Sociocultural Adaptation and Job Satisfaction as Mediators Between Cultural Competence and Intention to Stay Among Vietnamese Workers in Taiwan

Taiwan’s rapid economic growth and resulting labor shortages led to policy openings for foreign workers across several sectors. By 2019, Vietnam constituted the second-largest source of Taiwan’s foreign workers, yet quantitative research on Vietnamese low-skilled contract workers remains limited. Despite regulatory protections, Taiwan faces persistent challenges with migrant workers fleeing jobs, with Vietnamese workers exhibiting higher tendencies to do so. Prior work highlights financial burdens, working conditions, and exploitation as drivers of turnover, but there is little empirical examination of how cultural and social competencies influence intention to stay (ITS) among low-skilled migrants. This study focuses on Vietnamese manufacturing workers in Taiwan to explore whether cultural competence (CQ) is associated with ITS and how sociocultural adaptation (SCA) and job satisfaction (JS) function in this relationship (RQ1). It further tests the moderating roles of local language proficiency (Chinese) and length of residence (LOR) in the links between CQ–SCA and CQ–JS (RQ2). The purpose is to clarify the mechanisms by which cultural capabilities influence retention-related outcomes and to inform policies aimed at improving migrant worker retention.

The study builds on cultural intelligence/competence (CQ) theory, defining CQ as the capability to function effectively across culturally diverse settings and comprising metacognitive, cognitive, motivational, and behavioral components. Sociocultural adaptation (SCA) reflects behavioral competence in handling daily life and interactions in the host culture. Prior research links CQ to SCA and SCA to outcomes such as ITS and turnover intentions. Job satisfaction (JS), defined as a pleasurable emotional state related to job values, consistently predicts turnover intentions and is influenced by cultural adjustment in cross-cultural contexts. Drawing on social exchange theory and Herzberg’s two-factor theory, the study proposes that CQ promotes SCA and JS, which in turn enhance ITS. The study formulates hypotheses: H1 and H2 posit that SCA and JS, respectively, mediate the CQ–ITS relationship; H3 and H4 posit that Chinese proficiency and LOR strengthen the positive association between CQ and SCA; H5 and H6 posit that Chinese proficiency and LOR strengthen the positive association between CQ and JS. Control variables include gender, age, education, marital status, and income to account for alternative explanations for ITS.

Design and sample: A cross-sectional survey using purposive sampling targeted Vietnamese migrant workers in Taiwan’s manufacturing sector working under fixed-term contracts with limited pathways to permanent residence. Data were collected via self-reported questionnaires at brokerage agencies and factory dormitories with assistance from brokers and company representatives. The final valid sample comprised N=636 respondents (64% male, 36% female; 61.5% married; predominantly aged 18–30 or 31–40). Pilot and instrument translation: Original English measures were translated into Vietnamese using forward–back translation, expert review, and cognitive pretesting with a small worker group to ensure clarity. A pilot test (n=64) demonstrated acceptable reliability (Cronbach’s alpha: CQ=0.87, SCA=0.89, JS=0.82, ITS=0.80). Measures: CQ was measured with the 7-item Mini-CQS (Ang et al., 2007) on a 7-point Likert scale (1=strongly disagree to 7=strongly agree). SCA was measured with the 8-item Revised Sociocultural Adaptation Scale (SCAS-R; Wilson et al., 2017) on a 5-point competence scale (1=not at all competent to 5=extremely competent). JS was measured with 5 items adapted from Depré et al. (1995) on a 5-point agreement scale; ITS with 4 items from Kim et al. (1996) on a 5-point agreement scale. Moderators: Chinese proficiency was self-rated on five levels (Not at all to Excellent). LOR was categorized: 1–3, 4–6, 7–9, 10–12, and over 12 years. Reliability and validity: In the main study, Cronbach’s alpha ranged 0.84–0.89 across constructs. CFA indicated acceptable fit (CMIN/df=2.922, CFI=0.937, TLI=0.929, IFI=0.937, RMSEA=0.055). Factor loadings exceeded 0.50; CR>0.70 and AVE>0.50 supported convergent validity. Discriminant validity was supported as inter-construct correlations were below the square roots of AVEs. Common method variance controls included prevalidated scales, reversed items, different response scales, confidentiality assurances, and anonymity; Harman’s single-factor test yielded 29.63% variance (<50%), suggesting CMV was not a major concern. Analysis: Descriptive statistics and VIF checks (all <10) supported assumptions. Hypotheses were tested using PROCESS Macro in SPSS (Model 9) to estimate a moderated mediation model with bootstrapped confidence intervals for indirect effects. Direct paths (CQ→SCA, CQ→JS, SCA→ITS, JS→ITS) and interactions (CQ×Chinese proficiency, CQ×LOR) were estimated, with control variables entered. ANOVA tested group differences in ITS by demographics (gender, education, marital status, age, income).

- Descriptives (N=636): Means (SD): CQ=5.01 (0.66) [range 3–7], SCA=3.18 (0.60) [1–5], JS=3.29 (0.59) [1–5], ITS=3.25 (0.70) [1–5). - Direct associations: Initial regression showed CQ positively associated with ITS (β=0.22, p<0.001). In the full PROCESS model, CQ→SCA (β≈0.12, p<0.001), CQ→JS (β=0.18, p<0.001), SCA→ITS (β=0.31, p<0.001), JS→ITS (β=0.57, p<0.001). CQ’s direct effect on ITS became nonsignificant (β=0.01, p>0.05), indicating full mediation. - Mediation (H1, H2 supported): Indirect effect via SCA: β=0.04, 95% CI [LLCI=0.0154, ULCI=0.648]; via JS: β=0.10, 95% CI [LLCI=0.0653, ULCI=0.1471]. - Moderation on SCA (H3, H4 supported): CQ×Chinese proficiency → SCA (β=0.11, SE=0.05, t=2.30, p<0.05, 95% CI [0.0165, 0.2099]); CQ×LOR → SCA (β=0.17, SE=0.05, t=3.07, p<0.01, 95% CI [0.0603, 0.2738]). - Moderation on JS (H5, H6 rejected): CQ×Chinese proficiency → JS (β=0.06, SE=0.05, t=1.14, p>0.05, 95% CI [−0.0407, 0.1538]); CQ×LOR → JS (β=−0.02, SE=0.05, t=−0.35, p>0.05, 95% CI [−0.1268, 0.0881]). - Moderated mediation: Indirect CQ→ITS via SCA was conditionally moderated by Chinese proficiency (index 95% CI [0.0051, 0.0764]) and by LOR (index 95% CI [0.0084, 0.0993]). - Group differences in ITS: No significant differences by gender (F(1,634)=0.76, p>0.05) or education (F(5,630)=0.59, p>0.05). Significant differences by marital status (F(1,634)=13.64, p<0.001; married M=3.33 vs single M=3.12), age (F(2,633)=6.38, p<0.01; 31–40 higher than 18–30), and income (F(3,632)=6.33, p<0.001; higher income groups had higher ITS). Overall, SCA and JS fully mediated CQ’s effect on ITS; language proficiency and LOR strengthened CQ’s positive link to SCA but not to JS.

The findings address the research questions by demonstrating that cultural competence influences intention to stay through two key pathways: sociocultural adaptation and job satisfaction. CQ enhances behavioral and motivational capacities that facilitate everyday functioning and social interactions in the host culture (SCA) and improve affective evaluations of work (JS). Both SCA and JS, in turn, substantially increase intention to remain employed, rendering the direct CQ→ITS path nonsignificant in the full model. The moderation results clarify boundary conditions: Chinese language proficiency and longer residence amplify how CQ translates into better sociocultural adaptation, reinforcing the acculturation process, yet these moderators do not alter the CQ→JS link, suggesting that job affect may be shaped more by workplace conditions and exchanges than by residence length or language alone. These results contribute to psychological and behavioral management literature by integrating cultural competence, adaptation, and work attitudes in a single moderated mediation framework focused on low-skilled migrant workers. They highlight the importance of developing CQ and fostering acculturation supports to improve retention outcomes in migrant labor contexts.

This study shows that among Vietnamese manufacturing workers in Taiwan, cultural competence increases intention to stay indirectly through sociocultural adaptation and job satisfaction. Local language proficiency and length of residence strengthen CQ’s impact on adaptation but not on job satisfaction, yielding a nuanced understanding of how cultural factors shape retention. The work advances theory by evidencing full mediation via SCA and JS in the CQ–ITS link and by identifying when and how contextual moderators matter. Practically, the results support interventions that build CQ, provide language training, and create supportive environments that facilitate adaptation and enhance job satisfaction to reduce turnover intentions. Future research should generalize across nationalities and sectors, apply longitudinal designs, incorporate multi-source data to mitigate common method bias, and examine undocumented workers to contrast retention mechanisms.

- Non-probability (convenience) sampling and reliance on self-report measures may introduce social desirability and common method bias, despite procedural remedies and a nonsalient Harman’s single-factor result. - Cross-sectional design limits causal inference; longitudinal studies are needed to capture temporal dynamics of CQ, SCA, JS, and ITS. - Sampling focused on Vietnamese workers in the manufacturing sector under the Vietnam–Taiwan labor export program; findings may not generalize to other nationalities or sectors. - Other industries (construction, domestic work, agriculture, forestry, fishery) and undocumented workers were not included; comparative studies across diverse professional, cultural, and language contexts are recommended.