Philippines. In addition, economic factors in the Philippines suggest that even financial professionals in relatively well-paying jobs see poverty frequently, know the risks of losing their jobs, and are less likely to move to other cities within their country in search for jobs (Amante, 1993; Gupta & Kleiner, 2001; Yang, Chao, Choi, & Zou, 2000).

  1. Hypothesis 4

    Global job satisfaction and satisfaction with facets of the job will explain a larger proportion of the variance in and be more strongly related to intention to quit in the United States than in the Philippines.

Method

Sample and procedures

We collected data via survey questionnaires from professional finance workers in the USA and the Philippines. We sent 875 questionnaires to all professional employees of three Minneapolis offices (735) and one Manila office (140) of international accounting and consulting firms. The USA firms were two “Big Six” firms and one other international firm, and the Manila firm was an independent, Philippine-founded “Big Six” affiliate, not affiliated with the USA firms in the study. This sampling follows Van de Vijver and Leung’s (1997) recommendation to make samples as similar as possible on relevant background characteristics; industry, profession, and education levels are similar in these two samples.

The questionnaires contained items pertaining to facet satisfactions, global job satisfaction, intention to quit, and demographics, as well as additional items for another purpose. A cover letter assuring confidentiality and explaining the project was included with the survey, as well as a postage-paid return envelope.

444 usable surveys were returned for a 51% effective response rate, and the final sample consisted of 327 American and 117 Filipino professionals. The response rate in the Philippines was significantly higher (84%) than that in the USA (45%); discussions with HR directors in all four offices indicated that this was likely due to the fact that the Philippine employees did not get surveyed as much as their American counterparts, and so it was a more interesting and novel experience, and additionally, that they are generally motivated to do what is asked of them by reputable outside professionals. To partially address the differing response rates, we compared the Philippines and the USA samples on demographic variables. There were significant differences between the American and Filipino samples for age (32 for the USA, 28 for the Philippines) and sex (46% male in the USA, 23% male in the Philippines; coded male = 0, female = 1). Therefore, we controlled for these two variables in all inferential tests. Nationality was coded Philippines = 0, USA = 1.

In the Philippines, there are several ethnic groups and eight major dialects on which the primary language is based, but English is the second official language, and the country has a largely English speaking population who use American-based English in their education and business organizations (Galang, 2004; Jocano, 1999; Teehankee, 2004). During the survey design stage, we learned that virtually all possible respondents in the Philippines spoke fluent English; therefore, we surveyed Filipino workers in English. We asked the HR director in the Philippines office in which we surveyed workers to read the survey and suggest any changes to accommodate local usage, but she suggested none.

Measures

Global job satisfaction, facet satisfactions, and intention to quit were assessed. A five-point Likert scale ranging from strongly dissatisfied or strongly disagree (1) to strongly satisfied or strongly agree (5) was used for all items. For each scale, responses were averaged across items for each individual.

Overall global job satisfaction was measured with four items similar to the ones used by Hackman and Oldham (1976). Coefficient alpha was 0.91 in the USA sample and 0.75 in the Philippines sample. Although the alpha was smaller in the Philippines sample, 0.75 and 0.91 are both considered acceptable reliabilities. The intention to quit scale from the Michigan Organizational Assessment Questionnaire (Cammann, Fichman, Jenkins & Klesh, 1979) was adapted to assess individuals’ desire to remain with the organization. The scale contained four items. Coefficient alpha was 0.93 in the USA sample and 0.84 in the Philippines sample.

Facets of job satisfaction

Seven separate facets of job satisfaction were assessed for this study to represent classic intrinsic and extrinsic facets as well as two new facets. Five of these seven were MSQ facets, including three intrinsic facets—advancement, achievement, and responsibility, and two extrinsic facets—compensation and job security. Three items for each facet were chosen from the original five items to meet the goal of balancing psychometric reliability and validity with a shorter survey length to motivate higher response rates. The items retained were worded most clearly and were most representative of the underlying construct of interest based on prior research (e.g., Rothausen, 1999). Finally, benefit and flexibility satisfaction were measured with scales used in Rothausen (1999). Advancement satisfaction measured the degree of satisfaction with chances for advancement and growth; coefficient alpha was 0.97 in the USA sample and 0.92 in the Philippines sample. Achievement satisfaction measured the degree of satisfaction with the feeling of accomplishment on the job, being able to do something worthwhile and seeing the results; coefficient alpha was 0.87 in both samples. Responsibility satisfaction measured the degree of satisfaction with the chance to make decisions and use personal judgment on the job; coefficient alpha was 0.84 in the USA sample and 0.73 in the Philippines sample. Compensation satisfaction measured the degree of satisfaction with remuneration relative to the work; coefficient alpha was 0.94 in the USA sample and 0.90 in the Philippines sample. Job security satisfaction measured the degree of satisfaction with the level of certainty that the job will be there in the future; coefficient alpha was 0.91 in the USA sample and 0.75 in the Philippines sample. Benefits satisfaction measured the degree of satisfaction with benefits and their fairness; coefficient alpha was 0.93 in the USA sample and 0.89 in the Philippines sample. Flexibility satisfaction measured the degree of satisfaction with flexibility in scheduling work hours; coefficient alpha was 0.85 in the USA sample and 0.89 in the Philippines sample.

Data analysis

Preliminary data analysis, including Kolmogorov–Smirnov (Lilliefors significance correction) and Shapiro–Wilk tests, indicated that all the variables were normally distributed (Shapiro, Wilk, & Chen, 1968). Hypotheses were tested using correlational and regression techniques. Two separate hierarchical regression models were run to examine additional variance explained by intrinsic, traditionally measured extrinsic, and not typically measured extrinsic facet satisfactions on global job satisfaction and intention to quit for both the USA and the Philippines samples, with gender and age entered as control variables in the first step of the equations. To test Hypothesis 1, we examined the last step of this hierarchical regression, and examined differences between the correlation coefficients of the facet satisfactions with global job satisfaction and with intention to quit for the Philippines and the USA sample (Van de Vijver & Leung, 1997). To test Hypotheses 2 and 3, we examined the second and third steps of the regression equations, and examined differences between the correlation coefficients of the facet satisfactions with global job satisfaction and with intention to quit for the Philippines and the USA sample. To test Hypothesis 4, the final variance in intention to quit was compared between the Philippines and the USA, and an additional regression with an interaction term was run (Van de Vijver & Leung, 1997).

Examination of regression diagnostics, residual histograms, and partial residual plots suggested that regression assumptions were met in both samples, including the absence of collinearity and the presence of linearity and homoscedasticity. We did find some influential data. Thus, two outliers from the Philippines sample were deleted from the regression model for overall job satisfaction, and one outlier was excluded from the regression model for intention to quit.

Results

Table 1 contains the means, standard deviations, reliabilities, and correlations for the facet and global job satisfactions, intention to quit, and demographic variables for the USA and the Philippines samples. Examination of this table indicates that the reliabilities for the measures are acceptable in both samples and the facets are moderately intercorrelated in both samples, as expected, with the one exception being that satisfactions with responsibility and compensation were not significantly related in the Philippines sample. Facet interrelationship patterns were also similar between samples, although one difference in pattern is the strength of the relationship between compensation and benefits satisfactions. This relationship is much stronger in the Philippines sample than the USA sample, and is reflected in the correlation patterns between the two samples of these facets with flexibility and security. Subsequent analysis leads us to believe that the meaning of “benefits” is different in the USA, where it has a specific definition as non-cash compensation, and the Philippines, where the “benefits” of a job may be less distinct from compensation, security, and flexibility.

Table 1 Means, standard deviations, correlations, and reliabilities for the USA and the Philippines.
Full size table

Examination of Table 1 also shows that the facets are correlated with the outcome variables in the directions expected in both national samples, with the one exception being in the Philippines where advancement satisfaction is not related to intention to quit. The Philippine respondents in general report higher satisfaction levels and slightly lower variance in responses. This may be due to the acquiescent response style sometimes found in collectivist cultures and is not unexpected (Johnson, Kulesa, Cho, & Shavitt, 2005; Van de Vijver & Leung, 1997). Given the overall measurement equivalence in MSQ- and JDI-derived satisfaction measures despite slight cultural distance effects, and especially between English-speaking countries, we are confident of the general equivalence of the satisfaction measures (Liu et al., 2004). In addition, there is enough variance in reported levels of satisfaction in both samples to test our hypotheses, and all distributions conform to expectations of normality.

The first hypothesis was that extrinsic facet satisfactions not traditionally measured would explain more variance in and be more strongly related to overall global job satisfaction and intention to quit in the Philippines than in the USA. Examination of Table 2 shows that these facets did not explain any additional variance in global job satisfaction in the USA, but in the Philippines explained an additional 8% (p < 0.001). This is supportive of Hypothesis 1 and the notion that there are additional unmeasured facets which determine overall job satisfaction for workers in some countries. However, examination of Table 3 shows that these facets explained no additional variance in intention to quit in either the USA or the Philippines. Table 4 contains comparisons of differences in the correlation coefficients between facet satisfactions and the two dependent variables in the Philippines and the USA samples. Examination of Table 4 shows that the correlation coefficients of both benefits and flexibility satisfactions with overall global job satisfaction were significantly stronger in the Philippines sample, but that the correlation coefficients of both benefits and flexibility satisfactions with intention to quit were not significantly different in the samples, showing the same pattern as the regression results. Thus, overall, Hypothesis 1 was supported for global job satisfaction but not supported for intention to quit.

Table 2 Hierarchical regression analyses for global job satisfaction.
Full size table
Table 3 Hierarchical regression analyses for intention to quit.
Full size table
Table 4 Correlations between facet satisfactions, global job satisfaction and intention to quit for the USA and the Philippines.
Full size table

The second hypothesis was that intrinsic facet satisfactions would explain more variance in and be more strongly related to global job satisfaction and intention to quit in the USA than in the Philippines. Examination of Table 4 shows that the correlations for intrinsic facets with global job satisfaction were not significantly different between the two samples, but all three were significantly different for intention to quit in the expected direction. Regression results for global job satisfaction, which are presented in Table 2, show that intrinsic facets yielded an R 2 change of 0.54 (p < 0.001) in the USA sample and 0.37 (p < 0.001) in the Philippines sample for global job satisfaction. Regression results for intention to quit, presented in Table 3, show that intrinsic facets explained significant additional variance (3%, p < 0.001) above that explained by global job satisfaction for the American sample, but not for the Filipino sample. Overall, these results support Hypothesis 2 for intention to quit and although not conclusive, show similar patterns for overall global job satisfaction.

The third hypothesis was that traditionally measured extrinsic facets would explain more variance in and be more strongly related to overall global job satisfaction and intention to quit in the Philippines than in the USA. Examination of Table 4 shows that the only significant difference in correlations between these facet satisfactions and the two dependent variables was for security with intention to quit, and that was in the direction opposite than that expected. In addition, examination of Table 2 shows that in the USA sample, these facets explained an additional 3% (p < 0.001) of variance in overall job satisfaction, and 5% (p < 0.01) in the Philippines sample. Examination of Table 3 indicates that these facets explained an additional 1% (p < 0.05) of variance in intention to quit in the USA, while they explained an additional 4% (p < 0.001) in the Philippines. Hypothesis 3 was not supported.

The fourth hypothesis was that global job satisfaction and satisfaction with facets would explain a larger proportion of the variance in and be more strongly related to intention to quit in the USA than in the Philippines. The total sample models shown in Table 3 indicate that an interaction term for nationality and job satisfaction was significant (β = −0.31, p < 0.05). The difference in relationship between global job satisfaction and intention to quit across countries is pictured in Figure 1. In addition, examination of Table 3 shows that although global job satisfaction was strongly related to intention to quit in both samples, all satisfaction measures explained 53% of the variance in intention to quit in the USA but only 29% in the Philippines. These results show support for Hypothesis 4.

Figure 1
figure 1

Intention to quit as a function of the interaction between overall job satisfaction and nationality. The values represented by the dots are 1 SD either side of the mean. The solid line represents the Philippines and the dotted line represents the United States

Full size image

Examination of Table 2 also reveals that, although we did not anticipate it, women in the Philippines sample appear to be more dissatisfied than men, whereas the same effect was not found in the USA.

Discussion

Although global and facet job satisfactions have been the subject of much research, little of it has examined whether the facets that comprise this multidimensional construct are similar across cultures and nations, or whether the facets most commonly measured are equally explanatory in cultures different from the USA. In this sample, the relationships of the parts to the whole of job satisfaction were different in the Philippines than in the USA. Specifically, in the Philippines, but not in the USA, more variance was explained by adding extrinsic facets not measured by the most commonly used job satisfaction measures. It may be that commonly used measures should be modified to include additional extrinsic facets related to collectivist values. Our results also replicate and extend others’ findings that satisfaction with intrinsic facets may “fail to work” in some countries (Huang & Van de Vliert, 2003). Intrinsic facet satisfactions explained significant variance in overall global job satisfaction for both samples, but more in the USA than in the Philippines sample, intrinsic facets were clearly more related to intention to quit in the USA than in the Philippines, and nation significantly impacted how global job satisfaction related to intention to quit.

This study has at least four important contributions to the literature, as well as implications for the practice of management around the globe. First, the traditionally measured facets may explain outcomes better for populations that more closely mirror the population in which the measures were developed. These populations were more individualistic and autonomous than many other populations. In this study, the variance in overall global job satisfaction in the Philippines sample was better explained by adding additional extrinsic facets, whereas this same effect was not found for the USA sample. One explanation for these findings is that having values or responsibilities that emphasize aspects of life other than work may mean that certain facets of the job that facilitate these roles, including flexibility and various types of rewards, are more important, whereas having more individualistic values and fewer responsibilities outside of work may mean that intrinsic facets of the job, such as autonomy and achievement, are more important.

Second, asking which facet satisfactions are more important for global satisfaction and which are more important for intention to quit yielded different answers. For example, examination of Tables 2 and 3 shows that in this American sample, satisfaction with achievement, responsibility level, compensation, and security were all strongly related to global job satisfaction, but only advancement and compensation satisfactions were related to intention to quit after accounting for global job satisfaction. Similarly, in this Philippines sample, achievement, benefits, and flexibility were strongly related to global job satisfaction, but only compensation satisfaction were significant for intention to quit above and beyond global satisfaction.

Third, the results of this study support conclusions drawn by Huang and Van de Vliert (2003) that intrinsic aspects of the job matter more in more individualistic and lower power distance societies. Perhaps more importantly, although Huang and Van de Vliert concluded that extrinsic factors matter in similar degrees across countries, our results show that this may be because the most commonly used facet measures, the JDI and the MSQ—both developed in the USA—have not been updated to include work-family concerns, more collectivistic values, or the concerns of those in poorer nations. We recommend that future researchers continue to explore facets which may be more important to outcomes for people with less individualistic and higher power distance values.

Fourth, our findings add to the body of evidence that suggests that job satisfaction measures and facet models are robust across countries and that how satisfaction relates to intention to quit is robust—that is, predictive—across countries (see Judge et al., 2002 and Liu et al., 2004 for reviews). However, the results of this study and others (e.g., Rothausen, 1994; Thomas & Au, 2002) point out that the relative importance of different facets to global satisfaction and the importance of job satisfaction to intention to quit vary when people have different core values. Our results support Thomas and Au’s (2002) conclusion that in collectivist societies, workers are less likely to respond to job dissatisfaction with thoughts of exit than they are in individualist societies.

Other contributions of this study to the cross-national literature include using a true cross-national sample—that is, workers living and working in the two countries—and surveying working professionals and staff about job attitudes and behaviors. In addition, this study contributes information about job attitudes and behaviors in the Philippines, a relatively less studied country in cross-national organizational behavior research (Tsui et al., 2007). This study also directly compares the Philippines to the USA, two countries that have many on-going transactions and relationships (Audea et al., 2005).

This study has implications for practice. Although Galang (1999, 2004) concludes that HRM practices are transferable to the Philippines from the USA and other Western countries, our findings caution managers to pay attention to differences in practices that affect job facets, as they may relate differently to overall satisfaction and intention to quit in the Philippines. For example, research reported by Selmer and de Leon (2001) shows that Filipino employees who find a lack of concern and cooperation among co-workers are absent more often and that the ability to care for family and kin through “compadre nepotism” is important to Filipinos and may be included in the Filipino understanding of benefits. This may help explain why we found more importance of benefits satisfaction as well as flexibility satisfaction and such a strong relationship between these two facet satisfactions in our Philippines sample. Our findings lend support to de Leon’s (1987: 28) conclusion that an “indigenous style of Filipino management is viable and imperative, which recognizes the salience of collective identities in (organizations).”

In conjunction with other research discussed above, our results suggest that people in vertical collectivist societies may sacrifice their well-being on the job in favor of other rewards of the job. Thomas and Au (2002) found that horizontal individualists were more likely to respond to low satisfaction with exit, whereas vertical collectivists were more likely to respond with neglect. Flexibility and benefits, although perhaps not factors in their intention to quit, will add to employees’ satisfaction and well-being in these countries and can thus be tools for managers to keep workers satisfied overall to capitalize on their loyalty and prevent neglect.

These findings contribute to helping leaders in multinational enterprises develop a better understanding of subtle differences in organizational behavior around the world, which in turn can help their organizations perform better and have better relations with their employees. Even when there are historical and ongoing links, there may be deep cultural differences, and the fit between the organizational approach and the national culture is important to employees’ attitudes toward work (Testa, Mueller, & Thomas, 2003). Specifically, employers may want to explore the impact of facets of jobs and organizations that are more important to those in the horizontal collectivist cultures in which they operate.

Although not anticipated, this study also showed that in this sample, Philippine women were less satisfied with their jobs than were Philippine men, though this pattern was not found in the USA sample. This could be related to the fact that over 56% of Filipino employers preferred men and only 12% preferred women in a survey reported by Selmer and de Leon (2001). Employment of mothers in the Philippines is lower than in the USA (Doan & Popkin, 1993), and research reviewed by Skeen, Paguio, Robinson, and Deal (1988) suggests that husbands in the Philippines feel that most women are better off in the home and that people think less of a man if his wife works, which may impact women’s attitudes. Despite this, however, Burke (2001) studied women’s career experiences and satisfaction in five counties including the Philippines and concluded that women there had similar patterns of job attitudes to women in Canada, Norway, Bulgaria, and Singapore; however, he did not report levels of attitudes. All these authors note that there is little research on gender differences in the Philippines in the work setting. Future research should further explore gender impacts on job attitudes and behaviors in the Philippines.

One limitation of this study is that we examined differences across nations and therefore cannot draw conclusions as to the relative importance of various national factors, such as cultural values and economic conditions, to explaining our findings. In addition to cultural value differences, the USA and the Philippines have different economic environments. The Philippines is characterized as poor, as evidenced by low individual income, high rates of unemployment and underemployment, depressed real wages, and significant wage differentials across labor market segments (Amante, 1993). Huang and Van de Vliert (2003) found that intrinsic job characteristics related more strongly to satisfaction in both high individualist and low power distance nations, but also in more affluent countries and countries with stronger social welfare and security programs. The role of economics and security is powerful; lower order needs, such as material needs, are usually filled for a large majority of the population in affluent countries, so they can “afford” to care about intrinsic facets more (Huang & Van de Vliert, 2003). Although direct measurement of such variables is preferred, most cross-national organizational behavior studies have not included direct measures and yet they do contribute to cross-national organizational behavior knowledge (Tsui et al., 2007; Van de Vijver & Leung, 1997). Future researchers, however, should directly measure cultural values and economic circumstances so that the relative importance of these underlying potential causal factors can be examined.

Another potential limitation of this study is that we collected self-reported data in a single period of time. This prevented us from measuring actual turnover instead of turnover intentions, and introduces the possibility of common method variance, which may inflate the relationship of facet satisfactions to both outcomes of interest. However, the findings of different patterns between the two countries suggest that the differences are powerful enough to be found even with any possible common method variance. Another limitation is that a large number of predictor variables, which is common in facet satisfaction research, combined with relatively small sample sizes, prevented us from using multiple moderated regression techniques to explore the interaction of independent facets on the outcome variables between nations. However, the use of separate models for each nation along with correlational analyses allowed us to interpret the data with confidence. The use of only two countries allowed us to explore the results in a more detailed manner, as has been done by other scholars (e.g., Thomas & Au, 2002), but it, along with the significant historical relationship between these two countries, introduces the question of generalizability. Future research should replicate this study in other cultures and nations.

We hope this study provides a point of departure for more research on job attitudes and behavioral intentions in the Philippines and between the USA and the Philippines and other nations and cultures. This area presents opportunities for future researchers to explore the role and importance of classic intrinsic and extrinsic, as well as previously unconsidered, facet satisfactions to organizational behavior across nations.