1 Introduction

In recent years, the information technology outsourcing (ITO) industry has matured into a large-scale global professional industry providing essential information technology (IT) products and services (Karimi-Alaghehband & Rivard, 2020). More and more companies today rely on outside third parties to provide critical information technology (IT) products and services, and the growing ITO trend is likely to continue into the future. For comparison, in 2022, the ITO market in the United States was estimated at $144.8 billion (Statista, 2022), whereas China, the world's second-largest economy, was forecast to reach an estimated market size of US $82.3 billion in 2027 (Business Wire, 2020). The global interest in ITO by both academia and practitioners has continued to escalate as IT has become more pervasive with the digital transformation of business (Susarla & Mukhopadhyay, 2019). As the use of outsourcing has increased, companies have turned to contracts as a means to manage the client-vendor relationship (Barthelemy, 2001; Berggren et al., 2001; Hundertmark et al., 2008). However, the formal contract may not be enough to manage the complex ITO relationship and position the arrangement for success.

Understanding outsourcing success has been a focus of theoretical and empirical research in information systems for more than 20 years. Over the years, academic researchers have paid significant attention to the role organizational commitment plays in outsourcing success between client and vendor firms (e.g., Karimi-Alaghehband & Rivard, 2020; Kranz, 2021; Oshri et al., 2015a, 2015b; Claybaugh & Srite, 2009; Goo et al., 2008; Koh et al., 2004; Sull & Spinosa, 2007). Against this backdrop, however, outsourcing presents new challenges to the vendor-employee relationship that has not been systemically investigated in the earlier ITO research. We believe that a grounded theory-based literature review, business observations, and best practices in the industry are needed to help investigate the evolving theoretical foundations of the area and create a roadmap for advancing theoretical and practical understanding of the key ITO phenomena (Vaast & Walsham, 2013; Wiesche et al., 2017).

By definition, the vendor, vendor employees, and the client firm are interconnected through outsourcing projects. In such a relationship, vendor employees experience labor relations and career activities being managed by the vendor employer, whereas the working content and the performance evaluation are often determined by the client firm according to the specific project requirements. Outsourcing allows the client firm to achieve the human resource purpose of "not for ownership, but for work" through project outsourcing, which leads to differentiation tensions between the vendor employees and the inner staff of the client firm (Xiao et al., 2015). In many firms, the ITO vendor employees are regarded as the core IT human resources of the client firm who have contracted with the vendor. The client firm relies heavily on the performance of the vendor employees since their quality, capability, work efficiency, and stability determine the success of outsourcing projects due to their direct services to the client firm and, in the end, reflect the core competitiveness of the vendor. With the client's heavy reliance on the performance of the vendor and, ultimately, the vendor's employees, the commitment level of the vendor employees to the vendor firm becomes critical. As ITO often leads to differentiation in treatment between vendor employees and the client employees with whom the vendor employees work on a daily basis, the situation is even more complicated. Unless the problem of organizational commitment between the vendor and its employee is settled, the performance and service capabilities delivered to the client firm will be significantly affected.

Thus, as client firms turn more and more to outsourcing, there is still a lot to learn with regard to the multi-faceted role of the contract that is applied to managing ITO. A significant amount of research has focused on the role of the contract in governing the outsourcing relationship between the client and vendor firm (e.g., Chang et al., 2017; Goo et al., 2009; Gopal & Koka, 2012; Gulati & Nickerson, 2008; Mani et al., 2012; Willcocks et al., 2011). Despite the rapid growth in ITO, information systems (IS) research has largely focused its attention on the formal transactional or financial outcomes designated in the formal complete contract (Ang & Straub, 1998; Aubert et al., 2004; Lacity & Willcocks, 1998; Wang, 2002). However, recent research suggests that there is more to contract management outside of the formal structure. Research has suggested that many outsourcing outcomes are impacted heavily by the relational components of the vendor-client relationship – a key differentiator for ultimate project success and failure (Dutta et al., 2017; Goo et al., 2008; Westner & Strahringer, 2010). In this regard, a growing number of studies have just begun to recognize the importance of both the transactional and relational components of a psychological contract included in the overall ITO agreement (Haried & Ramamurthy, 2009). Outsourcing studies have suggested that both relational and transactional/financial factors are necessary contractual components (Lee et al., 2004), and as a result, our study introduces and applies the psychological conceptualization of the contract into the context of ITO and commitment.

Recent studies have examined the complex interrelationships between transactional and relational governance mechanisms and their impacts on the ITO project success (e.g., Karimi-Alaghehband and Rivard, 2020; Kranz, 2021; Oshri et al., 2015a, 2015b), few studies to the best of our knowledge investigate the organizational commitment of vendor employees under different psychological contract matching modes. Thus, in this research, we follow the systematic procedure of the grounded theory (e.g., Vaast & Walsham, 2013; Wiesche et al., 2017; Wolfswinkel et al., 2013) and develop a theoretical framework to examine the key phenomena of ITO. The research aims to address the role of the relationship and expand the relational view to include the psychological contract, where the parties involved establish a mutual perception of the psychological expectations, including the responsibility for each other and the commitment involved in the reciprocal exchange (Coyle-Shapiro & Kessler, 2002). The study contributes to the ITO literature by examining the psychological contract applied to manage the unique vendor-employee relationship that becomes even more complicated with the involvement of the ITO client.

The paper is organized as follows. Based on empirical findings and theoretical perspectives of a grounded theory-based literature review, business observations, and best practices in the industry, we develop a theoretical framework to conceptualize the role of the psychological contract, upon which the research model and hypotheses are developed. Next, we discuss the data collection and measures. We then present a discussion of the results and concluding remarks.

2 Theoretical Foundation

The grounded theory traces its origin back to the work of Glaser and Strauss (1967), in which the methodology was defined as “an initial, systematic discovery of the theory from the data” (p. 3), serving as a highly inductive approach to theory building. The grounded theory has been well-accepted in IS research for literature review and theory development (Vaast & Walsham, 2013; Wiesche et al., 2017; Wolfswinkel et al., 2013). As the qualitative design and method are rather promising for investigating organizational dynamics and building theoretical perspectives for the newly emerging phenomena (Sarker et al., 2018), the grounded theory is especially appropriate for a new research discipline such as this research, where the critical ITO factors and models have yet to be fully substantiated.

This research thus relies upon the grounded theory as a literature review method and retrieves key literature on psychological contract and organizational commitment in prior ITO studies and business observations and best practices in the industry to produce an ITO theoretical framework in the organizational context. Upon synthesis of theoretical perspectives and empirical findings of the literature review and business observations and best practices in the industry, we identify critical factors and interrelationships and develop the following theoretical framework to theorize the role of the psychological contract.

2.1 Theoretical Framework

ITO is defined as the practice of transferring IT assets, leases, staff, and management responsibility for the delivery of IT services from the internal department to third-party vendors (Hirschheim & Lacity, 2000). The literature has investigated reasons for ITO failures and difficulties surrounding the complexity and causal ambiguity of coordinating projects across diverse outsourcing activities, particularly in the relationships between a client firm and its outsourcing vendors (e.g., Barthelemy & Quelin, 2006; Poppo & Zenger, 2002; Ravindran et al., 2015). Many ITO studies have also focused on the role of the contracts from a transactional or a financial lens (e.g., Ang & Straub, 1998; Aubert et al., 2004; Lacity & Willcocks, 1998; Wang, 2002). However, a growing stream of literature has developed highlighting the relational view of the ITO contract (e.g., Dutta et al., 2017; Haried & Ramamurthy, 2009; Palvia et al., 2011; Westner & Strahringer, 2010; Goo et al., 2008). Over the last few decades, from the ITO partnership perspective, IS literature has identified major factors that have significant impacts on ITO success, such as partnership quality, commitment and trust, contract scope, and contract duration (Karimi-Alaghehband and Rivard, 2020; Grover et al., 1996; Lee et al., 2004). Meanwhile, in the business environment, there is an evolving shift where the client firm seeks to structure ITO initiatives as a partnership for long-term value creation, focusing on innovation and strategic performance rather than arms-length transactional arrangements (Kranz, 2021; Lacity & Willcocks, 2013; Oshri et al., 2015a, 2015b). As outsourcing levels increase, there is a need to investigate how the traditional relationship between an employer organization and its employee has changed due to outsourcing (Könning et al., 2019). Accordingly, in the ITO industry, the vendors, customers/client companies, and vendor employees are interconnected through outsourcing arrangements, and the dynamics involved have shifted the employment role of the vendor and its employees. The core of ITO is delegating work from the client firm to the vendor for the client firm. As such, the work's quality often depends on the vendor employee's performance and interpretation of the work expectations (Barrett & Oborn, 2010). When evaluating the client-vendor relationship, the vendor firm manages the labor relations and human resource activities for the employees, but the work content and evaluations are often determined by the client firm according to the specific project requirements. Considering the salary, work environment, and promotion training cannot meet the expectations of vendor employees, the turnover rate of professional and technical staff remains high, hindering the development and overall performance of outsourcing firms (Lo, 2013). In many cases, business contracts are applied as a universal tool to manage the complicated relationships that inevitably emerge within the process of ITO creation and delivery.

Against the backdrop of the growing global ITO industry, this research focuses on the developing outsourcing industry in China. Chinese ITO vendors provide an attractive research context to investigate ITO relationships because, compared to the multinational vendors and long-established Indian vendors, Chinese vendor firms are still in an early stage and are working to move up the value chain (Su, 2013, 2015). Earlier studies on Chinese ITO vendor firms suggest a need for vendors to select an appropriate capability development approach in order to move up the value chain (Du et al., 2020). Given that clients are expecting ITO vendors to handle more sophisticated tasks (Oshri et al., 2015a, 2015b, 2018a, 2018b), vendors are under increasing pressure to perform. Thus, guided by the two concepts—psychological contract and organizational commitment, we conducted a cross-sectional survey of multiple Chinese ITO vendor firms to explore the organizational commitment of vendor employees under different psychological contract matching modes. In this manner, we address the need to examine the role of the psychological contract as a tool in developing organizational commitment and positioning ITO for success.

3 Research Model And Hypothesis Development

Drawing upon the theoretical framework, we develop a 2 × 2 (i.e., (Psychological Contract: Relational vs. Transactional) x (Stakeholder: Vendor vs. Employee)) model to explore the organizational commitment of ITO vendor employees under different psychological contract matching patterns (Fig. 1). The four types of matching patterns include the following: Employee Dominating (ED), Vendor Dominating (VD), Mutual Relational (MR), and Mutual Transactional (MT). Among the four types, mutual relational and mutual transactional belong to the matching mode, whereas employee-dominating and vendor-dominating to the mismatching pattern. The specific definitions are as follows:

  1. (1)

    Employee Dominating: The employee adopts a relational psychological contract, while the vendor adopts a transactional psychological contract.

  2. (2)

    Vendor Dominating: The employee adopts a transactional contract, while the vendor adopts a relational psychological contract.

  3. (3)

    Mutual Relational: Both the employee and vendor adopt a relational psychological contract.

  4. (4)

    Mutual Transactional: Both the employee and vendor adopt a transactional psychological contract.

Fig. 1
figure 1

Employee-Vendor Psychological Contract Model 

Full size image

3.1 Organizational Commitment

In the light of the theoretical framework and the above research model, the organizational commitment construct measures the employee's willingness to make endeavors for an organization, the desire to stay in the organization, and the acceptance of organizational goals and values (Igbaria & Wormley, 1992; Porter & Lawler, 1968). Organizational commitment has been shown to be critical in the ITO industry. Without organizational commitment, the vendor-employee relationship is doomed to fail (Claybaugh & Srite, 2009; Goo et al., 2008; Koh et al., 2004; Sull & Spinosa, 2007). Furthermore, as organizational commitment is explicitly defined as "the relative strength of an individual's identification with, and involvement in a particular organization" (Mowday et al., 1979, p. 226), three defining characteristics stand out: (1) a high degree of commitment and acceptance of organizational goals and values (recognition), (2) the employee's willingness to make efforts for to achieve organizational goals (input), and (3) the employee's desire to maintain organizational membership (loyalty) (Porter et al., 1974). Thus, organizational commitment reflects the degree of the employee's psychological attachment to the organization.

The benefits of organizational commitment are well documented in the literature. Various approaches have been developed to investigate factors affecting an employee's organizational commitment. Studies suggest that organizational factors and characteristics such as communication, fairness, interpersonal relationships (Goo et al., 2008; Haried & Ramamurthy, 2009; Lacity et al., 2011), training (Fontinha et al., 2014;), corporate culture (Lee, 2001; Rottman & Lacity, 2006), and development opportunities (Chaudhuri & Bartlett, 2014) significantly affect an employee's organizational commitment. Accordingly, Dias and Silva (2016) indicated that insufficient attention to an employee’s career development leads to low organizational commitment. This necessitates important actions to address low organizational commitment. Research also suggests that employees with higher organizational commitment have a lower turnover rate and higher work performance than those with lower organizational commitment (Mowday et al., 1979), making employees' organizational commitment increasingly important in today's highly competitive labor environment. Particularly in the context of project management, stakeholders' organizational commitment is ranked as one of the most critical factors for project success (Hyväri, 2006). Given the highly recognized importance of organizational commitment and its significance in affecting outsourcing relationship performance, both ITO practitioners and researchers have a strong need to understand how the psychological contract influences ITO relationships.

3.2 Psychological Contract

According to the theoretical framework and our research model, the role of the outsourcing contract is one of the most important and widely researched topics, where many of the problems can be traced to the effectiveness and management of the contract (Barthelemy, 2001; Berggren et al., 2001; Hundertmark et al., 2008). Accordingly, many outsourcing decisions have traditionally focused on the transactional or financial outcomes designated in the formal complete contract (Ang & Straub, 1998; Aubert et al., 2004; Lacity & Willcocks, 1998; Wang, 2002). Hart (1988) defined a complete contract as outsourcing firms writing a detailed, complete contract to cope with all possible contingencies that might be caused by bounded rationality and uncertainty. Particularly when outsourcing is viewed as a permanent strategy for a client firm, the client relies heavily on the complete contract (Aubert et al., 2017). A growing area of research has shown that outsourcing relationships that rely on the complete contract are associated with the effectiveness of contract management (Kim et al., 2013; Srivastava & Teo, 2012) and contractual strategic benefits (Balaji & Brown, 2014). However, upon the client's heavy reliance on outsourcing, outsourcing outcomes are determined heavily by the relational components of the vendor-client relationship – a key differentiator for ultimate project success and failure (Dutta et al., 2017; Goo et al., 2008; Westner & Strahringer, 2010). In this regard, a growing number of studies have just begun to recognize the importance of both the transactional and relational components of a psychological contract included in the overall ITO agreement (Haried & Ramamurthy, 2009). As prior outsourcing studies have suggested that both relational and transactional/financial factors are necessary contractual components (Lee et al., 2004), our study introduces and applies the psychological conceptualization of the contract into the context of ITO.

Our theoretical framework and research model both suggest the essence of the psychological contract is that the stakeholders who are involved and sign the contract establish mutual perceptions of the psychological expectations of the contract, including work responsibilities of each other and commitment that are essential in reciprocal business exchanges (Rousseau, 1982). Depending on different organizational concerns and perspectives, a contract can be viewed as relational or transactional (Anderson & Schalk, 1998; Rousseau, 1990). The relational view focuses on the emotional basis, trust, and obligation of the two contractual parties of stakeholders. In contrast, the transactional lens aims to reach an agreement between the two parties upon transactional and financial indicators, while the private intimate relationship is not taken into account. Research has suggested that the point of view of a psychological contract may be different among the internal staff of the firm and its external contracted employees, and the point of view of the contract influences the working attitudes of employees (Yan et al., 2002; Wang & Jiang, 2005; Men & Robinson 2018; Coyle-Shapiro & Kessler, 2002). For example, studies found that both obligations and fulfillment of obligations of employees correlate highly with those of the employer firm (Coyle-Shapiro & Kessler, 2002). Additional research has extended the viewpoint of the psychological contract to examine the relationship between vendors (organizations) and external customers (individuals) (Oshri et al., 2015a, 2015b; Kranz, 2021) and particularly to explore the relationship between the ITO vendor and client firms (Koh et al., 2004; Oshri et al., 2015a, 2015b; Kranz, 2021).

Our literature review shows that the psychological contract is often viewed as the internal cause of both organizational commitment and job satisfaction for explaining the relational changes between the two parties (Robinson, 1996; Morrison & Robinson, 1997; Robinson et al., 1994; Turnley & Feldman, 1999) . Interestingly, as ITO adds new dynamics and tensions to the organizational environment, the traditional employee-employer relationship is challenged. For example, the ITO vendor's employees now engage in the position responsibilities and manage the relationship with their unique expectations and commitment for both the client and the vendor firms. From the employee's perspective, a relational psychological contract positively influences both task and contextual performance (Long et al., 2015). Moreover, Lioliou et al. (2014) show that a relational psychological contract correlates higher with an employee’s emotional, continuous, and normative organizational commitment than a transactional psychological contract. We thus propose the following hypotheses:

  • H1: Overall, the employee’s relational psychological contract correlates higher with organizational commitment than a transactional psychological contract.

  • H1a: In the matching pattern, the employee’s relational psychological contract correlates higher with organizational commitment than a transactional psychological contract.

  • H1b: In the mismatching pattern, the employee’s relational psychological contract correlates higher with organizational commitment than a transactional psychological contract.

In light of the theoretical framework and our research model, the story of the employee-vendor relationship can be told differently from the vendor's perspective. According to Coyle-Shapiro and Kessler (2002) and Meyer and Allen (1991), employer evaluations of organizational commitment are mainly based on their obligations to the employee. The obligations of the employer have a positive effect on the fulfillment of obligations of employees. Meanwhile, as a major component of organizational commitment, affective commitment has a stronger positive influence on organizational commitment than normative or continuance commitment (Jaros, 1997; Ying & Shengtai, 2009). As such, the vendor’s relational psychological contract can enhance the affective obligations of employees and thus make more contributions to organizational commitment, improving the overall level of employee obligation fulfillment. Consequently, the vendor’s psychological contract can be viewed as a major factor in evaluating organizational commitment. The above arguments lead to the following hypotheses:

  • H2: Overall, the vendors' relational psychological contract correlates higher with organizational commitment than a transactional psychological contract.

  • H2a: In the matching pattern, the vendors' relational psychological contract correlates higher with organizational commitment than a transactional psychological contract.

  • H2b: In the mismatching pattern, the vendors' relational psychological contract correlates higher with organizational commitment than a transactional psychological contract.

3.3 Organizational Commitment under the Matching Situation

In light of our research model, the proposed psychological contract model describes two situations where the vendor firm and its employees experience a matching condition within either a transactional or a relational psychological contract. A matching condition means that the perceptions of a psychological contract by the vendor and employees of the vendor firm match each other. In a situation where the vendor-employee psychological contract reaches equilibrium, both parties maintain an interdependent relationship in a mutual relational situation (Rousseau, 2000). Under the pattern, the IT vendor usually has an established care system for its employees. Meanwhile, employees are also willing to focus their care and attention more on their vendor firm, resulting in a higher level of employee loyalty and a lower turnover rate (Li & Sun, 2006; Coyle-Shapiro and Kessler, 2002). Earlier studies have found that employees are more likely to resign when the psychological contract balance is broken (Lirong et al., 2015). In the condition of the imbalance in the psychological contract, it may cause the gradual decline of organizational commitment of employees, leading to a higher turnover rate. Therefore, we propose:

  • H3: The employees' organizational commitment in the matching condition is higher than that in the mismatching condition.

  • H4: In the matching condition, organizational commitment in the mutual relational situation is higher than that in the mutual transactional situation.

3.4 Organizational Commitment under the Mismatched Situation

According to our research model, an employer's failure to meet or fulfill its obligations is more significantly associated with the decline of its employee’s relational obligations (Robinson et al., 1994). As an important factor of the employee’s psychological contract, relational obligations are more susceptible to negative effects, including those of a situation where the vendor-employee psychological contract mismatches – that is, employees perceive a relational psychological contract, whereas the vendor employer holds a different transactional contract view. In this condition, ITO vendors pay less attention to organizational communication or the working environment in which employees' relational needs are satisfied, leading to their organizational commitment being damaged. Thus, we propose:

  • H5: In the mismatching condition, organizational commitment in the employee-dominating situation is higher than that in the vendor-dominating situation.

4 Research Method

4.1 Meaurement Development

In order to test the research hypotheses of this study, we turn to the empirical findings and theoretical perspectives of our grounded theory-based literature review and develop a survey instrument for data collection. Most of the measurement items of the survey were adapted from existing literature on psychological contracts (Rousseau, 1998, 2000) and organizational commitment (Porter, 1974; Mowday et al., 1979).

Our instrument design is based on the Rousseau (1998, 2000) construct measurement items for the survey. The initial questionnaire included 35 items, aiming to identify three types of psychological contracts. Considering the actual situation of Chinese corporate employees, Li and Guo (2002) further screened the items of the psychological contract questionnaire. At the same time, they divided the psychological contract patterns into four types, consisting of two dimensions for employees and two dimensions for organizations. We further categorized the psychological contract questionnaire used by previous scholars into four types: enterprise relational, enterprise transactional, employee relational, and employee transactional. The questionnaire items for the enterprise-type psychological contract are all from the perspective of the company, measuring the importance attached by the company to aspects such as corporate atmosphere, employee opinions, communication mechanisms, and salary benefits. On the other hand, the questionnaire items for the employee psychological contract are from the perspective of employees, measuring their importance attached to aspects such as salary benefits, company culture, and work atmosphere.

The organizational commitment measurement items used in this study are derived from Porter et al. (1974) and Mowday et al. (1979). The measurement of organizational commitment needs to be approached from three perspectives based on previous scholars' research: affective commitment, continuance commitment, and normative commitment. Affective commitment refers to employees strongly endorsing the company's values and goals; continuance commitment reflects employees' willingness to continue working in the current organization in the future; normative commitment involves employees caring about the future direction of the company and being willing to invest more effort in the company's development.

Study design and pilot studies were carried out following the standard procedure in the literature (e.g., MacKenzie et al., 2011; Moore & Benbasat, 1991), including developing the conceptual definitions of the constructs, item adaptation, and modification to fit the current study, assessment of content validity of measurement items, measurement model specification, and assessments of the structural model. All measurement items were measured using a seven-point Likert-type scale (1 = "Strongly disagree" to 7 = "Strongly agree"). Table 10 in Appendix 1 presents the principle constructs and measurement items encompassed in this study.

4.2 Pilot Study

All measurements followed a refinement procedure with several pretests. The instruments were preliminarily tested and revised by four IS faculty, and one IS master student for clarity and accuracy. Then a pilot study was conducted by inviting individuals with outsourcing project experience to complete the questionnaire. Thirty-two questionnaires were distributed in the pilot study. Based on feedback and recommendations, the instrument was refined for wording clarification. The pilot assessment of the reliability of the instrument was then performed. The items that failed the test or were highly cross-loaded on other constructs were dropped. In the pilot, the respondents were also asked to make notes of problematic survey items. Based on the notes, modifications were made mainly to the layout and phrasing of the survey questions.

4.3 Survey Administration

Data collection followed the standard process of random sampling. The questionnaires were distributed to employees of ITO vendors through printed mail and a URL in WeChat as well. In total, 650 questionnaires were sent out, and 562 valid responses were collected. The valid return rate was 86.8%.

The questionnaire consisted of demographic characteristics of employees, including working years, position, and vendor characteristics, including qualifications and the total number of employees. The sample distribution is shown in Table 1.

Table 1 Sample distribution 
Full size table

The number of employees surveyed showed a considerable level of dispersion. Specifically, 22.52% of the companies had fewer than 50 employees, indicating that they were relatively small in scale. Meanwhile, 42.73% of the companies had employee counts ranging from 50 to 500, and 16.49% had employee counts between 500 and 1000. Furthermore, 18.26% of the companies had over 1000 employees. Regarding the question of whether the surveyed IT companies had CMM/CMMI certification, 37.77% of the companies had the certification, while a larger proportion (62.23%) of the companies did not have this certification. This indicates that there was still room for further development in terms of qualification certification and scale for IT service outsourcing companies. Although some companies had reached a certain scale, many others were continuously growing and expanding. In addition, the analysis of the surveyed vendor employees can be summarized as follows: Most of the vendor employees in the sample came from middle and lower-level managers or regular staff positions, accounting for 30.5% and 51.6% of the total sample, respectively. This finding can better illustrate the situation of mid-level managers and implementers in the subsequent study of organizational commitment among vendor employees.

Furthermore, the work experience of the surveyed vendor employees was consistent with their job positions' distribution. The majority of them had work experience of less than 5 years, with 50.35% having less than 3 years of experience and 31.38% having 3 to 5 years of work experience in the total sample.

We tested the differences between the mail and online respondents. The results indicate that there were no significant differences between the two groups of respondents regarding the demographic dimensions and principal constructs (Armstrong & Overton, 1977; Pavlou & Fygenson, 2006). We also assessed the nonresponse bias of the dataset with two tests (Pavlou & Fygenson, 2006). The results of the first test showed that there were no significant differences across demographic groups upon comparing CMM/CMMI qualification levels, number of employees, position, and working experience. In the second test, we tested the mean differences of principal constructs across the demographic groups, which also showed no significant differences at the significance level, p < 0.05.

4.3.1 Data Analysis

In this study, the structural equation modeling (PLS-SEM) approach was used to analyze the data and validate both the measurement and structural models (Hair et al., 2017). The PLS-SEM approach was chosen due to its ability to test structured causal relationships among constructs with multiple measurement items and validate the reliability and validity of questionnaires with the analytical advantages of Partial Least Square Regression (PLS). Since 2005 when SmartPLS was released (Ringle et al., 2005), the PLS-SEM analysis using SmartPLS has become a key multivariate analysis method in IS research. We chose SmartPLS 3.3.3 as our PLS-SEM software (Ringle et al., 2021).

4.3.2 Measurement Model

The reflective scales of the survey were assessed through conventional tests of convergent validity, discriminant validity, and instrument reliability. Prior IS research has provided well-established criteria and rules to test convergent validity and discriminant validity (MacKenzie et al., 2011). Internal consistency and reliability were evaluated using composite reliability and average variance extracted (AVE) (Table 2) (Fornell & Larcker, 1981). As is shown in Table 2, Cronbach's α coefficients of all latent variables were above 0.895. The values of Composite Reliability (CR) were all above 0.920. The values of Average Variance Extracted (AVE) were all above 0.670. All estimates are significant at p < 0.001, indicating the adequacy of the said validity and reliability of the questionnaire scales of the study.

Table 2 The loads, Cronbach's α, AVE, and CR of the questionnaire scale
Full size table

Convergent validity refers to the extent to which the measures of a variable act as if they are measuring the underlying theoretical construct because they share variance (Schwab, 1980). Our study assessed the convergent validity using the PLS-SEM internal consistency score. Under the assessment, all factor loadings should exceed 0.7, and the average variance extracted (AVE) for each construct should exceed 0.5 (Fornell & Larcker, 1981). Moreover, discriminant validity is used to measure the extent to which measures of two constructs are empirically distinct (Davis 1989). For verifying discriminant validity, the square root of each construct's AVE should be higher than its correlations with other constructs (Chin 2003). As shown in Table 3, the square roots of all AVEs are much larger than all the cross-correlations.

Table 3 Discriminant validity and correlation coefficients between the latent variables
Full size table

In recent years, the PLS-SEM has undergone significant enhancements in the estimation accuracy of path coefficients (Hair et al., 2017) and "consistent PLS (PLSc) to estimate linear and nonlinear latent variable models, a bootstrap-based test to statistically assess overall model fit, measures of overall model fit, and the heterotrait-to-monotrait (HTMT) ratio of correlations to assess discriminant validity" (Benitez et al., 2020, p. 2). In Table 4, we report the values of the heterotrait-to-monotrait (HTMT) of the study. As is shown, all values are lower than 0.90, indicating the discriminant validity of the reflective scales of the study (Benitez et al., 2020). The above-referenced tests demonstrated that all measures have adequate convergent and discriminant validity.

Table 4 Discriminant validity and heterotrait-to-monotrait (HTMT) values
Full size table

5 Model Assessment And Hypothesis Test

5.1 Identification of Psychological Contract Matching Patterns

This study adopted the following procedure to identify the psychological contract matching patterns. We calculated the weights of related factors based on the standard factor loadings in Table 2. The calculation formula is shown in Eqs. (1)-(2) below. Among them, Eq. (1) shows the quadratic sum of all factors' contribution degree for a latent variable equal to 1. Equation (2) demonstrates the sum of all factors' weight for a latent variable equal to 1. fij implies the standard loadings of the factor j on the latent variable i. wij implies the weight of the factor j on the latent variable i. We then used the method of weighted averages to obtain the aggregate scores of each latent variable based on the calculated weights and factor scores.

$${S}_{ij}={f}_{ij}/\sqrt{\sum_{j=1}^{n}{f}_{ij}^{2}}$$
(1)
$${w}_{ij}={S}_{ij}/\sum_{j=1}^{n}{S}_{ij}$$
(2)

Next, we compared the difference in the latent variables between the transactional and relational psychological contracts for both employees and vendors. If the absolute value of the difference was less than 0.1, we determined that the identification of the psychological contract matching pattern was not clear. Only when the vendor's and employee's psychological contract matching pattern was clear would the identification of the psychological contract for an ITO vendor succeed. For each investigated individual, if the score of the vendors' transactional psychological contract was 0.1 greater than the vendors' relational psychological contract, the individual considered the vendor's psychological contract as transactional; otherwise, we considered it as relational. The same procedure was applied to the employee's psychological contract. The results of identification are shown in Table 5 and 6.

Table 5 The Summary of The One-Sided Psychological Contract Matching Patterns
Full size table
Table 6 The Summary of The Two-Sided Psychological Contract Matching Patterns
Full size table

5.2 Research Variables

We set the latent variables of employees' and vendors' psychological contracts as dependent variables. Further, variables such as working experience, management position, Capability Maturity Model/ Capability Maturity Model Integration (CMM/CMMI) qualification, and the total number of employees were taken as independent variables separately. Through the ANOVA analysis, we found some latent variables showed significant differences from those independent variables. Therefore, we set these latent variables as control variables. Among them, "over five years: was regarded as the reference variable of working years; "ordinary staff" was regarded as the reference variable of position; "greater than 500" was regarded as the reference variable of the total number of employees; "not passed in CMM/CCMI certification" was regarded as the reference variable of qualification.

5.3 Structural Model Testing

This section reports the assessment of the structural model of the study, including estimates of path coefficients and R-square values. Among the assessments, path coefficients indicate the strength, as well as the statistical significance of relationships between independent and dependent variables, and R-square values refer to the variance explained by the independent variables (Chin, 1998). We also evaluate the overall fit of the structural models through the bootstrap-based test of overall model fit and SRMR.

We tested the base model of four psychological contract types and their influences on organizational commitment (Fig. 2). The path coefficients show that the vendor transactional and relational psychological contracts and the employee relational psychological contracts all have significant positive impacts on organizational commitment. However, the testing results did not support the hypothesized positive association between the employee transactional psychological contract and organizational commitment. As to the model fit, the test shows that the value of SRMR (0.058) was below the suggested threshold of 0.080 (Benitez et al., 2020), indicating an acceptable model fit such that the proposed model is well suited for confirming and explaining the relationship between psychological contract types and organizational commitment.

Fig. 2
figure 2

Testing Results of the Base Model. (***p < 0.01, **p < 0.05; *p < 0.1;)

Full size image

In the following section, three models were tested by adding the pattern-matching variables. In Model 1, the dummy variable was added to verify whether the employees' and vendors' psychological contract of the sample matched and the reference variable was "mismatch". In Model 2, the dummy variable was added to represent the matching pattern when employees' and vendors' psychological contract were mismatched, and the reference variable was "vendor-dominating". In Model 3, the dummy variables were added to show the matching pattern in the matching situation, and the reference variable was "mutual transactional". As shown in Table 7, the R Square of the three models is all above 0.54, indicating that all models were capable of explaining the changes in organizational commitment.

Table 7 Indicators of model fit
Full size table

Using Bootstrap testing of SmartPLS 3.3.3, the path coefficients and related T-values were detected and reported in Fig. 3 below. The T-value of "vendors' and employees' psychological contract matching- > organizational commitment" in Model 1 is 0.517, suggesting that the influence of “vendors' and employees' psychological contract matching" on organizational commitment was not significant.

Fig. 3
figure 3

Testing Results of Model 1. (*p < 0.1; **p < 0.05; ***p < 0.01)

Full size image

Moreover, as is shown in Fig. 4, the path coefficient of "employee-dominating in mismatching situation- > organizational commitment" is 0.178, and the T-value is 1.988, indicating that the employee-dominating pattern has a more significant positive influence on organizational commitment than the vendor-dominating pattern.

Fig. 4
figure 4

Testing results of Model 2 (Mismatching Situation). (*p < 0.1; **p < 0.05; ***p < 0.01)

Full size image

In the same vein, the testing results of the third model are shown in Fig. 5. The path coefficient of "mutual relational—> organizational commitment" is 0.064, and the T-value is 2.654, indicating that the mutual relational pattern has a more significant positive effect on organizational commitment in the matching situation. That is, under the same condition, the organizational commitment of the mutual relational pattern is slightly higher than the mutual transactional one.

Fig. 5
figure 5

Testing results of Model 3 (Matching Situation). (*p < 0.1; **p < 0.05; ***p < 0.01)

Full size image

Additionally, we ran the T-test in the advantage of Lin's research method (Lin, 2011) and derived the significance of the path coefficient differences with the path coefficients and standard deviations from the Bootstrap testing formula (Chin, 2000). The formula is as shown in Eq. (3) below, where l1 and l2 represent different path coefficients, respectively. The testing results are shown in Table 8 to be followed.

Table 8 Comparison of path coefficients in matching and mismatching situations
Full size table
$${\text{t}}=\frac{{l}_{1}-{l}_{2}}{\sqrt{\frac{{({n}_{1}-1)}^{2}\times {SE}_{1}^{2}-{({n}_{2}-1)}^{2}\times {SE}_{2}^{2}}{{n}_{1}+{n}_{2}-2}}}\times \frac{1}{\sqrt{\frac{1}{{n}_{1}}+\frac{1}{{n}_{2}}}}$$
(3)

As is shown in Table 8, under the matching situation, the vendors' relational psychological contract has a significantly stronger influence on organizational commitment than the transactional psychological contract. In the mismatching situation, however, it was not significant. As for the employees' psychological contract, the T-values of the two models (situations) were 3.103 and 3.302, respectively, and the differences in the path coefficients are 0.847 and 0.494, respectively, indicating that employees' relational psychological contract is more effective than the transactional contract for organizational commitment. Overall, as is shown in Table 9, all the hypotheses were supported except for H2b and H3.

Table 9 Hypothesis testing results
Full size table

6 Discussions

In the context of the ITO relationship, this study investigated the impact of the psychological contract on organizational commitment across different contractual matching patterns between the ITO vendor firm and employees of the vendor firm. Drawing upon empirical findings and theoretical perspectives of the grounded theory-based literature review of psychological contract and organizational commitment, business observations, and best practices in the industry, we theorized the matching patterns of the vendor-employee psychological contract into four types: Vendor Dominant, Employee Dominant, Mutual Relational, and Mutual Transactional. We specified the theoretical patterns in the context of the ITO industry and developed our research hypotheses. The survey data was collected from emerging Chinese ITO firms. Our findings demonstrate that the vendor employees' relational psychological contract has a significantly higher positive influence on organizational commitment than the transactional psychological contract. This remains true for both matching and mismatching contractual modes.

6.1 Theoretical Contributions

In general, our findings contribute substantially to the limited research on psychological contracts and organizational commitment in ITO theory building. Our findings on the matching patterns of the psychological contract are valuable because this perspective demonstrates how the different relational psychological contract types significantly impact employee organizational commitment to the ITO vendor organization. This finding suggests that commitment to an organization is intertwined with maintaining a mutually reciprocal relationship between employees and ITO vendors.

First, our findings show that no matter whether the psychological contract between the organization and its employees matches or not, the employees' relational psychological contract significantly contributes to their organizational commitment to the employer organization more than a transactional psychological contract does. According to (Robinson, 1996; Turnley & Feldman, 1999), an employee's relational psychological contract encourages them to consciously and practically agree with the culture and values of the organization in which organizational commitment is deeply rooted. Our study confirmed that relational contracts are characterized by high levels of commitment to the organization (McDonald & Makin, 2000; Morrison & Robinson, 1997).

Second, our findings demonstrate that when a psychological contract from the outsourcing vendor's perspective matches the perspectives of the employees, compared to a transactional psychological contract, the level of the employees' organizational commitment increases under the circumstance that the relational psychological contract is strengthened. A relational psychological contract on the employees' side is perceived as the obligation to their employer of loyalty, as employees wish to build a long-term relationship with their employers. This finding is consistent with (McDonald & Makin, 2000; Rousseau, 1989).

In addition, the data analysis of the dummy variable demonstrates that the influence of the employee-dominating model on organizational commitment is significantly stronger than that of the vendor-dominating one. Under an unbalanced organization-dominating situation in which the employee shares the relational psychological contract where the vendor does not take much account of the working environment or the employee's perspectives, the aftermaths decrease or even ruin organizational commitment. The flip side of the coin is that, within the employee-dominating context with a reasonable salary, the vendor's relational psychological contract can significantly promote the process of perceiving, accepting, and agreeing on vendor organizational culture for employees in the long term. The findings and observations are relatively consistent with those of Men and Robinson (2018).

6.2 Practical Implications

The goal of this research was to develop an improved understanding of the impact the psychological contract has on organizational commitment across different matching patterns between the outsourcing vendor and the vendor's employees. For ITO vendors who are interested in establishing a cooperative win–win working environment, the findings of this study hold several critical managerial implications for ITO vendors. Upon the matching pattern of the psychological contract, ITO employees are generally satisfied with their present working conditions. Under the situation and given the vendor pays appropriate attention to the employees' expectations and enhances their decision engagement, the employees' work initiative can be stimulated higher, which further boosts their organizational commitment to the ITO vendor. However, upon the mismatch pattern of the psychological contract, the ITO vendor needs to recognize whether their employees' core needs are "economic" in nature or derived from long-term reciprocal expectations to increase employees' organizational commitment and decrease the turnover rate among employees (McDonald & Makin, 2000; Men & Robinson, 2018).

Furthermore, when taking into account the complicated dynamic process of contractual matching and/or mismatching, the findings are rather interesting. Our findings indicate that the employee's organizational commitment under the matching condition is not necessarily higher than that under the mismatching condition. The finding in this regard has unique implications for both ITO academia and business practice. Currently, the market of ITO is booming as vendor firms are expanding, and the number of IT professionals is growing in the workforce. Moreover, due to the increasingly competitive business environment in the ITO industry, vendors need to offer employees stronger and more comprehensive benefits and wage packages. This leads to ITO professionals under a transactional psychological contract being more likely to job-hop. Similarly, ITO vendors under the same transactional psychological contract tend to search for a workforce of more economical – an ITO trend that further accelerates employee turnover (Men & Robinson 2018; Priya & Sundaram, 2016). Thus, under the situation of the mutual transactional psychological contract, ITO vendors are suggested to add value to the job offer, salary, promotion, and training opportunities to mitigate employee turnover and, even worse, the organizational brain drain (Fontinha et al., 2014; Men & Robinson, 2018).

Lastly, our findings show that, under the matching situation, the vendor’s psychological contract of the mutual relational model exercises a more positive impact on organizational commitment than that of the mutual transactional model. When the mutual relational matching pattern is present, the vendor employees are concerned more about affective factors, by which they appreciate more attention from the employer and desire more internal communication. In turn, the ITO vendor holding the same psychological contract is more likely to establish a cooperative win–win working environment and involve them in internal activities. All these efforts significantly help to meet employees' expectations and eventually match them with organizational values and culture (Fontinha et al., 2014; McDonald & Makin, 2000).

6.3 Research Limitations and Future Works

In any research project, choices made by the researchers create limitations in interpreting the results. Any limitation encountered was addressed to guard against and minimize impact. Limitations encountered consisted of the limited sample size and the analysis of only Chinese vendor firms. This limited sample size from Chinese vendors suggests a call for caution in generalizing the findings. Since Chinese vendor firms are still in an early stage and striving to move up the value chain (Su, 2013, 2015), there is a future research opportunity to compare our findings to more established ITO vendors such as those located in India.

Future research has an opportunity to compare our findings to other ITO vendor locations to examine if culture plays a role in the impact of the psychological contract. We encourage future studies to generalize the psychological contract findings of this study to review organizational innovation in the ITO industry. In this regard, the research of Susarla and Mukhopadhyay (2019) demonstrates that the design of the outsourcing contract is critical to fostering innovations that are produced through an ITO relationship. Future work is therefore encouraged to review how vendor-employee psychological contracts can be engaged to drive organizational innovations, especially in the ITO industry. These limitations notwithstanding, we believe that the study has provided meaningful insights into the importance of the psychological contract in ITO. These and other limitations should be kept in mind when considering the findings.

7 Conclusion

Overall, this paper reviews the impact of the psychological contract on organizational commitment across different matching patterns between the vendor and the vendor's employees. The study's expanded view of the psychological contract offers some insights into how the vendor and vendor-employee perspectives differ based on psychological contract factors. Our findings demonstrate that the employee's relational psychological contract has a higher positive influence on organizational commitment than an employee's transactional psychological contract in both matching and mismatching situations. We urge both researcher and practitioner communities to build on our findings to increase our understanding of the role the psychological contract plays in organizational commitment.