According to, for example, McConnell (2011) the self is a collection of multiple, context-dependent selves resulting in multiple identities. We furthermore categorize and define ourselves in terms of individual attributes, involving a personal self/identity (Hogg and Terry, 2000; Klein, 2014; Knez, 2016), and social attributes, involving a collective self/identity (Jackson et al., 2006; Knez, 2016). Given this, we can say that work identity involves two levels of abstraction as two different knowledge structures (Kihlstrom et al., 2003), namely, personal and collective work identity (Riketta, 2005; Pate et al., 2009; Miscenko and Day, 2016). In line with the self-categorization theory (e.g., Turner et al., 1987) personal work identity has the personal career or occupation as its focus of identification while collective work identity has the organization as a whole as its focus of identification (van Dick and Wagner, 2002; Riketta and Van Dick, 2005).

Thus, the foci of identification refer to variations of the identity concept “that are defined at different levels of abstraction in applied organizational context” (Millward and Haslam, 2013, p. 51). Personal and collective work identities are relatively independent of each other, meaning that having a strong identity at one level does not rule out a similarly strong identity on another level (Brewer and Gardner, 1996; Pate et al., 2009). It is furthermore suggested that these constructs might differ along the dimension of inclusiveness (Sluss and Ashforth, 2007; Miscenko and Day, 2016), resulting in different self-related definitions and interpretations (Sedikides et al., 2011; Knez, 2016). In particular, personal work identity involves classifications of “My profession/work” and, for example, “I/Me a teacher of psychology.” It articulates the psychological need to distinguish oneself from others (Brewer and Gardner, 1996) “in order to preserve the personal self, the personal story and its memories” (Knez, 2016, p. 3).Therefore, it is associated with individual self-related cognitions, interests, values, attitudes, motivations and behaviors (Ybarra and Trafimow, 1998; Johnson et al., 2006). By consequence, the stronger the personal work identity the stronger personal goals, preferences and needs will be (Brewer and Gardner, 1996; Ybarra and Trafimow, 1998; Ellemers et al., 2004).

In line with our psychological prerequisite to belong to a social group (Brewer and Gardner, 1996), collective work identity involves the We-descriptions “in order to be part of the collective self, the collective story and its memories” (Knez, 2016, p. 3). Accordingly, this phenomenon involves cognitions, interests, values, norms, motivations and behaviors related to the identity of a group or an organization (Ybarra and Trafimow, 1998; Johnson et al., 2006). According to social identity theory (Tajfel and Turner, 1979, 1986; see Hogg, 2012 for a review), the knowledge of belonging to a certain group includes furthermore a depersonalization of the individual self and an emotional attachment to the group/organization. Hence, the more “collectivized” an individual is the more s/he will demonstrate loyalty, acceptance, adherence to norms, values, beliefs and decisions communicated by the group/organization (Hogg and Terry, 2000; Ellemers et al., 2004; Johnson and Jackson, 2009). This suggests that normative and motivational mechanisms of the We will be generated by the depersonalized individual (Brewer and Gardner, 1996; Ybarra and Trafimow, 1998; Ellemers et al., 2004).

Along the lines of Knez (2014) and Knez (2016) proposed a conceptual model of the work-related self, a personal work identity including emotion and cognition categorizations of the work, based on the autobiographical memory research (e.g., Kihlstrom and Klein, 1994; Wilson and Ross, 2003; Conway and Holmes, 2004; Conway et al., 2004; Klein et al., 2004; Knez, 2017; Knez and Nordhall, 2017; Knez et al., 2017). The emotion component encompasses the process of work-related attachment/belongingness/closeness, and the cognition component involves the processes of work-related coherence, correspondence, mental time, reflection and agency. This theoretical frame of reference furthermore implies that that the work identity (encompassing above mentioned cognitive and emotional processes) is a higher order cognitive construct (Law et al., 1998; Stajkovic, 2006), a knowledge structure resulting in a personal, autobiographical work-related experience.

Knez (2016) predicted that the emotional component will precede the cognitive one in establishing personal work identity (see Knez and Eliasson, 2017 for similar argument). Finally, Knez (2016) model is general in its formulations, meaning that the psychological mechanisms accounting for the person-work bonding can be applied in both personal and collective accounts. However, in contrast to personal identity, organizational identity is more of a cognitive structure (Harquail and King, 2003) conceptualized as “a product of the dialectic relationship between collective, shared cognition on one hand and socially structured individual cognitions on the other” (Corley et al., 2006, p. 88). This suggests that the cognitive component in organizational identity, in terms of incorporation, identification and assimilation, might precede the emotional component, in terms of affective commitment, esteem and pride (Mael and Ashforth, 1992; van Knippenberg and Sleebos, 2006).

In sum, the concept of work identity used in this study involves two theoretical perspectives, levels of abstraction (in general terms foci of identification, e.g., Reichers, 1985; van Knippenberg and van Schie, 2000; Millward and Haslam, 2013): (1) personal work identity originated from a cognitive, autobiographical memory perspective (e.g., Knez, 2016); and (2) collective work identity emanated from a social identity perspective (e.g., Ashforth and Mael, 1989). As a result, we broaden the multiple focus of identification concept as a part of everyday organizational life to include personal, autobiographical work-related experiences (personal work identity). By that we suggest that personal career/occupation can be treated at both the autobiographical memory level (individual) and the organizational/workgroup level (social). This means that career/occupation at the former level “can be conceptualized as one of the life goals that we strive for and find meaning in Gini (1998), analogous with Sisyphus rolling the boulder up the hill (Camus, 1942)” (Knez, 2016, p. 2).

Research on work motivation has previously shown associations between work motivation and well-being (Baard et al., 2004; Steers et al., 2004; Gagné and Deci, 2005), turnover intentions (Houkes et al., 2003), organizational commitment (Tremblay et al., 2009), and work performance and productivity (Steers et al., 2004; Rich et al., 2010), as well as to organizational and interpersonal work environment (Latham and Pinder, 2005) such as leadership (Castanheira et al., 2016), social support (Castanheira, 2016), and satisfaction (Gillet et al., 2016). Also, individual differences in perceived control, self-esteem, ego-development, defensive functioning and basic need satisfaction, have been shown to associate with work motivation (Gagné and Deci, 2005; Latham and Pinder, 2005).

One of the major theories in motivation, Self-Determination Theory (Deci and Ryan, 2000, 2002; Ryan and Deci, 2000a, 2002), has been broadly applied in the occupational work-context. According to this account (Gagné and Deci, 2005) occupational work per se has the potential to promote different psychological needs and thereby influence personal self-determined, i.e., intrinsic, motivation (Ryan and Deci, 2000b; Deci and Ryan, 2010; van den Broeck et al., 2016). Furthermore, Self-Determination Theory highlights individual rather than group-based needs and motivation and by that has the individual as its focus (Gagné and Deci, 2005; Riketta and Van Dick, 2005; Ryan and Deci, 2012; Bjerregaard et al., 2015).

Three basic psychological needs are assumed to form the foundation on which self-determined work motivation relies (Deci and Ryan, 2000; van den Broeck et al., 2016): (1) Competence, involving effectiveness, confidence and capacities in one's daily work and in the interplay with colleagues, managers, customers and clients (Deci and Ryan, 2000, 2008; Elliot et al., 2002; Gagné and Deci, 2005); (2) Relatedness, refers to feelings connected to others (colleagues, managers, customers, clients etc.) and being cared for by, and caring for, others (Deci and Ryan, 2000, 2008; Gagné and Deci, 2005); and (3) Autonomy, involves experiences of own behaviors and actions as expressions of the self (Deci and Ryan, 2000, 2008; Gagné and Deci, 2005).

Finally, previous findings indicate that the more one identifies with occupational work the stronger self-determined work motivation will be, and the stronger one perceives the views of one's organization to be the less one's self-determined motivation is presumed to be (Deci and Ryan, 2000; van den Broeck et al., 2016). Given all this, in this study the concept of work motivation was operationalized as the work self-determined motivation.

Organizational justice has been shown to associate with both individual and organizational outcomes (Greenberg, 2011; Colquitt et al., 2013), such as work performance (Wang et al., 2015), job satisfaction (Greenberg, 2011), organizational commitment (López-Cabarcos et al., 2015), counterproductive behavior Cremer (2006), turnover intention (Aryee et al., 2002), organizational citizenship behavior (Blader and Tyler, 2009), health related factors like sick leave, stress related problems, cardiovascular problems, burnout and emotional exhaustion (Greenberg, 2010; Ndjaboué et al., 2012; Robbins et al., 2012; Piccoli and De Witte, 2015), and anxiety and depression (Spell and Arnold, 2007).

Several predictors of organizational justice have also been indicated, such as trust in the supervisor or organization (Aryee et al., 2002), perceived equity and equality within the organization (Colquitt et al., 2005), needs (see Greenberg and Cohen, 1982), job security, complexity and status (Ambrose and Arnaud, 2005) moral and ethical standards (Folger et al., 2005), perceived organizational support (Bies, 2005), and expectations (Colquitt et al., 2005).

This construct has furthermore been shown to comprise four independent dimensions (Colquitt, 2001; Ambrose and Arnaud, 2005; Colquitt and Shaw, 2005; Greenberg and Colquitt, 2005; Ambrose and Schminke, 2009; Nicklin et al., 2014):

Previous research strongly suggests including all four justice dimensions in studies related to organizational context issues (Bies, 2005). However, since this phenomenon emanates from the organizational work-environment and thus has the organization, i.e., the collective as its focus (Riketta and Van Dick, 2005), it is reasonable to assume that organizational related issues like organization-related identity, loyalty, shared values and preferences, may be more critical for the organizational justice (Cremer, 2006; Blader and Tyler, 2009). Finally, in this study the concept of organizational justice was operationalized as organizational pay justice, because it originates from an organizational framework (Greenberg and Colquitt, 2005; Andersson-Stråberg et al., 2007) and by that might link with collective work identity (see Present Study below for this line of reasoning).

In line with self-categorization theory (e.g., Turner et al., 1987) and the general notion of foci of identification and its associated effects, the relationship between work identity and other work-related constructs might be stronger when foci of categorization between the constructs match; thus, when both the type of identity (predictor) and the outcome (criterion variable) are conceptually related (van Dick and Wagner, 2002; Riketta and Van Dick, 2005; Olkkonen and Lipponen, 2006).

Within the frameworks of social identity theory (Tajfel and Turner, 1979, 1986; see Hogg, 2012 for a review), the two predominating foci of identification (and outcomes) have been work-group (relational level) and organization (collective level), both involving a social focus of identification. This means that the personal focus of identification has to a large extent been neglected in previous research (van Dick and Wagner, 2002; Riketta, 2005; Riketta and Van Dick, 2005). Given this, the aim of the present study was to explicitly test the general notion of foci of identification and its associated effects within and beyond the social focus of identification. This was done by investigating both the personal and the collective focus of identification in relation to work-related motivation and justice.

By consequence and given that personal and collective work identity might tap psychological processes at different levels of abstraction (Haslam et al., 2000; Sluss and Ashforth, 2007), we hypothesized that these constructs will associate differently with work-related motivation and justice.

It has, for example, been reported that collective work identity is not associated with individual internal motivation (van Knippenberg and van Schie, 2000), and that a positive association between collective work identity and organizational-related motivation and behaviors is stronger in a collectivistic vs. an individualistic organizational culture (Lee et al., 2015). Also, the more one identifies with one's occupational work tasks (personal/individual identification) the stronger one's basic need satisfaction is, and the stronger one identifies with one's organization (collective identification), the weaker one's self-determined motivation is (van den Broeck et al., 2016).

The two different structures of work identity have been, furthermore, related to different motivational (van Knippenberg and van Schie, 2000; Ellemers et al., 2004; Johnson et al., 2010; He et al., 2014) and attitudinal (Ybarra and Trafimow, 1998) outcomes. Previous studies have, however, largely addressed the motivational issues on behalf of the organizational collective as opposed to the personal motivation to perform well. Also, the focus of identification has to a large extent been framed in terms of social identity; thus, relational and/or collective work identity (van Knippenberg, 2000; van Dick and Wagner, 2002; Bjerregaard et al., 2015).

Most previous findings on the relationships between work identity and organizational justice have, in line with the group-engagement model (Tyler and Blader, 2003; Blader and Tyler, 2009), focused on justice as a decisive predictor of different levels of work identity. This was done to clarify the role of identity-based information incorporated in organizational justice (Lipponen et al., 2011), as well as the mediating role of work identity in relation to the effects of organizational justice on different outcomes (Olkkonen and Lipponen, 2006; Cho and Treadway, 2011; He et al., 2014). Finally, interaction effects of social identity and organizational justice dimensions on different work-related outcomes have additionally been investigated (Cremer, 2006; Lipponen et al., 2011; Smith, 2012).

Regarding the relationship between work identity and organizational justice, it is suggested that the self-constitutes one of the predicting foundations for self-reported justice (Skitka, 2003). Several findings have indicated that this association primarily concerns collective work identity (Olkkonen and Lipponen, 2006; Blader and Tyler, 2009; Cho and Treadway, 2011; Ehrhardt et al., 2012; He et al., 2014).

Despite the previous research on work-related motivation and justice no previous research has, as far as we know, investigated the role of personal and collective work identity in predicting these constructs. Accordingly, and in line with the general notion of foci of identification and its associated effects (van Dick and Wagner, 2002; Riketta and Van Dick, 2005; Olkkonen and Lipponen, 2006) the aim of the present study was to investigate the relationships between personal and collective work identity, as predictors, and work-related motivation and justice as criterion variables. This involves both matching variables and two levels of explanation; that is, a link between: (1) Personal work identity and work-related Self-determined motivation (personal level of explanation); and (2) Collective work identity and Organizational pay justice (collective level of explanation).

Given the above, we predicted:

Digitized questionnaires were distributed by e-mail to 2905 members of the Swedish trade union “The National Union of Teachers” (in Swedish “Lärarnas Riksförbund”) working in the south and middle part of Sweden, during April and May 2016. A total of 768 questionnaires were returned, with a response rate of 26%. Ninety-nine percent of the participants had an educational function, that is, they worked as teachers or similar. Also, the participants worked within the municipal sector (92.1%), had university studies as their highest level of education (68%), were in permanent employment (95.2%), had full-time jobs (80.5%) and were female (75.5%). Mean age was 46.3 (SD = 10.07, range 24–67) and mean employment time within the organization was 14 years (SD = 10.2).

Chairpersons of 11 municipal associations of the Swedish trade union The National Union of Teachers were contacted and informed of the purpose of the present study. They were asked to invite their union members to participate in the present study, of which they approved. Due to Swedish juridical restrictions that do not permit a chairperson of a union to distribute individual e-mail addresses of the members, a web-link to the questionnaire was distributed to the members by the chairpersons. The questionnaires were accompanied by a covering letter that described the purpose of the study and informed the participants that completion of the questionnaire was taken as an indication of their consent to participate in the present study, and that this was voluntary and that confidentiality and anonymity was assured. After completion of the questionnaire the participants were asked to fill in their name and address if they wanted to receive a cinema ticket as compensation for their participation. They were informed that nobody but the researchers of the present study would have access to their names and addresses.

Finally, an ethical application was reviewed and approved by the Swedish regional ethical committee of Uppsala University (Dnr 2015/423).

In line with the four hypotheses, four types of multiple hierarchical regression analyses were performed in order to investigate the role of personal and collective work identity in predicting motivation and justice at work:

In all four analyses, we controlled additionally for the effects of: monthly income; school sector (public vs. private); years of employment; and permanent employment (no vs. yes). These variables have previously been addressed as potential confounders, related to work self-determination motivation (Amabile et al., 1994; Vansteenkiste et al., 2007; Howard et al., 2016; Ryan and Deci, 2017) and organizational pay justice (Ambrose and Schminke, 2003; Andersson-Stråberg et al., 2007; Fortin, 2008). In order to specify a fixed order of entry to control for the effects of potential confounders, in the four multiple hierarchical regression analyses, the covariates were entered in step one and predictors in step two.

Together, personal- and collective work identity predicted work self-determined motivation, with an additional explained variance of 24% after controlling for the four potentially confounding variables. As predicted (Hypothesis 1), personal compared to collective work-related identity was shown to positively associate with work self-determined motivation (see Table 2; see also Appendix 1 for the first step regression statistics).

Together, emotion- and cognition components of personal work identity predicted work self-determined motivation, with an additional explained variance of 35% after controlling for the four potentially confounding variables. In agreement with Hypothesis 2, emotion- compared to cognition component of personal work identity was shown to positively associate with work self-determined motivation (see Table 3; see also Appendix 1 for the first step regression statistics).

Together, personal- and collective work identity predicted all organizational pay justice dimensions, with an additional explained variance of 11.1% in procedural pay justice, 1.3% in distributive pay justice, 4.2% in interpersonal pay justice and 4.4% in informative pay justice, respectively, after controlling for the four potentially confounding variables. Consistent with Hypothesis 3 collective- compared to personal work identity was shown to positively associate with all organizational pay justice dimensions (see Table 4; see also Appendix 2 for the first step regression statistics).

Together, emotion and cognition components of collective work identity predicted the organizational pay justice dimensions with an additional explained variance of 11% in procedural pay justice, 1.4% in distributive pay justice, 4.3% in interpersonal and 4.5% in informative pay justice, respectively, after controlling for the four potentially confounding variables. In line with Hypothesis 4, cognition vs. emotion component of collective work identity showed a stronger positive relationship with organizational pay justice, with exception for the distributive justice dimension. In other words, the cognition component of collective work identity accounted for a stronger association with three dimensions of organizational pay justice than did the emotion component of collective work identity (see Table 5; see also Appendix 2 for the first step regression statistics).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.