Abstract
Purpose
This study emphasises the importance of both literal and figurative interpretations in explaining training and adjustment (cross-cultural) diversity in a multinational enterprise (MNE) context.
Design/methodology/approach
The study employs structural equation models applied to employees relocated to a Malaysian subsidiary. It emphasises the importance of distinguishing between measurement, analysis and interpretation. This distinction highlights the need to not only directly quantify variables but also indirectly interpret their relationships to address diversity.
Findings
The findings underscore the significance of figurative interpretation in understanding diversity, particularly at the micro level. Literal interpretation focuses on direct analysis, while figurative interpretation involves an indirect analysis where diversity may not be immediately evident. The study suggests that considering both macro and micro levels is essential in understanding training and adjustment diversity.
Originality/value
This study offers a novel perspective by incorporating both literal and figurative interpretations to deepen the understanding of diversity in international business contexts. It uniquely positions itself by distinguishing between the direct quantification of variables and the indirect interpretation of their relationships. This dual approach enables a more comprehensive understanding of how training, adjustment, compliance with standards and commitment to expectations interact within MNEs. By advocating for figurative interpretation, the study encourages a more holistic analysis that extends beyond surface-level metrics, providing valuable insights for future research. It also paves the way for practitioners to develop strategies that not only address diversity at both macro and micro levels but also foster adaptability within culturally diverse work environments.
Keywords
Citation
Mandrinos, S., Liew Sat Lin, C. and Ilma binti Salleh, N. (2024), "Explaining training and adjustment diversity in multinational enterprises: literal and figurative interpretations", Journal of Trade Science, Vol. 12 No. 4, pp. 284-303. https://doi.org/10.1108/JTS-05-2024-0032
Publisher
:Emerald Publishing Limited
Copyright © 2024, Symeon Mandrinos, Constance Liew Sat Lin and Nurul Ilma binti Salleh
License
Published in Journal of Trade Science. Published by Emerald Publishing Limited. This article is published under the Creative Commons Attribution (CC BY 4.0) licence. Anyone may reproduce, distribute, translate and create derivative works of this article (for both commercial and non-commercial purposes), subject to full attribution to the original publication and authors. The full terms of this licence may be seen at http://creativecommons.org/licences/by/4.0/legalcode
1. Introduction
Training and adjustment are particularly important aspects of human resource management (HRM), especially in the context of expatriate assignments (cross-cultural) for multinational enterprises (MNEs), where employees are required to relocate to foreign subsidiaries to fulfil specific roles (Meyer et al., 2020). Successful adjustment through HRM involves efforts to uncover the deeper, more complex mechanisms that influence training. The challenge lies in integrating diverse forms of interpretation to develop robust strategies for managing cross-cultural adjustment.
Training and adjustment in the ever-evolving landscape of international business play pivotal roles in the success and sustainable development of MNEs (Littrell et al., 2006; Meyer et al., 2020). In particular, training and adjustment are critical not only for immediate operational success but also for fostering long-term sustainability within MNEs by enhancing their ability to integrate and thrive in diverse cultural environments (Kedia and Mukherji, 1999). These efforts prepare employees to operate effectively in foreign environments by enhancing their cultural awareness and ability to adjust, which helps avoid misunderstandings and fosters smoother operations and stronger relationships with local partners (Elg and Ghauri, 2023). Similarly, they are essential for the sustainable integration of MNEs into local contexts, reducing uncertainty that could hinder business operations (Liou and Rao-Nicholson, 2021). Investing in training demonstrates an MNE’s commitment to support sustainability and may open up new business opportunities through cross-cultural adjustment.
Training can be separated into pre-to-departure training conducted before the departure and post-to-departure training conducted after the arrival (Littrell et al., 2006; Selmer, 2001). This type of training prepares employees to adjust in diverse locations and manage host subsidiaries (Littrell et al., 2006; Okpara et al., 2021). In this study, we use literal and figurative interpretations to explain training and adjustment in macro-level and micro-level diversities (see Figure 1). Interpretation lies in the depth of analysis and the extent to which diversity is considered beyond the surface level. Literal interpretation focuses on the direct without extending interpretations further, while figurative interpretation follows an indirect analysis where diversity may not be immediately apparent (Keysar, 1989; Giora, 2002).
Despite existing HRM literature on training and adjustment, there is limited research on how figurative interpretation can complement literal interpretation to provide a holistic understanding in a cross-cultural context. Traditional analyses rely on direct measurements without delving into the deeper (indirect), latent constructs that influence diversity. Thus, a significant gap exists in understanding how both literal and figurative interpretations can inform the complexities of training and adjustment in MNEs. This gap highlights the need for further investigation and provides a rationale for this study to explore the importance of incorporating both literal and figurative interpretations in understanding training and adjustment diversity.
In particular, for both MNEs and employees, diversity is of central importance at both macro and the micro levels (Mendenhall and Stahl, 2000). It encompasses a wide range of variations, including but not limited to markets, culture, standards and expectations (Rugman and Verbeke, 2000; Shenkar et al., 2008; Tung and Verbeke, 2010). On the macro level, firms interpret many distant routes because of diversity in countries, entry modes, markets and relationships within and beyond local undertakings (Johanson and Vahlne, 2009; Forgren et al., 2015), while at the micro level, employees interpret diversity in both local endeavours and ventures extending beyond regional borders (Williams and Gregoire, 2015; Maitland and Sammantino, 2015; Baack et al., 2015). With an emphasis at the micro level, employees themselves, rather than the MNEs they work for, play a significant role in interpreting diversity.
Diversity is associated with training and adjustment, which can be directly measured through a first-order measurement. Unlike a direct measurement, higher-order measurement is indirectly assessed through the first-order measurement and serves as a proxy to explain the latent construct. Thus, higher-order measurement coexists alongside first-order measurement within a research analysis (Edwards and Bagozzi, 2000; Jarvis et al., 2003). Commonly employed to interpret first-order measurement is the reflective approach (Hair et al., 2006). With a reflective approach, any changes in one variable (i.e. pre-to-departure training) are anticipated to trigger corresponding changes in all measured variables (i.e. post-to-departure training, accuracy) when they are related to different facets of the overarching variable (i.e. training). This enables the completion of the measurement part of the analysis (Bollen, 1989, 2002).
The reflective approach relates to literal and figurative interpretations in the context of first-order and higher-order measurements as a two-way process. Literal interpretation is a type of basic interpretation that emphasises direct without extrapolating indirect interpretations. On the other hand, figurative interpretation involves indirectly analysing a variable (i.e. training) by considering deeper relations beyond surface appearance (i.e. adjustment). This standpoint involves grappling with the inherent differences in diversity, suggesting that adjustment (cross-cultural) requires more than just a direct measurement and interpretation of training (cross-cultural).
Based on measurement, direct interpretation explains diversity when employees undergo micro-level commitment due to macro-level compliance. By acknowledging the impact on commitment (micro level), figurative (indirect) interpretation underscores the importance of looking beyond the literal (direct) interpretation of compliance (macro level). An indirect figurative interpretation goes beyond the direct aspects of training and adjustment when the higher-order measurement is based on macro-level standards and expectations. In essence, the progression from literal to figurative interpretation explains the multifaceted nature of commitment and underscores the imperative of considering both macro and micro perspectives in understanding training and adjustment.
First-order and higher-order pertain to the level of measurability, while literal and figurative interpretations pertain to the depth of analysis. Measurement involves the process of quantifying actual variables, while interpretation involves examining the meaning of the data. While measurement focuses on obtaining reliability and validity, interpretation focuses on uncovering relationships to address a hypothesis. Therefore, order is typically addressed first in the analysis of the study, followed by a consideration of literal and figurative interpretations to examine the data.
Following the above rationale, research to investigate the relationships between training, adjustment, compliance (standards) and commitment (expectations) at both macro and micro levels should highlight the complexities inherent in interpreting compliance and commitment beyond direct correlations. This approach provides insights that can inform MNEs’ strategies for fostering a supportive and inclusive work environment in diverse cultural settings. By considering indirect figurative interpretations, we support three hypotheses through the exploration of both macro-level standards and micro-level expectations. Thus, this study expands our understanding in multinational contexts regarding compliance and commitment. Furthermore, by acknowledging the complexities of literal and figurative analyses, we invite further interpretations that can inform MNEs’ strategies in global workplaces.
The rest of the paper is organised as follows: We begin with the literature review, followed by methodological evidence and the analysis of the study. We conclude with a summary of the key findings and implications in addition to the directions for future research. Finally, the limitations section acknowledges potential shortcomings of the study.
2. Literature review
2.1 Training and adjustment
This research explores the exchange between training and adjustment of expatriate assignments (Black and Porter, 1991) while highlighting the importance of literal and figurative interpretations. Expatriate assignments represent a critical aspect of MNEs, facilitating global mobility and the expansion of organisational reach. Within this context, the exchange of training and adjustment plays a pivotal role in ensuring the success of HRM endeavours. Assignments that prepare employees to adjust in a host country (Meyer et al., 2020) offer pre-departure (Littrell et al., 2006) and post-departure training (Selmer, 2001). Training is designed for employees embarking on assignments and serves as a preparatory measure to facilitate successful adjustment to new environments. Meyer et al. (2020) underscore the significance of comprehensive training, which encompasses both pre-departure and post-departure components. Pre-departure training, as emphasised by Littrell et al. (2006), aims to equip employees with the necessary knowledge and accuracy to explain potential changes associated with international pressures and firms’ expectations. Post-departure training, as highlighted by Selmer (2001), provides ongoing support and guidance to facilitate continued adjustment to international relocations.
Central to the effectiveness of expatriate training and adjustment is the interaction between compliance-enforcement and commitment-inducement mechanisms. Training and adjustment depend on the interaction between compliance-enforcement and commitment-inducement mechanisms (Su et al., 2018; Miao et al., 2021). These studies explore the interplay between pressure mechanisms, emphasising their role in shaping coherence within MNEs established standards to ensure adherence to their expectations. In particular, compliance-enforcement mechanism involves an exchange between macro and micro levels to ensure MNEs and employees’ adjustments comply with established standards (Kopowski et al., 2021). Commitment-inducement mechanism focuses on employee commitment to facilitate micro-level adjustments that align with MNEs expectations (Su et al., 2018). Compliance plays a crucial role in upholding macro-level standards (Su et al., 2018) and nurturing commitment expectations (Miao et al., 2021). Compliance refers to meeting expectations, which may differ from obligations that have not yet been fully codified in standards and formal responsibilities (Durand et al., 2019). Compliance sets the minimum standards to be considered acceptable, and enforcement ensures that employees’ expectations meet these basic standards.
2.2 Interpretation: literal vs. figurative
The interpretation of training and adjustment within the expatriate context involves both literal and figurative analyses. While literal interpretation focuses on surface-level observations and adherence to established norms, figurative interpretation facilitates a more in-depth understanding of expatriate adjustment. Thus, the interpretation of micro-level adjustment may vary depending on macro-level standards and expectations. Variation in interpretation occurs due to both literal and figurative analyses of training, as employees may interpret adjustment differently. Literal interpretation is anticipated to demonstrate accuracy to MNEs' expectations, while figurative interpretation provides an analysis of measurements extending beyond the direct view of macro-level standards and expectations (Gibbs, 2007; Gibbs and Colston, 2012). Empirical evidence supporting this claim may be limited, but it underscores the complexity of adjustment (Black and Mendenhall, 1990; Tung, 2008). The macro- and micro-level diversities are central to considering literal and figurative interpretations.
Literal interpretation refers to a straightforward analysis, while figurative interpretation considers analysis beyond basic expectations. Macro-level standards encompass overarching expectations, and compliance with these standards ensures that analyses give an in-depth explanation for what is initially apparent. This view suggests that considering in-depth relations beyond the most obvious expectations may contribute to a more accurate adjustment. However, adhering to macro-level standards through literal interpretation focuses on direct “meeting” expectations, whereas a figurative interpretation may consider indirect relations, leading to a complex and multifaceted adjustment. Literal interpretation, while valuable for ensuring alignment with macro-level standards, may overlook the intricacies of micro-level adjustment changes. This is particularly relevant in HRM, where contexts characterised by cultural diversity and contextual complexity and a figurative interpretation will allow an in-depth analysis of adjustment accuracy, taking into account individual differences and contextual pressures.
2.3 Human resource management
Requiring a careful balance between the theoretical frameworks of HRM and the practices of training and adjustment in cross-cultural diversity (Ran and Huang, 2019), expatriate assignments involve implementing strategies to optimise adjustment through tailored training for diverse cultural environments (Heidemann and Søndergaard, 2023). In this context, HRM incorporates pre-departure and post-departure training to prepare expatriates for the potential pressures they may face (Hechanova et al., 2003). Pre-departure training provides expatriates with the necessary knowledge to anticipate and manage pressures and differences, enabling them to work effectively in the host country. This includes macro-level standards and expectations as well as developing micro-level cultural awareness for living and working in the new host location (Sheldon and Elliot, 1999).
2.4 Hypotheses
Standards and expectations, as elucidated by Sheldon and Elliot (1999), converge in both synergistic and differing ways. First, when employees commit themselves to directly interpreting how to pursue expectations as per the condition of compliance, they are also committed based on shared expectations after adopting the highly respected and protected standards (Sheldon and Elliot, 1999). Second, employees are not always directly interpreting adjustment; instead, they may consider a figurative approach that allows them to reconcile macro-level standards with micro-level expectations. In other words, employees may find a balance between macro and micro levels that provides a standard for adjustment capable of maintaining the relationship between training expectations and diversity.
This assertion is grounded in the fact that micro-level variation establishes an interdependence with commitment and compliance while seeking independent adjustment. For instance, diversity prompts macro-level MNEs (Bingham and Davis, 2012; Bingham and Eisenhardt, 2011) and micro-level employees (Drummond, 1995; Staw, 1976) to seek a direct interpretation that aligns with diversity. However, an indirect interpretation may lead to diversity of expectations in the foreign country, deepening the micro-level variation (Baack et al., 2015; O'Grady and Lane, 1996).
Furthermore, training that incorporates adjustment is a dual-phase approach that aims to ensure expatriates are equipped with the necessary knowledge to meet both macro-level standards and micro-level expectations. From this lens, awareness and perceived pressure play pivotal roles in predicting micro-level adjustment, highlighting the complexity–adjustment relationship beyond direct measurement. These factors influence adjustment, suggesting that a comprehensive understanding of awareness and pressure can shape the adaptation outcomes.
In light of the diverse relationship between standards, expectations and interpretation, we hypothesise that:
An indirect (figurative) interpretation extends beyond the direct (literal) interpretation of macro-level standards and micro-level expectations.
Training incorporating pre-departure and post-departure components improves expatriate adjustment in international assignments.
Awareness and pressure predict micro-level adjustment, highlighting the complexity–adjustment relationship beyond direct (literal) measurement.
The above hypotheses underscore the complexity of expatriate adjustment processes and the need for training and interpretation within MNEs. Hence, the above content reflects the diversity of the relationship between standards (macro level), expectations (micro level) and interpretation (literal figurative). By expanding on these insights, we aim to provide a comprehensive understanding of the intricate orientation shaping expatriate assignments, and through empirical research, we seek to contribute to the existing literature by offering practical insights for MNEs on the complexities of global talent mobility.
3. Method
3.1 Research approach
This research is based on the case study approach (Yin, 2018). According to academic research, a case study is an appropriate approach for exploring small sample data (Rashid et al., 2019), with a case and common form of research that usually examines a single country over a period of time, often for the purpose of explaining a final outcome or showing the effects of changes that occurred during the time examined (Geddes, 2003: p. 117). Its representativeness can be found within the collection of selected elements (Tsai, 2007).
3.2 Target population
In this study, the target population comprises employees who have been temporarily relocated to work in the Malaysian subsidiary of their MNEs. These expatriates are integral for understanding the complexities of cross-cultural training and adjustment, as they experience firsthand in an international context. The focus on expatriate employees provides direct insights into the actual experiences and challenges faced during international assignments. These individuals are representative of the broader phenomenon of global mobility within MNEs, making them ideal subjects to study training and adjustment.
3.3 Sample selection
A total of 127 respondents were selected using convenience and snowball sampling methods. These techniques were deemed appropriate due to the potential difficulty in accessing a dispersed population of expatriates. Convenience sampling allowed for easier and quicker collection of data from available respondents who fit the study’s criteria (Newton, 2024). Snowball sampling leveraged the networks of initial respondents to reach additional expatriates, thus broadening the sample pool (Newton, 2024).
3.4 Minimum conditions
Participants must be employees who are currently on an expatriate assignment in the Malaysian subsidiary of their MNE. They should have a minimum of one year of international assignment experience to ensure they have had adequate exposure to cross-cultural training and adjustment, and they should be proficient in English to comprehensively understand and respond to the questionnaire.
3.5 Questionnaire structure
Data from 127 respondents are collected via a questionnaire consisting of four sections. The first section comprises demographic data to gather information on age, gender, nationality, industry, job roles and lengths of employment and international assignments, while the remaining sections measure various aspects of the target population relevant to training, adjustment, compliance and commitment.
3.6 Demographic data
The demographic data provide insights into the respondents, revealing patterns in age distribution, gender representation, nationality, industry, job roles and lengths of employment and international assignments. In particular, the variable of age shows the majority of respondents fall into the age groups of 31–40 years old (37.8%) and 41–50 years old (38.6%). The age group of 21–30 years old represents 9.4% of the respondents, while the age group of 51–60 years old represents 14.2%. The data are skewed towards males, comprising 73.2% of the respondents, with females representing 26.8%. The most represented nationalities are Japanese (9.4%) followed by Indian (7.9%), British (7.9%) and Chinese (7.1%), whereas the rest have lower representation, ranging from 0.8% to 5.5%. The largest proportion of respondents work in the oil and gas industry (37.8%), followed by manufacturing (7.1%) and technology (7.9%). Other industries have smaller representation, ranging from 0.8% to 6.3%. Most respondents hold either senior roles (50.4%) or managerial roles (28.3%) in their current firms. The majority of respondents have been with their current firm for 6–15 years, with 6–10 years being the highest (36.2%). The most common length of international assignment is 3–4 years (44.1%), followed by 1–2 years (26.8%).
4. Analysis and findings
4.1 Reliability and adequacy
In this research, we have integrated the findings directly within the analysis section. The role of the findings section is to present the data systematically and logically, enabling readers to understand the foundation upon which the analysis is built. Following this approach, we connect data directly to the analysis, offering a structured presentation while minimising content repetition. In the study’s analysis, the rationale for the choice of tools depends on online distribution via Microsoft Forms, which was chosen due to its ability to reach a geographically dispersed participant pool efficiently. It ensures easy access for respondents, facilitating higher response rates. The platform allows for user-friendly design and anonymity, which can enhance the reliability of responses. In particular, selected by convenience and snowball sampling methods (Newton, 2024), numerical values are assigned in ordinal and categorical scales based on the strongly disagree to strongly agree scale (Liu and Keusch, 2017), and the questionnaire was made accessible online on Microsoft Forms for respondents to participate.
In detail, participants were invited to complete the questionnaire online via Microsoft Forms. The online format facilitated participation at the respondents’ convenience, which is particularly beneficial for a busy expatriate population. Completed questionnaires were downloaded into Microsoft Excel to ensure a structured data format. RStudio was employed to handle data cleaning and preparation, ensuring no missing data and that all sections of each questionnaire were complete. Microsoft Excel is important for initial data storage and management due to its wide acceptance, simplicity and compatibility with various data analysis tools, while RStudio is important for its powerful data analysis capabilities, allowing detailed statistical analyses to assess the reliability and internal consistency of the data. By systematically implementing these tools and processes, the study aims to ensure accurate, reliable and comprehensive data collection and analysis, providing robust insights into the training and adjustment experiences of expatriates in MNEs.
The RStudio programming language is used to describe the study’s reliability and internal consistency reliability. The value of Cronbach’s alpha (0.89) represents good internal consistency and reliability. The Guttman’s lambda-6 value (0.95) is the proportion of variance that is accounted for by the underlying construct that demonstrates internal consistency. The average inter-item correlation (0.37) provides an indication of the extent to which variables are related to each other. The signal-to-noise ratio (8.2) denotes the strength of the true score relative to measurement error. The average standardised error of 0.021 represents the average error associated with each variable, revealing reliability. The mean of 5.4 and standard deviation of 0.49 provide information about the distribution of responses across the scale, while the median inter-item correlation of 0.38 provides another measure of the strength of relationships, highlighting that our scale has good reliability and internal consistency.
The Kaiser–Meyer–Olkin (KMO) and the Bartlett’s test of sphericity measurements are used to assess the adequacy of data for factor analysis by assessing the proportion of variance among variables that might be common variance. Variables close to 1 are well suited for factor analysis, and thus, our measurement of 0.81 shows that data are suitable for factor analysis. Additionally, the individual KMO values for each variable are also relatively high, further supporting the adequacy of the data. The Bartlett’s test statistic and the value less than 2.22e−16 represent a significant p-value with suitable variables for factor analysis. Because our study assesses whether the actual variables are intercorrelated, it shows that our data are suitable to identify underlying latent factors.
4.2 Factor analysis
After assessing reliability and adequacy, we measured the underlying factors that directly interpret the patterns of correlations among observed variables. This approach helps identify indirect patterns and relations where diversity may not be immediately apparent, thus facilitating a figurative interpretation. In our analysis, the three-factor model has fewer degrees of freedom for the null model (52) compared to the two-factor model (64). A figurative interpretation signifies that the three-factor model is more parsimonious but can achieve a balance between simplicity and adequate interpretation of the observed data. The objective function refers to the criterion that seeks to optimise during the estimation process. The objective function value is lower for the three-factor model (1.83), suggesting better model fit compared to the two-factor model (3.62). The probability associated with the chi-square test is extremely low for both models, indicating that the observed data significantly deviate from what would be expected under the null hypothesis. However, the p-value is even lower for the 3-factor model, suggesting a better fit. The chi-square statistic is a measure of how well the estimated factor model fits the observed data. The chi-square value is lower for the three-factor model (217.26), implying better model fit compared to the two-factor model (431.32). The root mean square residual represents the average difference between the observed and predicted covariances. In this case, the three-factor model (0.03) indicates better model fit is lower compared to the two-factor model (0.07). The Tucker–Lewis index compares the fit of the estimated factor model to a baseline model, with values close to 1 suggesting good fit, while higher values indicating better fit. The index is higher for the three-factor model (0.808) compared to the two-factor model (0.655), suggesting better fit. This signifies that a higher index better represents complex interrelations among variables, including those that may be indirectly related or not readily observable. The root means square error of approximation (RMSEA) measures of how well the models fit the data, adjusted for the number of parameters in the model with values close to 0.05 or lower show good fit, with smaller values showing better fit. In our case, it is lower for the three-factor model (0.158), representing a better fit compared to the two-factor model (0.212). Finally, the Bayesian information criterion balances model fit and complexity, with lower values signifying better-fitting models. In this case, the value is lower for the three-factor model (−34.63) compared to the two-factor model (121.29).
The adequacy of the factor model helps to measure validity and reliability while providing information about the fit to the actual data, thereby enabling the evaluation of adequacy and comparison of alternative models. Hence, a literal interpretation explains that the three-factor model suggests better reliability and captures important variability that leads to a better representation of the underlying structure (see Figure 2). From this lens, the two models, MR1 and MR2, contain factor loadings of 1, while the MR3 model includes a variable without contribution. A literal interpretation suggests that variables with factor loadings of 1 may have little or no impact on later clustering results, and thus, we should choose to exclude these variables to reduce the dimensionality of our data, as they do not add any additional information to simplify our analysis. However, an indirect interpretation suggests that because these variables are perfectly represented or not represented by the extracted factors, they can still be used as the basis for clustering directly rather than using the factor scores. Factor scores represent a summary of the actual (original) variables based on factor loadings. Yet, by using actual variables, we can retain the information contained in those variables. This is because factor scores condense the measurement into a single score for each observation, potentially losing some information from the original variables in the process of creating these summary scores. The decision to indirectly look at the data and use factor scores and actual variables is to deal with the complexity of training and adjustment diversity.
Figure 2 provides a direct representation of the relationships between the factors extracted from the factor analysis, with the matrix representation showing the correlation coefficient between two factors. In a literal interpretation of the data, we observe that there are three factors that appear to be the preferred choice when observing the factor correlation matrix (see Table 1). In the first correlation matrix, the correlation coefficients of MR2 and MR3 with MR1 specify a moderate-to-strong positive correlation. In the second correlation matrix, the correlation coefficient of MR2 with MR1 directs a moderate positive correlation. Comparing the two matrices, both show a positive correlation between MR1 and MR2, suggesting that changes in one factor are associated with changes in the other factor, although the correlation is slightly higher in the second matrix (0.27 vs. 0.34). A direct interpretation of a third factor (MR3) in the first matrix introduces additional information about its relationships with MR1 and MR2. The correlations involving MR3 suggest a strong relationship between MR3 and the other factors compared to the relationship between MR1 and MR2. From a figurative interpretation, however, the presence of a third factor in the first matrix adds complexity to the relationships observed. This implies that, despite changes based solely on the factor relation offering a straightforward positive correlation, yet one factor may influence or be influenced by changes that indirectly may affect another factor through underlying variables that are not immediately apparent.
A comparison of factor analysis with the parallel factor analysis highlights the importance of considering diversity in forming a figurative interpretation (McDonald, 1985). Parallel factor analysis compares the eigenvalues obtained from actual data with those obtained from simulated (generated) data (Lim and Jahng, 2019). This comparison helps identify the presence of underlying factors in the data that might not be obvious through direct observation (Lim and Jahng, 2019; McDonald, 1985). A literal interpretation that excludes variables with factor loadings of 1 may simplify the analysis, yet it may discard figurative information embedded in those variables and thus risk overlooking valuable figurative information inherent in these variables. In this case, the figurative interpretation suggests the importance of retaining valuable information while measuring indirect relations inherent in the data. With literal interpretation, loadings close to 1 are typically considered highly related. A figurative interpretation, however, considers the diversity across variables. Rather than focusing on loadings close to 1, variables with lower loadings still contribute to the understanding of factor analysis (see Figure 3a). Hence, by incorporating loading diversity, a figurative interpretation helps avoid overlooking the measurement of eigenvalues of principal factors encoded in those variables by the simulated data.
4.3 Principal components
Next, we use principal component analysis as a reduction technique to identify factors that explain the correlations among variables. We calculate the mean correlation (0.41) to assess the eligibility of the variables to continue with the analysis. To decide on a principal component, the correlation needs to be above 0.3 or below −0.3, indicating strong evidence of correlation among variables. Further, we check whether principal components are independent and whether they capture the essence of the variables using the centre output, which represents the mean values used in the analysis (see Table 2). For each variable, the provided value is the mean across all observations. A literal interpretation of the mean values explains the central tendencies, where higher values demonstrate a positive response, especially for variables that likely represent other aspects (i.e. awareness) measured in the study, which may include other constructs of interest (i.e. perception and behaviour). From an indirect approach, however, a figurative interpretation can capture diversity beyond the identification of underlying factors. For instance, despite the variable “change” showing a lower mean value, it may still hold considerable importance in understanding diversity as it represents an aspect of complexity faced by expatriates.
In addition to the literal interpretation of parallel analysis and component centres, we create a scree plot to further visualise the eigenvalues, aiding in determining the number of components to retain when they begin to level off (6.42–2.53–1.16–0.91 …). Retaining components up to this point ensures that a significant amount of variance in the data is captured while avoiding generalisation. A literal interpretation examines the loading plot to understand the actual data and identify the variables most strongly related to each cluster. This direct view illustrates three eigenvalue relationships and provides a graphical representation of the three factor loadings, elucidating the patterns of correlations among the actual variables (see Figure 3b). However, a figurative interpretation will consider any broader implications of the three identified components and factors.
4.4 Structural equation model(s)
Incorporating the above, several key insights emerge that inform the decision to continue measuring our data with structural equation modelling (SEM). The factor analysis reveals correlations among the actual variables, suggesting specific factors that explain correlation, while the parallel analysis enhances understanding by comparing eigenvalues from actual and simulated data, shedding light on potentially hidden details. Similarly, the figurative interpretation of factor analysis emphasises retaining valuable information and explores indirect relations in the data, highlighting complexity beyond direct interpretations, while the principal component analysis further validates the correlations among variables and the presence of underlying components. In the same vein, centring the variables simplifies interpretation, while the scree plot aids in determining the number of components to retain, pointing towards a complex structure within the data. In this case, both literal and figurative interpretations coexist to contribute to understanding our data.
While literal interpretation focuses on direct relations, figurative interpretation considers broader implications and underlying patterns, surpassing simple correlations. Hence, considering these relationships, the decision to continue with SEM analysis allows for the examination of complex relationships among variables, accommodating both direct and indirect interpretations. The correlations and underlying patterns identified through factor and principal component analysis suggest that our data can be further explored using SEM, which facilitates hypothesis testing by examining the relationships between latent constructs and actual variables.
We measured a number of models, yet we interpret only three, with the first explaining the macro level (standards – expectations) regressed on the latent variables of accuracy, knowledge, training and pre-adjustment, whereas the second and third explaining the micro level (expectations) regressed on the latent variables of awareness, training, change, post-adjustment, pressure and knowledge. Table 3 suggests that model_a fits the data well, as shown by the high p-value, reasonably good comparative fit index (CFI) and RMSEA values and low standardised root mean square residual (SRMR) value. The “CFI” value ranges from 0 to 1, where values closer to 1 signify a better fit. A “CFI” of 0.95 or higher is generally considered a good fit, suggesting that the model_a explains the data well. The “RMSEA” measures the discrepancy between the observed covariance matrix and the model-implied covariance matrix range from 0 to infinity, with lower values indicating better fit. In our data, a value below 0.05 suggests a close fit. The “SRMR” value provides a measure of the average absolute difference between the observed and predicted correlations. A range from 0 to 1 with lower values specifies better fit. Because our value is below 0.08 is considered acceptable, showing that the model_a effectively reproduces the observed correlations. The parameter estimates of the relationships and variances highlight how the latent variables contribute to explaining the macro level. For accuracy, the estimated value is 1.00 with no standard error, suggesting perfect measurement of this latent variable. For knowledge, the estimated value is 0.450 with a standard error of 0.117; for training, the estimated value is 1.039 with a standard error of 0.450, while for pre-adjustment, the estimated value is 0.988 with a standard error of 0.128, specifying the strength of their relationships. Regressed on the above latent variables, the macro level involves compliance, thus meeting standards directly and ensuring that the actual data reflect established expectations. This literal interpretation, however, differs from a figurative interpretation that goes beyond the direct standards and considers diversity of accuracy, knowledge, training and pre-adjustment as variables of readiness to meet compliance expectations.
In a similar manner, model_b fits the data well, because of the non-significant chi-square test, high “CFI” value, low “RMSEA” value and low “SRMR” value (see Table 4). The parameter estimates of the relationships and variances highlight how the latent variables contribute to explaining the micro-level. The “CFI” value of 0.997 shows a very good fit of the model. The “RMSEA” value of 0.019 below the typical threshold of 0.05 suggests a close fit of the model. The “SRMR” value of 0.043 is below the commonly accepted threshold of 0.08 and reveals a good fit and suggests that model_b reproduces the observed correlations among the latent variable of awareness with an estimated value of 1.00. For training, the estimated value is 0.643 with a standard error of 0.167, suggesting the strength of the relationship between the latent variable training and its indicators. Next, the negative value of change (−0.308) with a standard error of 0.273 suggests an inverse relationship with the latent variable. For post-adjustment, the estimated value is 1.083 with a standard error of 0.185, representing the strength of the relationship with the latent variable, in addition to the variable pressure. The value is 1.214 with a standard error of 0.192, showing a strong positive relationship with the latent variable. For knowledge, the value is 1.450 with a standard error of 0.210, representing a strong positive relationship with the latent variable to explain the micro level.
From a literal interpretation, the outcome affirms the reliability of the model and its ability to capture the relationships among the micro-level variables accurately, and the parameter estimates provide further information on how the latent variables contribute to explaining the micro-level expectations. However, the figurative interpretation underscores diversity in relation to the negative value of the “change” variable, suggesting that, as the level of change decreases, there is a corresponding increase related to micro-level expectations. This inverse relationship highlights the need to consider an indirect interpretation while addressing change.
For model_c, we exclude knowledge from its variable list, and the p-value associated with the chi-square test is 0.284, demonstrating that the value is not statistically significant at the level of 0.05 (see Table 5). This suggests that the model fits the data well. The “CFI” value is 0.970, which suggests a good fit in addition to a reasonable fit of the model to the data because “RMSEA” value is 0.060, which is slightly higher than the typical threshold of 0.05. This aligns with the “SRMR” value of 0.047 that is below the commonly accepted threshold of 0.08, showing a good fit of the model. The parameter estimates highlight the latent variables of awareness and training, contributing to explaining the micro level. Yet, the negative estimate of variable change is −0.269 with a standard error of 0.301, suggesting a negative relationship with the latent variable, though it is not statistically significant, while the post-adjustment value is 1.233 with a standard error of 0.254, demonstrating a strong positive relationship with the latent variable. Finally, the value of pressure is 1.090 with a standard error of 0.231, signifying a strong positive relationship with the latent variable.
In the case of model_b and model_c, the inclusion and exclusion of the “knowledge” variable have implications for understanding the relationship between standards and expectations. The statistical indicators, including the chi-square test, “CFI”, “RMSEA” and “SRMR”, affirm the reliability of both model_b and model_c. They suggest that these models accurately capture the micro-level expectations. The parameter estimates in both models directly interpret how latent variables, such as awareness, training, post-adjustment and pressure, contribute to explaining micro-level expectations. In model_b, the inclusion of the “knowledge” variable also highlights its significant role in influencing micro-level expectations. The negative value of the “change” variable underscores the complexity of managing training and adjustment processes while addressing change. This highlights the need to consider an indirect interpretation based on potential obstacles that may arise during training and adjustment expectations. The inclusion or exclusion of “knowledge” shapes the interpretation of the analysis. In model_b, where “knowledge” is measured, it contributes to fostering commitment among employees. On the other hand, in model_c, it suggests that awareness, training, post-adjustment and pressure play more prominent roles in influencing micro-level expectations without the confounding effects or overlapping variance that “knowledge” might introduce. Thus, a literal interpretation may not contribute to how variables interact, and therefore, a figurative interpretation allows consideration of the differences in how micro-level expectations are influenced by the inclusion of “knowledge”, which may introduce confounding variance that dilutes the relation of other factors.
Following the above findings, this study demonstrates that the results are meticulously crafted, showcasing a comprehensive analysis of the collected data through rigorous statistical techniques, including factor analysis and SEM. The presentation of findings effectively elucidates the correlations among variables and the underlying factors that shape expatriate training and adjustment interactions. Similarly, the integration of both literal and figurative interpretations adds depth to the analysis, allowing for an understanding of compliance and commitment across macro- and micro-level expectations. This dual approach not only strengthens the robustness of the results but also highlights the complexity and multifaceted nature of training and adjustment.
5. Discussion and implications
This study discusses both literal and figurative interpretations, highlighting their distinct contributions to understanding and analysing data. Incorporating both concepts allows for a better understanding of the multi-dimensional aspects of training and adjustment diversity. The factor analysis and SEM both illustrate how literal interpretations focus on direct relationships, such as standards and expectations, while figurative interpretations delve into deeper, less obvious connections, like those related to diversity and indirect relations. Thus, the study’s analysis supports H1 by emphasising that indirect interpretation captures additional complexities beyond direct correlational measures. Similarly, H2 and H3 can be confirmed based on the study’s analysis complexities in the existing discussion of training and the role of awareness and pressure in predicting adjustment diversity. The statistical analyses shed light on these aspects, making them relevant extensions of our findings. In a nutshell, our study underscores the importance of incorporating both literal and figurative interpretations to fully understand expatriate training and adjustment diversity. This dual approach supports the stated hypotheses and advances our understanding of how to optimise HRM practices in multinational contexts, thereby fostering sustainable development in MNEs through better-equipped and culturally aware employees.
Our study explores relevant complexities of cross-cultural training and adjustment diversity by examining expatriates working in the Malaysian subsidiaries of MNEs. Previous studies have focused on how well expatriates align with their work environments, highlighting the importance of training and adjustment in ensuring a good fit between individuals and their work environment (Nolan and Morley, 2013). Training guarantees that expatriates possess the necessary knowledge to align with standards and adhere to expectations, such as job demands and environment and fit in new cultural settings, thereby enhancing person-job fit under pressure (Nolan and Morley, 2013). Thus, explaining the relationship between an expatriate and the international environment involves aligning training and adjustment to the various dimensions of a cross-cultural environment. Similarly, the drive to adjust to new cultural contexts requires training to equip expatriates with the necessary knowledge to interact effectively across cultures (Ramalu et al., 2010). This compatibility, which includes cultural awareness and exposure to different environmental pressures, refers to how expatriates acclimate to new cultural environments. Involving both inner and outer adjustment, this process in diverse settings can enhance adaptability to foreign subsidiaries and improve expatriate efforts in facilitating continued adjustment to international relocations (Ramalu et al., 2010).
The findings of this research also align with and expand upon the previous studies, emphasising the significance of pre-departure and post-departure training in facilitating expatriate adjustment. Our findings corroborate this by demonstrating that comprehensive training, encompassing both pre-departure and post-departure components, is crucial for successful expatriate adjustment. Additionally, the interaction between compliance-enforcement and commitment-inducement mechanisms, highlighted by Su et al. (2018) and Miao et al. (2021), plays a critical role in shaping coherence within MNEs’ established standards to ensure adherence to their expectations. This study reaffirms their findings by showcasing how these mechanisms facilitate training and adjustment by balancing macro-level standards and micro-level expectations. Moreover, our approach of integrating literal and figurative interpretations adds depth to the analysis, as supported by Gibbs (2007) and Gibbs and Colston (2012). The ability of figurative interpretation to uncover deeper constructs beyond literal definitions is critical, especially when considering the complexity of expatriate training and adjustment.
This study demonstrates the importance of incorporating figurative interpretation alongside literal interpretation in understanding the complex relation between training and adjustment (cross-cultural). By considering an indirect figurative interpretation of the data, we support all hypotheses through the exploration of both macro-level standards and micro-level expectations. By incorporating a figurative interpretation, we highlight the complexity inherent in compliance and commitment, hence going beyond direct correlations. This theoretical perspective broadens the scope of current research by advocating for holistic approaches that integrate multiple layers of interpretation. Similarly, our hypotheses are upheld as our analysis considers an indirect significance, shaping both macro-level standards and micro-level expectations. Moreover, our discussion of first-order and higher-order measurement highlights the importance of distinguishing between levels of measurability and depth of analysis. This distinction underscores the need to not only quantify variables but also examine their meaning and relationships to address research hypotheses effectively. By acknowledging this gap and advocating for a holistic approach that integrates measurement and interpretation, our study lays the groundwork for further investigation into these critical areas. Hence, our study emphasises the importance of considering both literal and figurative interpretations when analysing complex phenomena. By embracing a figurative interpretation, researchers and practitioners can gain a holistic understanding of the underlying mechanisms in cross-cultural training and adjustment while also expanding our understanding in multinational contexts regarding compliance and commitment. Studies exploring these aspects could provide valuable insights for practitioners seeking to foster a supportive and inclusive work environment across diverse cultural settings. Additionally, our study opens avenues for future research. By acknowledging the complexities of literal and figurative analyses, we invite further interpretations that can inform MNEs’ strategies in the global arena.
6. Limitations and future research
While the analysis presented offers valuable insights into the relationships between variables and the levels of compliance and commitment, it is essential to acknowledge specific limitations. First, the study’s reliance on a specific sample of employees temporarily relocated to work in the Malaysian subsidiary of multinational corporations may limit the generalisability of the findings. Therefore, caution should be exercised when applying these findings to other contexts or populations. Second, the size and representativeness of the sample offer another limitation, as a larger and more diverse sample could enhance the generalisability of our findings and mitigate the risk of biases associated with a smaller sample size. Third, the use of ordinal and categorical scales for measuring variables may limit the precision of the data collected. Future research could consider incorporating more objective measures or mixed-method approaches to triangulate findings and enhance the validity of the results. Finally, while SEM can explore complexity, it is important to recognise that this approach has inherent limitations. For example, factor analysis assumes that observed variables are linearly related to latent factors. Failure to meet this assumption – as a direct interpretation suggests – could impact the accuracy and reliability of the SEM estimates. Therefore, future research could consider a figurative analytical approach that is sensitive enough to interpret international business phenomena and their potential contributions.
Figures
Figure 1
Interpretation
Figure 2
Factor analysis diagram
Figure 3
Scree plots
Factor correlation
| Factor correlation | |||||
|---|---|---|---|---|---|
| MR1 | MR2 | MR3 | MR1 | MR2 | |
| MR1 | 1.00 | 0.27 | 0.62 | 1.00 | 0.34 |
| MR2 | 0.27 | 1.00 | 0.43 | 0.34 | 1.00 |
| MR3 | 0.62 | 0.43 | 1.00 | ||
Source(s): Table by authors
Component centres
| Pre_orient | Post_orient | Pre_adjust | Post_adjust | Change | Accuracy | Enjoyment |
| 5.322835 | 5.19685 | 5.244094 | 5.440945 | 4.614173 | 5.598425 | 5.661417 |
| Uncertainty | Awareness | Social | Training | Knowledge | Pressure | Stand_Expect |
| 5.503937 | 5.653543 | 5.811024 | 5.559055 | 5.645669 | 5.566929 | 5.314961 |
Source(s): Table by authors
Macro level
| model_a | p-value (chi-square) | cfi | rmsea | srmr |
|---|---|---|---|---|
| 0.810 | 1.000 | 0.000 | 0.012 | |
| estimate | std.err | p(>|z|) | ||
| Latent variables | Accuracy | 1.000 | ||
| Knowledge | 0.450 | 0.117 | 0.000 | |
| Training | 1.039 | 0.450 | 0.000 | |
| Pre_adjust | 0.988 | 0.128 | 0.000 | |
| Variances | Accuracy | 0.091 | 0.027 | 0.010 |
| Knowledge | 0.434 | 0.056 | 0.000 | |
| Training | 0.120 | 0.031 | 0.000 | |
| Pre_adjust | 0.389 | 0.055 | 0.000 | |
| macro_level | 0.306 | 0.055 | 0.000 | |
| macro_level = ∼Accuracy + Knowledge + Training + Pre_adjust | ||||
Source(s): Table by authors
Micro level
| model_b | p-value (chi-square) | cfi | rmsea | srmr |
|---|---|---|---|---|
| 0.399 | 0.997 | 0.019 | 0.043 | |
| estimate | std.err | p(>|z|) | ||
| Latent variables | Awareness | 1.000 | ||
| Training | 0.643 | 0.167 | 0.000 | |
| Change | −0.308 | 0.273 | 0.260 | |
| Post_adjust | 1.083 | 0.185 | 0.000 | |
| Pressure | 1.214 | 0.192 | 0.000 | |
| Knowledge | 1.450 | 0.210 | 0.000 | |
| Variances | Awareness | 0.218 | 0.033 | 0.000 |
| Training | 0.383 | 0.050 | 0.000 | |
| Change | 1.276 | 0.161 | 0.000 | |
| Post_adjust | 0.304 | 0.044 | 0.000 | |
| Pressure | 0.269 | 0.043 | 0.000 | |
| Knowledge | 0.148 | 0.038 | 0.000 | |
| micro_level | 0.166 | 0.044 | 0.000 | |
| micro_level = ∼ Awareness + Training + Change + Post_adjust + Pressure + Knowledge | ||||
Source(s): Table by authors
Micro level
| model_c | p-value (chi-square) | cfi | rmsea | srmr |
|---|---|---|---|---|
| 0.284 | 0.970 | 0.060 | 0.047 | |
| estimate | std.err | p(>|z|) | ||
| Latent variables | Awareness | 1.000 | ||
| Training | 0.760 | 0.197 | 0.002 | |
| Change | −0.269 | 0.301 | 0.371 | |
| Post_adjust | 1.233 | 0.254 | 0.000 | |
| Pressure | 1.090 | 0.231 | 0.000 | |
| Variances | Awareness | 0.225 | 0.041 | 0.000 |
| Training | 0.359 | 0.050 | 0.000 | |
| Change | 1.281 | 0.161 | 0.000 | |
| Post_adjust | 0.257 | 0.054 | 0.000 | |
| Pressure | 0.324 | 0.054 | 0.000 | |
| micro_level | 0.159 | 0.049 | 0.001 | |
| micro_level = ∼ Awareness + Training + Change + Post_adjust + Pressure | ||||
Source(s): Table by authors
Statements: All authors declare that they have no conflicts of interest.
References
Baack, W.D., Dow, D., Parente, R. and Bacon, R.D. (2015), “Confirmation bias in individual-level perceptions of psychic distance: an experimental investigation”, Journal of International Business Studies, Vol. 46 No. 8, pp. 938-959, doi: 10.1057/jibs.2015.19.
Bingham, C.B. and Davis, J.P. (2012), “Learning sequences: their existence, effect, and evolution”, Academy of Management Journal, Vol. 55 No. 3, pp. 611-641, doi: 10.5465/amj.2009.0331.
Bingham, C.B. and Eisenhardt, K.M. (2011), “Rational heuristics: the ‘simple rules’ that strategists learn from process experience”, Strategic Management Journal, Vol. 32 No. 13, pp. 1437-1464, doi: 10.1002/smj.965.
Black, J.S. and Mendenhall, M.E. (1990), “Cross-cultural training effectiveness: a review and a theoretical framework for future research”, Academy of Management Review, Vol. 15 No. 1, pp. 113-136, doi: 10.5465/amr.1990.11591834.
Black, J.S. and Porter, L.W. (1991), “Managerial behaviors and job performance: a successful manager in Los Angeles may not succeed in Hong Kong”, Journal of International Business Studies, Vol. 22 No. 1, pp. 99-114, doi: 10.1057/palgrave.jibs.8490294.
Bollen, A.K. (1989), Structural Equations with Latent Variables, Wiley, NY.
Bollen, A.K. (2002), “Latent variables in psychology and the social sciences”, Annual Review of Psychology, Vol. 53 No. 1, pp. 605-634, doi: 10.1146/annurev.psych.53.100901.135239.
Drummond, H. (1995), “De-escalation in decision making: a case of disastrous partnership”, Journal of Management Studies, Vol. 32 No. 3, pp. 265-281, doi: 10.1111/j.1467-6486.1995.tb00776.x.
Durand, R., Hawn, O. and Ioannou, I. (2019), “Willing and able: a general model of oganizational responses to normative pressures”, Academy of Management Review, Vol. 44 No. 2, pp. 299-320, doi: 10.5465/amr.2016.0107.
Edwards, J.R. and Bagozzi, R.P. (2000), “On the nature and direction of relationships between constructs and measures”, Psychological Methods, Vol. 5 No. 2, pp. 155-174, doi: 10.1037/1082-989x.5.2.155.
Elg, U. and Ghauri, P.N. (2023), “Towards a global sustainability approach: challenges and opportunities for multinationals”, in Ghauri, P.N. and Elg, U. (Eds), Creating a Sustainable Competitive Position: Ethical Challenges for International Firms, Emerald Publishing, Leeds.
Forgren, M., Holm, U. and Johanson, J. (2015), “Knowledge, networks and power - the uppsala school of international business”, in Foorsgren, M., Holm, U. and Johanson, J. (Eds), Knowledge, Networks and Power - the Uppsala School of International Business, Palgrave Macmillan, Basingstoke, Hampshire, UK.
Geddes, B. (2003), Paradigms and Sand Castles: Theory Building and Research Design in Comparative Politics, University of Michigan Press.
Gibbs, R.G. (2007), Analysing Qualitative Data, Sage Publications, London.
Gibbs, W.R. and Colston, L.H. (2012), Interpreting Figurative Meaning, Cambridge University Press, NY.
Giora, R. (2002), “Literal vs. figurative language: different or equal?”, Journal of Pragmatics, Vol. 34 No. 4, pp. 487-506, doi: 10.1016/s0378-2166(01)00045-5.
Hair, J.F., Black, B., Babin, B., Anderson, R.E. and Tathan, R.L. (2006), Multivariate Data Analysis, Prentice Hall, Upper Saddle River, NJ.
Hechanova, R., Beehr, T. and Christiansen, N. (2003), “Antecedents and consequences of employees’ adjustment to overseas assignment: a meta-analytic review”, Applied Psychology. An International Review, Vol. 52 No. 2, pp. 213-236, doi: 10.1111/1464-0597.00132.
Heidemann, C. and Søndergard, M. (2023), “Systematic cross-cultural management education: a quasi-experimental analysis of guided experiential learning during intercultural simulations”, Cross Cultural and Strategic Management, Vol. 30 No. 2, pp. 413-439, doi: 10.1108/ccsm-01-2022-0007.
Jarvis, C.B., Mackenzie, S.B. and Podsakoff, P.M. (2003), “A critical review of construct indicators and measurement model misspecification in marketing and consumer research”, Journal of Consumer Research, Vol. 30 No. 2, pp. 199-218, doi: 10.1086/376806.
Johanson, J. and Vahlne, J.E. (2009), “The uppsala internationalization process model revisited: from liability of foreignness to liability of outsidership”, Journal of International Business Studies, Vol. 40 No. 9, pp. 1411-1431, doi: 10.1057/jibs.2009.24.
Kedia, L.B. and Mukherji, A. (1999), “Global managers: developing a mindset for global competitiveness”, Journal of World Business, Vol. 33 No. 3, pp. 230-251, doi: 10.1016/s1090-9516(99)00017-6.
Keysar, B. (1989), “On the functional equivalence of literal and metaphorical interpretations in discourse”, Journal of Memory and Language, Vol. 28 No. 4, pp. 375-385, doi: 10.1016/0749-596x(89)90017-x.
Kopowski, S., Mazzioni, S., Dal Magro, C.B. and Rosa, F.S. (2021), “National culture and corporate social responsibility”, Revista Braileira Delelott Gestao Delelott Negocios, Vol. 23 No. 3, pp. 488-502, doi: 10.7819/rbgn.v.23i3.4115.
Lim, S. and Jahng, S. (2019), “Determining the number of factors using parallel analysis and its recent variants”, Psychological Methods, Vol. 24 No. 4, pp. 452-467, doi: 10.1037/met0000230.
Liou, R.-S. and Rao-Nicholson, R. (2021), “Multinational enterprises and sustainable development goals: a foreign subsidiary perspective on tackling wicked problems”, Journal of International Business Policy, Vol. 4 No. 1, pp. 136-151, doi: 10.1057/s42214-020-00080-8.
Littrell, L.N., Hess, K.P., Riedel, S. and Paley, M. (2006), “Expatriate preparation: a critical analysis of 25 years of cross-cultural training research”, Human Resource Development Review, Vol. 5 No. 3, pp. 355-388, doi: 10.1177/1534484306290106.
Liu, M. and Keusch, F. (2017), “Effects of scale direction on response style of ordinal rating scales”, Journal of Official Statistics, Vol. 33 No. 1, pp. 137-154, doi: 10.1515/jos-2017-0008.
Maitland, E. and Sammantino, A. (2015), “Managerial cognition and internationalization”, Journal of International Business Studies, Vol. 46 No. 7, pp. 733-760, doi: 10.1057/jibs.2015.9.
McDonald, R. (1985), Factor Analysis and Related Methods, Howe, East Sussex, Psychology Press.
Mendenhall, M.E. and Stahl, G.K. (2000), “Expatriate training and development: where do we go from here?”, Human Resource Management, Vol. 39 Nos 2-3, pp. 251-265, doi: 10.1002/1099-050x(200022/23)39:2/3<251::aid-hrm13>3.0.co;2-i.
Meyer, K.E., Li, C. and Schotter, A.P. (2020), “Managing the MNE subsidiary: advancing a multi-level and dynamic research agenda”, Journal of International Business Studies, Vol. 51 No. 4, pp. 538-576, doi: 10.1057/s41267-020-00318-w.
Miao, R., Bozionelos, N., Zhou, W. and Newman, A. (2021), “High performance work systems and key employee attitudes: the roles of psychological capital and an interactional justice climate”, International Journal of Human Resource Management, Vol. 32 No. 2, pp. 443-477, doi: 10.1080/09585192.2019.1710722.
Newton, P.M. (2024), “Design, run, and interpret survey-based research in the fields of academic integrity and misconduct”, in Eaton, S.E. (Ed.), Second Handbook of Academic Integrity, Springer International Handbooks of Education, Cham.
Nolan, E.M. and Morley, M.J. (2013), “A test of the relationship between person–environment fit and cross-cultural adjustment among self-initiated expatriates”, International Journal of Human Resource Management, Vol. 25 No. 11, pp. 1631-1649, doi: 10.1080/09585192.2013.845240.
O'Grady, S. and Lane, W.H. (1996), “The psychic distance paradox”, Journal of International Business Studies, Vol. 27 No. 2, pp. 309-333, doi: 10.1057/palgrave.jibs.8490137.
Okpara, J.O., Kabongo, J.D. and Lau, W.K. (2021), “Effects of pre departure and post arrival cross-cultural trainings on expatriates’ adjustment: a study of Chinese expatriates in Nigeria”, Thunderbird International Business Review, Vol. 63 No. 2, pp. 115-130, doi: 10.1002/tie.22185.
Ramalu, S., Subramaniam, Che, R., Kumar, N. and Uli, J. (2010), “Doing business in global arena: an examination of the relationship between cultural intelligence and cross-cultural adjustment”, Asian Academy of Management Journal, Vol. 15, pp. 79-97.
Ran, S. and Huang, J. (2019), “Enhancing adaptive transfer of cross-cultural training: lessons learned from the broader training literature”, Human Resource Management Review, Vol. 29 No. 2, pp. 239-252, doi: 10.1016/j.hrmr.2017.08.004.
Rashid, Y., Rashid, A., Warraich, M.A., Sabir, S. and Waseem, A. (2019), “Case study method: a step-by-step guide for business researchers”, International Journal of Qualitative Methods, Vol. 18, pp. 1-13, doi: 10.1177/1609406919862424.
Rugman, A.M. and Verbeke, A. (2000), “Multinational enterprises and public policy”, in Millar, C., Grant, M.R. and Choi, C.J. (Eds), International Business: Emerging Issues and Emerging Markets, Palgrave Macmillan, London.
Selmer, J. (2001), “The preference for pre-departure or post-arrival cross-cultural training”, Journal of Managerial Psychology, Vol. 16 No. 1, pp. 50-58, doi: 10.1108/02683940110366560.
Sheldon, M.K. and Elliot, J.A. (1999), “Goal striving, need satisfaction, and longitudinal well-being: the self-concordance model”, Journal of Personality and Social Psychology, Vol. 76 No. 3, pp. 482-497, doi: 10.1037//0022-3514.76.3.482.
Shenkar, O., Luo, Y. and Yeheskel, O. (2008), “From ‘distance’ to ‘friction’: substituting metaphors and redirecting intercultural research”, Academy of Management Review, Vol. 33 No. 4, pp. 905-923, doi: 10.5465/amr.2008.34421999.
Staw, M.B. (1976), “Knee-deep in the big muddy: a study of escalating commitment to a chosen course of action”, Organizational Behavior & Human Performance, Vol. 16 No. 1, pp. 27-44, doi: 10.1016/0030-5073(76)90005-2.
Su, Z.X., Wright, P.M. and Ulrich, M.D. (2018), “Going beyond the SHRM paradigm. Examining four approaches to governing employees”, Journal of Management, Vol. 44 No. 4, pp. 1598-1619, doi: 10.1177/0149206315618011.
Tsai, L. (2007), “Solidary groups, informal accountability and local public goods provision in rural China”, American Political Science Review, Vol. 101 No. 2, pp. 355-372, doi: 10.1017/s0003055407070153.
Tung, R.L. (2008), “The cross-cultural research imperative: the need to balance cross-cultural and other methodologies”, Journal of International Business Studies, Vol. 39 No. 1, pp. 41-46, doi: 10.1057/palgrave.jibs.8400331.
Tung, R.L. and Verbeke, A. (2010), “Beyond Hofstede and GLOBE: improving the quality of cross-cultural research”, Journal of International Business Studies, Vol. 41 No. 8, pp. 1259-1274, doi: 10.1057/jibs.2010.41.
Williams, W.D. and Gregoire, A.D. (2015), “Seeking commonalities or avoiding differences? Re-conceptualizing distance and its effects on internationalization decisions”, Journal of International Business Studies, Vol. 46 No. 3, pp. 253-284, doi: 10.1057/jibs.2014.52.
Yin, K.R. (2018), Case Study Research and Applications: Design and Methods, SAGE publications.
Acknowledgements
This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.