Antecedents of organizational identification: implications for employee performance Open Access

Organizations make many efforts to meet their existing necessities and enhance employees’ performance to ensure productivity and conductivity at their workplace. However, organizations must also prepare for their upcoming necessities to confront impending challenges (Argote, 1999). The two main challenges of the organization are employees’ performance and motivation (Appelbaum, 2000). Employees are considered the tangible business assets that encourage the everyday activities and operations of an organization. Organizational viability and effectiveness rely upon how powerful and productive the workers are (Sparrow & Cooper, 2014). Employers' ability to appreciate workers' fulfillment of timetables and everyday obligations will significantly affect employees’ efficiency and performance. As employees are sustainable organizational resources, they become key drivers for change, particularly in a quickly evolving environment.

The performance of the employees is defined as a conduct that is different from the outcome of their work (Campbell, McHenry, & Wise, 1990). The outcome does emerge from the employee’s performance; however, more elements determine results. When explaining the attitudes and behaviors at the individual and organizational levels, organizational identification (OID) holds a unique value because of the involvement of various organizational identities (Piccoli et al., 2017). It roots an individual toward the achievement of goals and values in the organization; however, in return, an organizationally identified employee is likely to be interested in making decisions to benefit the entire organization rather than focusing on flourishing his self-interest (Lee, Park, & Koo, 2015). Deloitte’s (2018) survey reported employee performance (EP) statistics in different industries, namely, professional services (11%), financial services (8%), consumer business (13%), technology, media and telecommunications (12%), manufacturing sector (7%), public sector (12%) and energy and resources (12%), from different regions: Europe, USA, Asia, Africa, Oceania, Middle East and Nordic Countries.

The achievement of a high-level performance through profitability and productivity has been an organizational objective of great primacy. For this reason, a highly satisfied workforce is an absolute need. For several decades, employment issues have become an integral part of human relations and a significant concern for organizations operating at the national and international levels. The problem occurs when employees are not performing as predicted by the organizational planning (Bradler, Dur, Neckermann, & Non, 2016). Estes and Wang (2008) addressed the issues that may improve or bring down the employees’ performance and independently or jointly influence the performance of employees, such as customer satisfaction, loyalty, workplace incivility (WI), work environment and engagement.

On the other hand, in a recent study, Chiang and Hsieh (2012) predicted EP through support from the organization, supervisor and job satisfaction (JS). Going back to the late 1970s, Steers (1975) stated that there is no convincing argument that performance must essentially cause satisfaction, predominantly if performance goes unrewarded. More concretely, Ostroff (1992) discovered that organizations with satisfied employees were more compelling than organizations with disappointed ones.

EP is influenced by financial benefits, training plans, non-monetary benefits, supervisory support, organizational support and capability-building courses. Resultantly, EP upgrades organization benefits (Gull, Akbar, & Jan, 2012). The specialists and academic experts affirm that performance and results rely on organizations’ methods to deal with their workers (Delaney, 1996). Organizational outcomes can be enhanced by adopting special measures comprising employees’ job involvement, empowerment, JS, skill-based training, advancement programs and preventing harmful activities such as bullying and incivility at the workplace (Pfeffer, 1998).

OID is defined as “a feeling of psychological inclusion or belongingness to a particular firm or institution” (Ashforth, Harrison, & Corley, 2008, p. 229). It occurs when an individual uses organizational identity (same attributes, virtues and flaws) to define himself and considers the achievements and failures of the organization as their own (Dutton, Dukerich, & Harquail, 1994). A few past researchers have explored the mediating effect of OID on EP, but very little is known about this relationship for private-sector employees. OID has also been used to mediate perceived organizational performance and job performance (Carmeli, Gilat, & Waldman, 2007). Walumbwa et al. (2011) discovered the link between ethical leadership and employee while focusing on the mediating effect of OID. The relationship between favorable work experiences and affective organizational commitment was also mediated through OID (Stinglhamber et al., 2015).

Past studies used job mobility (Wu, Liu, Kwan, & Lee, 2016), organizational commitment, JS (Sy, Tram, & O'Hara, 2006), trust and rewards (Colquitt, Scott, & LePine, 2007; Pushpakumari, 2008) as predictors of EP. However, little is known, and there is no research in which EP is measured by WI, perceived supervisor support (PSS) and JS in a single model. There is partial research on the antecedents that affect EP in organizations, specifically, with the mediating impact of OID. Therefore, a quantitative study is required to ascertain the relationship between WI, PSS, JS and OID. Moreover, it is also important to analyze the mediating effect of OID on EP.

Thus, the first objective of this study is to determine the relationship between WI, PSS, JS and OID. Another objective of the study is to ascertain the relationship of OID with EP. It is speculated that the presence of OID will strengthen the performance of employees at the workplace. This study is significant for institutions and practitioners in two ways. Firstly, using a second-order model to inspect the relationship between the three antecedents of OID. Secondly, by linking OID as a mediator between those antecedents and EP, which has been, to some extent, ignored in the existing literature. This study further improves the current body of knowledge of the circumstances when OID can efficiently boost the performance of employees at the workplace.

WI is defined as “a low-intensity deviant behavior with ambiguous intent to harm the target, in violation of workplace norms for mutual respect” (Andersson & Pearson, 1999, p. 457). Target employees of WI tend to restrain tasks and actions beyond their job requirements, which is also supported by the social exchange theory that employees do not get along with other co-workers or heads if they ever offend them by any means (Andersson & Pearson, 1999). Researchers who studied the relationship between WI and OID also observed major changes in employees’ behavior who faced uncivil practices such as stress, depression, anxiety and sometimes, even suicide (Cortina, Magley, Williams, & Langhout, 2001). In addition, inappropriate behavior tends to happen in an uncivil workplace (Lim & Cortina, 2005). According to Fevre, Lewis, Robinson, and Jones (2012), incivility and unreasonable treatment are preferable over bullying, which may lead to a higher level of aggression (Pearson & Porath, 2009). Therefore, employees who experience uncivil practices usually get involved in revenge and disruption (Hodgins, MacCurtain, & Mannix-McNamara, 2014). Pearson and Porath (2009) reported a significant but negative relationship between WI and OID because when employees are victims, they lose their identity and dedication toward their organization. Thus, the following hypothesis has been developed:

PSS is defined as “the degree to which a subordinate feels that they are supported and respected by their supervisor” (Eisenberger, Huntington, Hutchinson, & Sowa, 1986, p. 500). Similarly, Kottke and Sharafinski (1988) further elaborated that “it also refers to the supervisor’s willingness to help the subordinate” (p. 1076). Being a representative of an organization, the supervisor’s responsibility is to direct and evaluate the performance of employees working under his supervision. According to Rhoades and Eisenberger (2002), supervisor support is strengthened when employees are aware that their assessments are often communicated to higher-level management; however, this support further leads to high commitment and JS. In addition, Johnston, Parasuraman, Futrell, and Black (2000) postulated that employees who perceive their supervisors as considerate and approachable perform much better. Researchers also found a positive link of supervisory concern in the past (DeCotiis & Summers, 1987; Zaccaro & Dobbins, 1989).

Moreover, an emotional connection exists between the employee and its organization established by the supervisor–employee relationship since the supervisor considers himself a representative of his subordinates when it comes to the organizational level. Studies (Tepper, 2000; Martinko, Harvey, Brees, & Mackey, 2013) reveal that abusive supervision can result in a negative effect on the employees, whereas leadership which empowers the employees facilitates and promotes employee creativity (Zhang & Zhou, 2014) and causes a high OID (Chan & McAllister, 2014). Van Knippenberg, Van Dick, and Tavares (2007) confirmed that supervisor support positively affects the employee’s identity with its organization. Social exchange theory suggests that employees would reciprocate through a positive attitude, considering it an obligation toward the supervisor and the organization. The authors discovered a positive relation of supervisor support and OID. The behaviors may include working hard, helping co-workers achieve organizational or departmental goals and quality work (Wayne, Shore, & Liden, 1997). Hence, based on the related literature, the following hypothesis is suggested:

JS is defined as “how people feel about their jobs and the different aspects of their jobs” (Spector, 1997, p. 513) and “the extent to which employees gain enjoyment from their efforts in the workplace” (Fogarty, 1994, p. 13). Satisfaction can be positive or negative depending on the individuals or work circumstances (Weiss, 2002). Thus, it can be said that people express themselves in their job performance (Kahn, 1990), and JS can be higher with OID (Newman, Dana, & Hulin, 2010) as the employees will have a psychological bond with their organization. However, a strong relationship between JS and OID has been reported in the literature (Schrodt, 2002; Wiesenfeld et al., 2001). Van Knippenberg and Van Schie (2000) revealed that JS is connected to numerous other organizational results, and OID is one of them. These results are consistent with those studies of Pratt (1998) and Van Dick, Wagner, Stellmacher, and Christ (2004), who found that positive identification with the organization enhances inspiration, work fulfillment, and commitment. Therefore, this study expects to ascertain that employees’ OID is positively affected by the JS of employees in the organization. Nevertheless, focusing on the latest times, it can generally be expressed that executives are looking for those who feel good in their workplace and like to work with individuals who have a positive perspective of the job (Bakotić, 2016). Workers who possess a high JS level tend to enjoy their job and are recognized in the organization through their tasks and excellent performance (Karanika-Murray, Duncan, Pontes, & Griffiths, 2015).

According to Adams’ equity theory (Adams & Freedman, 1976), employees compare their output/input ratio to their colleagues’. If these ratios are not equal, employees having higher ratios feel humiliated due to the assumption of being overpaid. On the contrary, employees having lower ratios face aggressiveness in their behavior due to the assumption of being underpaid. Therefore, they only feel satisfied when the ratios between the two are equal (LePine, Erez, & Johnson, 2002). The social identity theory (SIT) states that employees who feel content with their jobs often position themselves as a part of their organization and feel that its goals and achievements are their own (Ashforth & Mael, 1989). Thus, the employees create a positive bond with the organization.

Employees who sense injustice respond by endeavoring to change that unbalanced state. According to the statement of Salancik and Pfeffer (1978), “employees decide to what extent they are satisfied with their jobs by observing other employees' satisfaction levels” (p. 225). In most cases, employees think they can stay in the organization and maintain their social identity only if they are satisfied with their jobs. SIT refers to “the perception of belonging to a group within a social category” (Chen et al., 2015), and it is used to investigate and understand the growth of OID. In general, therefore, it seems that:

Employees’ performance is affected by their organization’s identification as it creates a bond between them, and it could further strengthen the link between the organization’s actions and individual outcomes (Karanika-Murray et al., 2015). This also matches the earlier observations of Dutton et al. (1994) that individuals who have a strong identity with the organization are anxious about their organization’s prosperity. Additionally, as noted by Dutton et al. (1994), “when people strongly identify with their work organization, their sense of survival is tied to the organization’s survival” (p. 242). In this manner, employees who identify themselves as representatives of the organization are expected to work instinctively to benefit it (Ouchi, 1980).

The argument of the current study is consistent with those mentioned above, which found out that OID has a positive relationship with EP which derives from the SIT (Walumbwa et al., 2011). Furthermore, Van Knippenberg, De Dreu, and Homan (2004) proposed that individuals with high OID tend to be more prominent supporters of their organization. Also, various studies in the past (Ouchi, 1980) indicate a strong and positive relationship between OID and EP. Consequently, employees with higher organizational identity will perform their tasks in a more acceptable way than those with lower organizational identity (Dutton et al., 1994). Thus, the following hypothesis has been established:

OID is defined as the “perception of oneness with or belongingness to the organization” (Ashforth & Mael, 1989, p. 21). Many investigations have indicated that higher identification is favorable for the individual as well as the organization. From an individual point of view, it shows upgraded work performance, high motivation, prosperity and enhanced well-being. However, from an organizational perspective, higher identification leads to low absenteeism with better performance (Van Dick et al., 2004). Nevertheless, Hoy, Smith, and Sweetland (2002) asserted that high identifiers are more motivated to acquire new skills and knowledge as it enables them to contribute to organizational success.

SIT is the theoretical solid basis for understanding OID, which suggests that individuals describe themselves by using groups and enhance their self-worth by using their social status in the organization (Kramer, 1993). Turner (1982) believes that once an individual is identified in a group, he tends to be committed to that group and produce high-performance standards for achieving the goals and objectives associated with the respective group. Research has revealed that WI is a significant predictor of OID (Pearson & Porath, 2009). On the other hand, it has also been reported that a positive relationship exists between JS, PSS and OID (Van Knippenberg et al., 2007). Therefore, our final hypothesis is stated below: (Figure 1)

This study research design is co-relational, also known as survey or non-experimental research design (Tharenou, Saks, & Moore, 2007). Following a probability sampling approach, we chose cluster sampling because the total population is known and groups have been categorized. As the population is 300, the research sample size has been calculated 160 using Krejcie and Morgan’s (1970) technique. However, owing to the issue of normal distribution, it was extended to 200. Thus, over 200 survey questionnaires were distributed among permanent business faculty members of private universities in Karachi with a minimum of one year of experience, out of which 175 valid sample forms were received. These questionnaires were distributed manually, and the participants were informed of the purpose of the study.

This study used a total of five variables measured on a five-point Likert type scale (1 = strongly disagree to 5 = strongly agree). All items were self-reported and responded in a non-contrived field-study setting. After a short pilot study, the five reliable and valid questionnaires were adapted into a single questionnaire to ensure that the language was self-explanatory and the questions were directly related to the study.

It refers to the different sorts of working environment abuse, for example, tormenting, inconsiderate treatment, antisocial behavior and social undermining (Torkelson, Holm, Bäckström, & Schad, 2016). A total of 12 items were adapted from Cortina et al. (2001).

It refers to the supervisor’s support to his employees (Kottke & Sharafinski, 1988). A total of seven items were adapted from DeConinck and Johnson (2009).

It refers to an element of the apparent connection concerning what one needs from one’s activity and what one sees it as offering (Locke, 1969). A total of seven items were adapted from Spector (1994).

It refers to how much a part characterizes themselves by similar traits that they recognize as a characteristic of the association (Dutton et al., 1994). A total of seven items were adapted from De Roeck and Farooq (2018).

It refers to the exercises and undertakings implemented by a worker proficiently and successfully (Saleem & Amin, 2013). A total of six items were adapted from Koopmans et al. (2012).

Face and content validity was conducted to ensure that the items under the construct were perfectly suitable, which meant they measured the construct with utmost adequacy. For this purpose, two subject experts were consulted to validate the tool and the questionnaire (Van der Stede, Young, & Chen, 2005). The final questionnaire is summarized in Table 1 below:

IBM’s SPSS version 22 and SmartPLS version 3.2 were used for data analysis. First, the data were screened to check the missing and out-of-range values by using SPSS. Then, multivariate outliers were eliminated through Mahalanobis distance suggested by Pallant (2011). After the removal of 23 outliers, the usable sample includes 175 responses. Afterward, the measurement model was developed to ascertain its construct reliability, convergent and discriminant validity. Finally, a five-point scale testing research hypotheses using the partial least squares – structural equation modeling (PLS-SEM) technique was created. Table 2 shows the demographic profile of the respondents. In total, 37% were male and 63% were female. The majority (48%) were in the age group of 31–40, 31% were between 21 and 30, 13% were between 41 and 50, and 6% were above 50 years. Moreover, 12% were graduates, 33% were postgraduates, 45% were MPhil scholars and 10% possessed doctoral degrees. The majority of the respondents, i.e. 52%, had 1–5 years of experience, 30% had 6–10 years of experience and 18% had more than 10 years of experience. All the respondents were from four different private universities in Karachi, Pakistan.

The measurement model is used to analyze the relationship between variables and Likert scale items. Consequently, this study is explanatory with a complex model that includes mediation and multiple variables at a time. To estimate such a complex model, the researchers used PLS-SEM using SmartPLS 3 (Ringle, Wende & Becker, 2015). Firstly, the estimations of factor loadings were inspected: the outer loadings were within the range of 0.732 and 0.932 (Table 3). Secondly, composite reliability (CR) was tested to check the degree to which the items of a particular variable describe the latent variable. As a result, all variables demonstrate good construct reliability because each CR value is in excess of 0.70 (Hair, Hult, Ringle, Sarstedt, & Thiele, 2017). Furthermore, convergent validity was also established for each variable because the average variance extracted (AVE) value is greater than the minimum threshold limit of 0.50 (Henseler, Hubona, & Ray, 2016).

As Table 4 shows, the discriminant validity was assessed using Fornell-Larcker criterion (Fornell & Larcker, 1981) and heterotrait-monotrait (HTMT) ratio of correlations (Henseler, Ringle, & Sarstedt, 2015). According to Kumar (2015), the diagonal values of AVE square root should be greater than other values in their respective row and column. It can be seen that all values are less than the suggested value that ranges between 0.18 and 0.72, hence, reflecting good discriminant validity (Figure 2).

Another estimation for testing discriminant validity is the HTMT proposed by Henseler et al. (2015), which states that the threshold values for estimating HTMT criterion should be less than 0.85. Table 5 demonstrates that all values are less than the suggested value ranges between 0.18 and 0.72, hence reflecting good discriminant validity.

After the outer model was clearly established as described in the previous section, the next step included inspecting estimations for the inner model considered in the study. For this purpose, the Smart PLS 2.0 software was used to test the hypothesized model. The inner model is used to analyze the relationship of one variable with the other variable. Hair, Ringle, and Sarstedt (2011) suggested using PLS-SEM for exploratory and complex models. To estimate such a complex model, the researchers used PLS-SEM with SmartPLS 3. For using SmartPLS, the first step was to examine path coefficients for testing the hypotheses proposed earlier. To generate path coefficients, PLS algorithm was used, then bootstrapping with a sample of 5,000, where the critical t-values of 5% significance level are more than 1.96 as suggested by Hair et al. (2011). Table 6 demonstrates that the results support H1–H4.

The model’s predictive relevance should be checked to analyze its quality (Hair, Black, Babin, & Anderson, 2010). The researchers of this study tested the model’s relevancy known as Q², which is the proportion of the total sum of squares captured by the model. The value of Q² should be greater than zero (Hair et al., 2011). Also, R² can be tested, which indicates how much of the dependent variable can be explained by the independent variable. Referring to Table 7, all values meet the minimum criteria which proves the relevancy of the studied model.

When two variables are connected through a third variable, it is said to be a mediating effect (Hair et al., 2013). As this study used Smart PLS to analyze the data, the researchers used a bootstrapping method by keeping a sample of 5,000 as suggested by Hair et al. (2011). For this study, the mediating effect was estimated using Baron and Kenny (1986), which requires a, b, c and c′ values. The value of “a” represents an independent variable’s impact on the mediating variable, “b” represents the impact of a mediating variable on dependent variables, whereas “c” represents the impact of an independent variable on the dependent variable. If all values are significant, then it is said to be a partial mediation among the variables. However, a bootstrapping method was also applied to estimate the result of indirect effect c′ (a × b). The decision of bootstrapping results is based on t-values, which should be greater than 1.96. Table 8 shows that H5a is not supported for having non-mediating effect, whereas H5b and H5c are supported as having mediating effect in this study (Figure 3).

Based on Zhao, Lynch, and Chen (2010), H5a was supported (t-values of a × b = 1.903), showing the presence of direct-only non-mediation of WI on EP (Table 9). Similarly, H5b and H5c were also supported (t-values of a × b = 2.371, 2.145) and displayed the presence of indirect-only mediation of PSS and JS on EP, respectively.

The results of H1 revealed that WI has a negative relationship with OID (β = −0.128, p-value = 0.013); thus, H1 was confirmed. These results are consistent with Baghini, Pourkiani, and Abbasi (2014) findings, which revealed that when faculty members experience uncivil behavior at work, they start losing their identity and trust toward the organization. In addition, workplace colleagues also play an integral part in boosting performance. If faculty members continuously go through ignorance and disrespectful remarks, they gradually start losing their identity in the institution.

The findings of H2 showed that PSS is positively related to OID (β = −0.325, p-value = 0.021); hence, H2 is supported. According to the findings of Shanock and Eisenberger (2006), supervision is an effective tool to promote good results as far as staff performance is concerned. In this sense, faculty members feel motivated and take initiatives in institutional activates that further pave their way in getting identified within the institution. The results also revealed that the majority of the faculty members are satisfied with their supervisors as they get intensive support for accomplishing major priorities, their goals and viewpoints are valued and their supervisors constantly notice them.

As to H3, the results revealed that JS has a positive relationship with OID (β = −0.280, p-value = 0.009); thus, H3 is confirmed. These findings are similar to previous research that ascertains that employees’ OID is positively affected by the JS of employees in the organization (Schrodt, 2002). The faculty members who possess a high JS level tend to enjoy their job and be recognized within the institution through their tasks and excellent performance (Karanika-Murray et al., 2015). Nevertheless, focusing on the latest times, it can generally be expressed that executives and supervisors are looking for those who feel good at their workplace and like to work with individuals who have a positive perspective of the job (Bakotić, 2016). Moreover, if faculty members are satisfied with their job and think they are being fairly paid for their skills and competencies, they feel proud of being represented by the organization.

The findings concerning H4 showed that OID is positively related to EP (β = −0.410, p-value = 0.000), therefore confirming it. Employees who consider themselves as part of their organization and feel strong ties with it tend to perform better than others. Hence, it is essential for organizations to integrate their employees to get the positive performance which will eventually increase the goodwill and sustainability of the organization. These results are similar to previous research (Schneider, Hanges, Smith, & Salvaggio, 2003). Remarkably, our study is among the few that have examined the relationship between OID and EP. The results endorse the concept that employees recognized in a particular organization for their efforts are likely to exhibit higher quality performance.

Lastly, the outcomes of mediation analysis revealed direct-only non-mediation between WI and EP and indirect-only mediation of PSS and JS with EP, respectively. According to H5a, OID will mediate the relationship between WI and EP. However, the results showed an insignificant indirect effect between the two, and only a statistically significant direct effect exists between these variables. It implies that due to the nature of WI is ambiguous and unintentional motivation to harm others, it will affect the performance of permanent faculty members of private institutions in Pakistan. Notably, it contrasts with the hypothesis though OID mediated the relationship of PSS with EP, and JS with EP, this study found that it has an insignificant indirect impact on EP. These findings may be deemed one of the study's limitations: the sample was drawn from only four private universities in Karachi. It is believed that a broader sample from private or public sector universities may find a statistically significant indirect effect of OID between the relationship of WI and EP.

The purpose of this study was to determine the relationship between WI, PSS, JS and EP to ascertain whether OID predicts EP. It also aimed at confirming the mediating role of OID between independent and dependent variables. The study took place in a non-contrived environment. It demonstrated the significance of PSS on OID and revealed that EP plays an essential role in shaping the organization’s vision.

This study can be further carried out by focusing on specific target respondents to get a more precise picture of the scenario. Lastly, it is suggested to explore WI due to the insignificant effect revealed in this study. The findings suggest that the manager or the immediate supervisor should recognize their employees and create a sense of identity that boosts their morale. If supervisors only criticize, then the employees will start hiding their mistakes. Therefore, it is necessary to maintain a balance between their strengths and weaknesses. Satisfied employees are more concerned about the targets they have to achieve and suggest new ideas with more positivity.

It is highly recommended that organizations work on the areas that lead to the enhancement of EP. Also, HR should create a healthy culture that promotes taking initiatives, open-door policies and discourage power distance. Supervisors should also acknowledge and appreciate the efforts of their subordinates and enforce recreational policies and employee engagement activities to prevent uncivil practices at the workplace. Finally, the organization should provide incentives to boost EP and introduce different activities to minimize WI and promote teamwork.