This study aimed to examine the relationship between temporary employees and employers (a non-dyad) role clarity and psychological contract in the Nigerian service industry. To address this aim, the following specific research questions were developed and highlighted below:

This study adopted a positivistic paradigm. This positivist worldview is epistemologically based on the assumption of objective reality that can be identified, understood and measured, such that the functional relationship can be derived from the causal and explanatory factors between the independent and the dependent variables (Park, Konge & Artino 2020; Schrag 1992). This study ontologically uses a hypothetical deductive approach to test hypotheses that are quantitative in nature (Schrag 1992). That is, this study adopted a quantitative approach underscored by a positivistic paradigm to examine the relationship between the constructs as perceived by employees and employers in the temporary employment relationship. Considering the primary objective of this study, the research design involved two studies, which included the temporary employee and employer perspectives in Nigeria, and this is in tandem with the existing research design adopted by Kim, Wang and Chen (2018). These two studies specifically adopted a survey research design that is cross-sectional in nature. The choice of survey research design was based on the quantitative description and tests for relations between variables (Creswell & Creswell 2018), and it is cross-sectional in nature because data were collected at one point in time (Creswell & Creswell 2018; Neuman 2014).

This study included employees and employers in temporary and direct contracts in the formal and informal service industries in Nigeria. For study 1: temporary employees in the exchange relationship, this direct temporary employee was considered because they do not create ambiguity of double employers with mutually incompatible obligations and competing expectations between the temporary agent and the client organisation (Handy et al. 2020). In addition, this study considered temporary employees because extant laws in Nigeria do not adequately protect them (Otuturu 2021). This gives the temporary employees some unpredictable obligations in the temporary employee-employer relationship. For study 2: employers in the exchange relationship, the employer in the direct temporary employment was deemed because the employer in triangular employment creates two employers in the employment relationship (Menatta et al. 2022), and there is legal ambiguity on who is responsible for the temporary employee in an exchange relationship (Hopkins 2015). In this study, the employers are businesses that employ one or more persons, play supervisory roles to their subordinates, and register as small, micro, medium and large-scale indigenous or multinational establishments. In other words, the manager or supervisor represents the view of the employer (Kim et al. 2018), and interacting on behalf of the employer in a direct temporary employment relationship has a significant relationship with how temporary employees perceive the employer or the organisation.

Because of the geographical spread of the study population and unavailable data on the list of employers and employees in direct temporary exchange relationships, the study adopted a cluster sampling technique for a representative sample by focusing on a cluster of participants in city (Lagos State, Nigeria) and industries (service) with the highest concentration of temporary employment in Nigeria (Ikeije 2020; Jawando & Adenugba 2015). Thereafter, a simple random technique was used to draw 354 participants and 105 direct temporary employees and employers, respectively, from the service industry in Lagos State, Nigeria. This sampling method was deemed because it helps with the proximity of the study sample and gives individual participants an equal chance of selection (Grønmo 2019)

For study 1: temporary employees, the participants’ mean age is 26 years (range = 18–50 years, standard deviation [s.d.] = 3.29). The sample consists mainly of males (53.1%, n = 188) working for 40 h per week or more full-time (51.7%, n = 183), in an internship type of temporary employment relationship (52.5%, n = 186) with an average of 2 years and 5 months in the temporary employment relationship and the service organisations. This indicates that the majority of the temporary employees are young and single males working as interns with an honours degree as their highest level of education in temporary fixed-term employment for less than 3 years in a direct temporary employment relationship within the service organisations.

For study 2: employer, the participants’ description shows that the average age of the employer in a temporary employment relationship is 42 years (range = 21–62 years, s.d. = 1.77). The sample consists mainly of males (64.76%, n = 68), who are married (81.90%, n = 86) with an honours degree (52.38%, n = 55) level of education, working in temporary fixed-term employment for 40 h per week or more full-time (76.19%, n = 80) and supervising daily or on-call-type of temporary employment (36.19%, n = 38). These employers have supervised temporary employment relationships at an average duration of 7 years and 5 months, and spent an overall average of 13 years and 6 months in their service organisations. This indicates that the majority of the employers are older adults and married males with an honour’s degree, supervising daily or on-call-type of temporary employment for less than 8 and 14 years in direct temporary employment relationship and in the service organisations, respectively.

Data were collected through a self-administered online survey via QuestionPro (2022) after consent was granted by participating organisations and individuals. The participants for study 1 included temporary employees, and for study 2 employers were randomly selected without matching the views of temporary employees with their immediate supervisor or employer. The online survey was completed at an average of 15 min for temporary employees and 14 min for supervisors representing the employers. The participants completed the survey anonymously after they had been informed about its voluntary nature, research purpose only and that incentives are not given for participation.

Role clarity: The 12-item scale for the role clarity questionnaire (RCQ) was a single-dimensional scale and newly developed after adapting the role clarity model of Linde et al. (2008) vis-à-vis role conflict and ambiguity questionnaire (MRCAQ) developed by Rizzo, House and Lirtzman (1970). Examples of items include: ‘There is available information on expectations between my supervisor and me in this employment relationship’ (temporary employee perspective) and ‘There is available information on expectations between me and my employee in this employment relationship’ (employer perspective). Therefore, the two study scales were rated on seven-point Likert scales (1-never, 2-almost never, 3-rarely, 4-sometimes, 5-often, 6-very often, and 7-always). The Cronbach’s alpha for this current study is 0.96 and 0.93 for the temporary employee and employer scales, respectively.

Psychological contract: A 25-item scale for the psychological contract questionnaire (PCQ) was a single-dimensional scale and newly developed citing five content dimensions of employee obligations and six content dimensions of employer obligations after adapting the model developed by Linde (2015) citing the work of Robinson et al. (1994) and De Vos, Buyens and Schalk (2002) on employee obligations. This measurement scale is similar to the scale adopted by Farnese et al. (2018) with a satisfactory reliability test where employer obligations’ Cronbach’s alpha scored 0.90 and employee obligations’ Cronbach’s alpha scored 0.91. Examples of items that were included in the online questionnaire are ‘me and my supervisor in this employment relationship met performance expectations on the job’ (temporary employee perspective) and ‘me and my employee in this employment relationship met performance expectations on the job’ (employer perspective). Therefore, the two study groups of participants were requested to indicate their perception of the PC rated on a seven-point Likert scale (1-never, 2-almost never, 3-rarely, 4-sometimes, 5-often, 6-very often and 7-always). The Cronbach’s alpha for this current study is 0.95 and 0.92 for the temporary employee and employer scales, respectively.

The measuring scales for role clarity and psychological contract in both studies (study 1 temporary employees; study 2 employers) were formulated based on existing scales. As a result, exploratory factor analysis (EFA) via principal component analysis (PCA) with the varimax rotation option was used to explore and extract factor loadings for each item in temporary employees’ and employers’ role clarity and psychological contract scales. This EFA with PCA was chosen because it helps to determine the dimensionality of the given set of variables and the contribution of the items to their construct domains on a single data set (Chumney 2012).

To assess the internal consistency of the scales, inter-construct correlations coefficient analysis for distinctiveness of the constructs and Cronbach’s alpha were used to examine the validity and reliability of the scales, where the Cronbach’s alpha coefficient (α) (≥ 0.70) is within the acceptable baseline thresholds (Hair et al. 2019), while Kaiser-Meyer-Olkin (KMO) and Bartlett’s Test values were used to establish the suitability of the data for EFA.

The descriptive statistics were analysed for the frequency distribution of the respondents, the mean and the standard deviation of the scales. The hypotheses were tested using the standardised regression weight of the structural equation modelling (SEM). This SEM was adopted in order to establish the relations between role clarity and PC in the direct temporary employment relationship from employers’ and employees’ perspectives (Pyrczak & Oh 2018). The results of the analysis are presented below.

Ethics clearance was obtained from the Faculty of Economic and Management Sciences’ Ethics Committee (EMS-REC), North-West University, South Africa, with ethics number NWU-01843-23-A4. This ethics clearance provides the opportunity to seek the consent of the participants through informed consent before completing the online questionnaire, which gives the participants the opportunity to withdraw voluntarily from the online survey. Also, the participants were assured of their confidentiality as the data collected would be used for the purpose of the study only.

The descriptive statistics considered the mean, standard deviation and correlation coefficient of the constructs. The mean score for each variable in Table 1 is the mid-point of 4, which shows that, on average, temporary employees and employers have a high perception of role clarity and PC in the exchange relationships.

As shown in Table 2 and Table 3, a total of 37 items were explored using a PCA for the two studies. These items represent two constructs (i.e. the PC and role clarity), and there are 12 items for role clarity and 25 items for PC for study 1 (temporary employees) and study 2 (employers). Since the measuring scales are single-dimensional constructs, the extracted factor loadings were based on a single-factor solution. These items for role clarity and PC were extracted at a minimum suppressed value (of 0.30) (Alhija 2010), and none of the items were excluded for temporary employees, except for two items (work extra hours and perform non-required tasks) that were excluded for the employer PC scale only because the factor loading of items not retained was less than 0.30.

That is, all 37 items (> 0.37) were retained for temporary employees, and 35 items (> 0.35) were retained for employers. The communalities (h²) and factor loadings show that the items in the constructs are correlated and have a significant contribution to the temporary employee and employer role clarity and PC constructs. The data were assessed for their suitability for EFA using the KMO measure and Bartlett’s Test of Sphericity for each scale across both samples. For the employer sample, the role clarity scale exhibited a KMO value of 0.867, and Bartlett’s test was statistically significant (χ² [66] = 901.867, p < 0.001). The PC scale for employers (23 retained items) showed a KMO of 0.833 with a significant Bartlett’s test (χ² [253] = 1319.822, p < 0.001). For the temporary employee sample, the role clarity scale had a KMO of 0.950 and a significant Bartlett’s Test (χ² [66] = 4233.203, p < 0.001). Similarly, the PC scale (25 items) for temporary employees had a KMO of 0.948 and a significant Bartlett’s test (χ² [300] = 5844.251, p < 0.001). As all KMO values were well above the recommended 0.60 threshold and all Bartlett’s tests were significant, the data were determined to be highly suitable for factor analysis.

Having established that the data are reliable and suitable for EFA as indicated by the Cronbach’s alpha, and KMO and Bartlett’s test values respectively, the structural model was tested using the maximum likelihood with the aid of SPSS AMOS 30 (Collier 2020) to test the significant relations between role clarity and the PC, as indicated in the path diagram and Table 4. For this study, the structural model relies on EFA because they are newly developed scales for a single data set for each of the constructs and EFA and confirmatory factor analysis (CFA) are inappropriate to be conducted on the same data set (Ocy, Sarifah & Riyadi 2025). Therefore, this study did not collect independent samples that would be required for a satisfactory CFA on the structural model (Henson & Roberts 2006). Further, the:

Given the results of Table 4, the hypothesised model implies that role clarity has a very significant relationship (β = 0.85, β = 0.61 and CR values of 10.636 and 4.101; p < 0.001) on PC for temporary employees and employers in the exchange relationship.

This study theoretically contributes to knowledge by: (1) advancing the understanding of the nexus between role clarity and PC in temporary exchange relationship; (2) empirically established that positive relationship exist between role clarity and PC from a separate perspective of the temporary employee and the employer in an exchange relationship; (3) establishing that what mostly defines role clarity and PC differ in perspectives of employee and employer in the temporary exchange relationship; and (4) providing a new conceptual and structural outlook to establish that temporary employee experience a strong link between role clarity and PC than the employer in the temporary exchange relationship.

The limited and lack of adequate protection in temporary contracts empowers the employer and the employee with the opportunity for ‘employment misbehaviours’, such as a temporary employee suddenly resigning and an employer terminating employment without prior notice, and the assignment or performance of tasks and responsibilities without clarity in the employment relationship.

Despite this, this study has practically made it evident that temporary employees and employers mostly want certainty in responsibilities and understanding in the guiding policies and procedures in role clarity, which to a large extent determines the fulfilment of obligations between the employer and employee in the temporary employment relationships.

In sum, temporary employees and employers would not want to risk the consequences of role ambiguity and unmet expectations in the exchange relationship if there is no availability, understanding and certainty in role performance for the fulfilment of PC despite the limited nature of temporary exchange relationships.

Despite the perceived gap in the literature to examine the relationship between role clarity and PC from a separate perspective of the temporary employees and employers, this study has identified some recommendations for future research based on the limitations of this study. These are highlighted below: