Although generally encouraged by many organizations, voice is not a required responsibility. Whether to speak up or not totally depends on individual preference. Researchers have explored factors that can attenuate the barriers of speaking up and inspire employee engagement. Typically, voice antecedents include situational factors and individual factors. As for situational factors, organizational climate and leadership improve psychological safety and sense of job meaningfulness, which in turn promote employee voice
. As for individual factors, employees with a proactive personality or highly extroverted ones perceive voice as an opportunity to serve their interests and to achieve organizational goals
. Employees’ abilities, such as communication competency and EQ, facilitate the exchange of ideas, leading to voice behavior
2.5. The Mediation Role of Self-Efficacy
From a psychological perspective, self-efficacy indicates individuals’ belief in their ability to perform a certain behavior in a given situation
[47]. Individuals with a high level of self-efficacy are inclined to take initiative, and individuals with a low level of self-efficacy have a tendency to withdraw
[47]. Individuals’ self-efficacy has been described as one of the most prominent, common predictors of work outcomes in organizations, including proactive
[48] and innovative behaviors
[49]. It also determines employees’ expended efforts and persistence in the face of challenges
[47].
The salient role of self-efficacy in voice has been reported in the literature, e.g., in
[50]. Employees low in self-efficacy doubt their capability to fulfill a task and therefore are reluctant to speak out; in contrast, employees high in self-efficacy are confident in the validity of their opinions and therefore speak their voice more effectively
[51]. Additionally, because of the intention to change the current situation, voice is perceived as a risky behavior
[41]. Self-efficacy empowers individuals to bear the risk
[52].
Culturally intelligent employees, while interacting with people from diverse cultures, realize the appropriateness of their communication and behavior, easily overcome cultural barriers, and smoothly adapt to a particular role expectation
[53]. Thus, their high-level of CQ is likely to help them to fit the job demands in the multicultural work setting, which enhances their general self-efficacy. Based on the results of prior research on the link between self-efficacy and voice, and demands-abilities fit theory, we consider that CQ allows hospitality employees to excel in the workplace, increasing their confidence in what they are doing. This sense of confidence subsequently encourages hospitality employees to actively express their thoughts. Thus, we theorize that self-efficacy is an underlying mechanism through which the favorable effect of CQ on voice occurs.
Hypothesis 4. Hospitality employees’ CQ is positively related to voice through self-efficacy. That is, hospitality employees’ self-efficacy mediates the relationship between CQ and voice.
Moreover, given H3 (CQ → voice → satisfaction) and the relationship just mentioned (CQ → self-efficacy → voice), we theorize a sequential mediation path from CQ to satisfaction via, first, self-efficacy and, then, voice (CQ → self-efficacy → voice → satisfaction) as follows:
Hypothesis 5. Hospitality employees’ self-efficacy and voice sequentially mediate the relationship between CQ and job satisfaction.
3. Methodology
3.1. Data
We collected data from employees working for restaurants in three cities (Los Angeles, Honolulu, and Seattle) in the western United States. The participating restaurants are full-service restaurants where customers are seated and the wait staff take orders and bring food and drinks to the customers at tables. The type of restaurants ranged from casual dining to fine dining. Quick service restaurants were excluded from the sample. Given that the focus of this study is on CQ and voice, establishments providing employees with an opportunity to make longer interactions with a variety of diners and in a more complex work environment, including colleagues from other cultures, are deemed more appropriate. Restaurant owners or managers who approved the data collection are familiar with the authors’ hospitality programs or alumni of the programs. In the cover letter of the survey, we explained the purpose of the study and guaranteed respondents’ anonymity. The respondents who were born and raised or have lived in the US for an extensive time as a citizen (without any language barrier) were qualified for this study. After distributing 500 questionnaires, we collected 285 usable ones (response rate: 57%). Most of the respondents were either servers or bartenders. Of the 285 respondents, 53% (N = 151) were women and 41% (N = 118) were men. They ranged in age from 18 to 63 years with a mean of 29.2. On average, respondents worked for 4 years in the restaurant industry and 2.1 years at their current restaurant.
3.2. Measures
We adopted Ang et al.’s scale to assess CQ. It comprises four sub-dimensions of CQ: metacognitive (4 items) (e.g., “I am conscious of the cultural knowledge I use when interacting with people of different cultural backgrounds”), cognitive (6 items) (e.g., “I know the cultural values and religious beliefs of other cultures”), motivational (5 items) (e.g., “I enjoy interacting with people from different cultures”), and behavioral (5 items) (e.g., “I change my nonverbal behavior when a cross-cultural situation requires it”)
[13]. To assess voice, we utilized the scale (6 items) developed by
[25]. Their scale focuses on voice as an extra-role behavior while in-role behaviors are expected or required behaviors. As this study concerns voluntary expressions that may promote the efficient operation of hospitality businesses (e.g., “I speak up in the group with ideas for new projects or changes in procedures”), the extra-role attribute of voice suits better for the proposed model.
We assessed employees’ self-efficacy (10 items), as suggested by Schwarzer and Jerusalem (e.g., “I am confident that I could deal efficiently with unexpected events”)
[54]. Job satisfaction was assessed with the scale (3 items) (e.g., “In general, I like working at my organization”) developed by Lawler et al.
[55]. All measurement items were rated on a 5-point Likert-type response scale: 1 = strongly disagree; 5 = strongly agree. Lastly, we inquired into respondents’ demographic and job characteristics: gender, age, job position,
and job tenure with the current employer, and industry tenure.
3.3. Statistical Analysis
The hypotheses were tested with partial least squares structural equation modeling (PLS-SEM). Unlike covariance-based SEM (CB-SEM) that aims to minimize the difference between the estimated theoretical covariance matrix and the observed sample, PLS-SEM focuses on maximizing explained variance in dependent variables, thereby increasing predictability of the model
[56]. In addition, PLS-SEM is desired for this study because: (1) the sample size is relatively small and (2) the data show non-normal distribution, which is typical of most social science studies
[56].
4. Results
4.1. Descriptive Statistics and Correlations of Study Constructs
Before testing the hypotheses, we examined the descriptive statistics of study variables and correlations among them. Table 1 shows means, standard deviations, correlations, and coefficient alphas. The mean of CQ represents the average score of four sub-facets (as a higher-order construct). The mean values of CQ, self-efficacy, voice, and job satisfaction are 3.54, 4.00, 3.64, and 3.91, respectively. CQ has a positive correlation with self-efficacy (r = 0.42, p < 0.01), voice (r = 0.44, p < 0.01), and job satisfaction (r = 0.39, p < 0.01). Self-efficacy has a positive association with voice (r = 0.49, p < 0.01) and job satisfaction (r = 0.45, p < 0.01). Voice has a positive correlation with job satisfaction (r = 0.39, p < 0.01).
Table 1. Descriptive statistics and correlations among the study variables.
Notes: N = 285. CQ, cultural intelligence. Coefficient alphas are shown in parentheses along the diagonal. Gender:
1, male; 2, female. Industry tenure: Months working in the restaurant industry. ** p < 0.01, * p < 0.05.
4.2. Measurement Model
CQ, self-efficacy, voice, and job satisfaction are latent constructs with reflective indicators. Factor loading values for each latent construct were assessed to establish convergent validity. All item loadings showed acceptable values ranging from 0.61 to 0.95
[57] (Table 2). Average variance extracted (AVE) is another barometer for convergent validity. It was shown to be greater than the recommended value of 0.5 for all latent constructs except for CQ. CQ is a second-order construct with four sub-facets. Although the AVE of CQ (0.45) is slightly below 0.5, this score is acceptable to support convergent validity
[57].
To test the psychometric properties of these constructs, we evaluated reliability and validity values in the measurement model
[56]. As for reliability, we checked Cronbach’s alpha as well as composite reliability. All latent constructs in the model demonstrated a value greater than the threshold of 0.7
[57][58], ensuring their internal consistency. In addition to convergent validity, discriminant validity was evaluated with the heterotrait-monotrait ratio (HTMT). All latent constructs showed HTMT values of less than 0.85, confirming discriminant validity
[59]. Reliability values and HTMT ratios are presented in Table 3.
Table 2. Measurement items and standardized factor loadings.
Note: All factor loading values are significant (p < 0.05).
Table 3. Reliability and validity of the study variables.
Note: The values under the diagonal show the HTMT ratio.
4.3. Common Method Bias
This study used a self-administered survey and collected data at one point in time. This method may introduce systematic error variance, known as common method bias, in the cross-sectional research design
[60], such as in this study. Min et al. recommended procedural remedy and statistical control for hospitality scholars to avoid the detrimental effect of common method bias
[60]. During data collection, participants were guaranteed anonymity and informed of the use of their data solely for academic research purposes. This procedural remedy often mitigates the concerns about common method bias
[60]. Moreover, following Kock’s logic, we adopted a full collinearity test to detect potential common method bias in a PLS-SEM model; when a variance inflation factor (VIF) shows a value greater than 3.3, common method bias is judged to exist
[61]. Our results showed estimations of VIF ranging from 0.08 to 1.51, revealing that common method bias is not an issue in the proposed relationships.
4.4. Structural Model and Hypotheses Testing
We assessed the structural model after the measurement model. In SmartPLS 3, bootstrapping 2000 subsamples were applied to test the proposed hypotheses. Figure 2 shows the results of parameter estimates and corresponding significance levels. CQ appears to be a salient predictor of hospitality employees’ voice (β = 0.27, p < 0.01) and job satisfaction (β = 0.31, p < 0.01). Thus, H1 and H2 are supported. Next, we tested the significance of a series of indirect paths: (1) from CQ to job satisfaction via voice (CQ → voice → job satisfaction), (2) from CQ to employee voice via self-efficacy (CQ → self-efficacy → voice), and 3) a sequential mediation path (CQ → self-efficacy → voice → job satisfaction). Mediation tests confirmed that CQ has a significant, positive indirect effect on job satisfaction via voice (β = 0.08, p < 0.01) and a significant, positive indirect effect on voice via self-efficacy (β = 0.17, p < 0.01). The sequential mediation effect of CQ on job satisfaction through self-efficacy and then voice was also significant (β = 0.05, p < 0.01). In summary, CQ improves self-efficacy, which leads to a greater chance to voice and ultimately increases job satisfaction. H4, H5, and H6 are, therefore, supported.
Figure 2. Parameter estimates of the research model. Notes: MetaCQ, meta-cognitive CQ; CogCQ, cognitive CQ; MotiCQ, motivational CQ; BehCQ, behavioral CQ. ** p < 0.01.
Some scholars argue the usefulness of analyzing the effect of each sub-facet of CQ, e.g.,
[62]. In line with this view, we conducted additional analyses to see if there are any major differential effects of CQ sub-facets on three outcomes—self-efficacy, voice, and satisfaction (see the results of regression in Table 4). Of the four sub-facets, three (meta, cognitive, and motivational) were generally found to be related to these outcomes. In addition, note that the direct effect of CQ on voice remains significant even after the inclusion of the mediator, self-efficacy. This indicates a partial mediation rather than full
[63]. All other indirect paths showed similar results, indicating partial mediations.
Table 4. Effects of CQ sub-facets on self, efficacy, voice, and job satisfaction.
Note: β, standardized beta coefficient; MetaCQ, meta-cognitive CQ; CogCQ, cognitive CQ; MotiCQ, motivational
CQ; BehCQ, behavioral CQ. ** p < 0.01. * p < 0.05. † p < 0.10.
5. Discussion
5.1. Theoretical Contribution
Based on person-environment fit, specifically demands-abilities fit, this study demonstrates that domestic contact employees’ CQ helps to express work-related opinions and make employees feel happier about their jobs in a diverse work environment. This desirable function of CQ works directly, as well as indirectly, through improved employees’ self-efficacy. The results of this study make the following theoretical contributions. First, it contributes to the voice literature. Prior studies have examined the factors that can influence employees’ voice behavior and have suggested a few personal (e.g., Big Five) and environmental/situational factors, e.g.,
[31][32]. In recent years, thanks to the growing interest in CQ, a couple of studies have revealed a glimpse of the relationship between CQ and voice through a mediation process different from this study
[42][43]. These prior studies targeted non-national employees working in other countries or migrant employees who had just moved to other countries without proper language skills; given their challenging position in a foreign land, it makes perfect sense that their leaders play critical roles in promoting their voice
[42][43].
Limiting the sample to a particular group of people impairs the generalizability of extant research
[64], because it is questionable whether or not the results may be applicable to a wide range of subjects. Unlike prior studies, we focused on domestic employees that are fully comfortable in their own countries and examined their qualifications for the job rather than the leaders’ role. The findings of this study add to the body of person-environment fit theory by conceptualizing CQ as a critical ability that contacts employees should possess to be successful voicers in domestic work environments such as hospitality. This relationship occurs through increased self-efficacy. To the best of our knowledge, this study is the first to explore the relationship between CQ and self-efficacy, more specifically, self-efficacy as a (partial) mediating mechanism between CQ and voice.
Broadly speaking, this study notes CQ as another intelligence just like EO or IQ that may serve as a source of workers’ self-efficacy in today’s environments. Because of globalization necessitating encounters with guests from other parts of the world, not to mention an increasingly diverse workforce, CQ is expected to play a more and more important role in employee confidence to perform their job better. CQ sub-facets that are mental activities-related contribute to the increase in employees’ self-efficacy (motivational CQ) and voice (meta, cognitive, and motivational CQs) rather than behavioral CQ. These results seem to suggest that mental sensitivity to feel the cultural difference from their own and willingness to learn about other cultures are true drivers of employee self-efficacy and voice behavior, as opposed to merely mimicking cultural behaviors. In other words, only when employees can comprehend and enjoy other cultures in service encounters does their perception of being capable at work increase. Overall, the result of this study serves as added evidence to support long-lasting, versatile person-environment fit theory.
5.2. Managerial Implications
This study has several practical implications for hospitality operators. The practitioners should strive to increase the organizational level of CQ because hospitality employees often encounter customers from other cultures and countries and work with multicultural or diverse co-workers. The organization may acquire CQ by recruiting employees with high levels of CQ. Recruiters could ask job applicants questions pertaining to direct (e.g., international trips) or indirect (e.g., reference groups such as friends or family members) cultural experiences and then ask what actions they have taken or will take in a variety of cultural scenarios.
It may not always be feasible to find such qualified applicants. The recent study by Min et al. argues that CQ can be developed through education
[64]. It will be wise for hospitality practitioners to provide regular cultural training to their contact employees to enhance their CQ. Some alternative options are available to improve employees’ cultural awareness and knowledge at minimal cost. For example, the organization can celebrate the major holidays of other countries (e.g., Lunar New Year, common to many Asian countries) through socials, or invite employees to collaborate with colleagues from other cultures on short-term group assignments or projects.
Although employee voice is recognized as a critical resource for a successful business
[43], business owners or practitioners are often ignorant of how to advocate employee voice. Managers, perceived as the embodiment of the organization
[65], should sincerely welcome and appreciate contact employees’ voice behavior. It may be great for superiors to share employee ideas adopted by the organization in the past. It sends an optimistic message to contact employees that their voice is least likely to be dismissed or neglected by their superiors.
In addition, any novel ideas should be frequently recognized—perhaps nominated for the “idea of the month” or the “idea of the year”—through internal company sites or newsletters. If possible, a small reward can be given to the employee who proposed the idea. It is also imperative to respond to employee opinions that are not accepted by the organization so that employees learn that the organization gives a fair amount of consideration to all of their suggestions. Just as hospitality practitioners tirelessly seek customer voice to improve the operation, employee voice should be taken just as seriously. The practitioners must keep in mind that self-efficacy is a key to encourage contact employees’ voice behavior. Employees’ self-efficacy is likely to increase through diligent, sincere responses. In conclusion, if hospitality companies wish to gain a competitive edge in a fast-changing global market, they must understand that the voice of employees with high levels of CQ deserves to be heard.
5.3. Limitations and Directions for Future Research
Given that this study was conducted in casual or fine dining restaurants, it is recommended to validate the findings in other hospitality or service sectors. It will also be interesting to conduct comparative studies to reveal which intelligence plays a more substantial role in voice behavior (e.g., IQ vs. EQ vs. CQ) in a variety of hospitality or service sectors. IQ may serve as basic intelligence for all types of work performance, including voice behavior, whereas CQ or EQ may be required for career success in certain job environments such as hospitality. In addition, it is plausible to examine interactions between different kinds of intelligence. Cote et al. argue that EQ compensates for IQ
[66]. They reveal that employees with low IQ are able to perform their tasks if they are emotionally intelligent. We predict similar results among employees with high CQ, especially in a multicultural work environment. CQ and IQ become compensatory; in other words, CQ becomes a stronger determinant of job performance as IQ decreases.
CQ enables people to perceive and understand cultures that are different from their own. Through interactions with customers of other ethnicities or from foreign countries, employees absorb new knowledge. This accumulated new knowledge may help them to see issues with fresh eyes. Employee voice can be promotive or prohibitive
[52]. Promotive voice offers suggestions to improve work practices; prohibitive voice is intended to prevent harmful events from occurring. Researchers may want to examine the relationships between CQ and these two types of voice. It is predicted that a CQ that is more receptive to a new culture encourages promotive voice rather than prohibitive, possibly stimulating innovative thoughts, which is critical in an increasingly volatile economic climate we experience.
6. Concluding Remarks
This study provides compelling evidence of the positive influence of CQ on employees’ voice behavior and job satisfaction, with self-efficacy serving as a mediator between CQ and voice behavior. Drawing upon the theoretical framework of demands-abilities fit, this study makes a valuable contribution to the extant literature by highlighting CQ as an essential individual capability in the hospitality industry. This study generally underscores the importance of hospitality employees’ knowledge and understanding of other cultures and the necessity of cultivating employees’ CQ. Thanks to CQ, hospitality employees are likely to feel more confident in daily operations. The organization, filled with highly culturally intelligent employees, can create a more conducive work environment, furthering each and everyone’s work engagement and job satisfaction. It is our hope that hospitality practitioners make the best use of employee voice for their organizations’ sustainable success in today’s fast-changing business environment, particularly while going through the COVID-19 pandemic.