The role of internal locus of control and social competence in knowledge worker innovativeness Open Access

Innovativeness, which can be defined as an individual’s intentional behaviour to generate, promote and actualise ideas to produce novel processes and products (e.g. De Jong and Den Hartog, 2008), stands as a cornerstone of organisational performance (e.g. Oldham and Cummings, 1996; Anderson et al., 2014; Shin et al., 2017; Hock-Doepgen et al., 2024). While innovativeness plays a critical role in enhancing organisational outcomes (e.g. Zhou and Shalley, 2003; Tsai, 2018), innovativeness also serves as a core element of autonomous knowledge work. This form of work is distinguished by challenges in competence, independence in problem-solving and the adept organisation of tasks (e.g. Martela et al., 2021; Zhao et al., 2023; Frèour et al., 2024). Autonomous work both stimulates and requires innovative work behaviour, such as creative thinking, taking initiative, promoting new ideas and translating ideas into actionable outcomes (e.g. Hisrich, 1990). Given the multifaceted nature of knowledge work (Jarrahi et al., 2023; Baer, 2003; Kelloway and Barling, 2000), the increasing number of knowledge workers (KWs) (e.g. Berg and Gmyrek, 2023) and their pivotal role in driving organisational performance (Seth and Lee, 2017; Zhang-Zhang et al., 2022) by bearing responsibility for knowledge acquisition, creation, processing and organisation (Edgar et al., 2015; Humphrey et al., 2007; Jacobs, 2017), it is crucial to examine the factors that enhance their innovativeness.

A significant body of prior research has examined the individual (Amo, 2006; Amar, 2004; Newman et al., 2018; Norena-Chavez and Guevara, 2020; Scott and Bruce, 1994) and contextual factors (Chen et al., 2016; Haider et al., 2023; Afsar et al., 2017; Shafique et al., 2020; Iqbal et al., 2022; Khari and Bali, 2024; Ononye, 2023; Muchiri et al., 2020) that influence employee innovativeness, thereby leaving innovativeness less studied within the particular context of knowledge work. The studies that have addressed innovativeness (e.g. Dul et al., 2011; Nisula et al., 2022) focused on contextual factors. For example, a study on designers’ innovativeness found that creativity skills fostered idea generation, whereas operational skills enhanced the implementation of ideas (Birdi et al., 2016).

Studies on employee innovativeness have highlighted that job design (Holman et al., 2012) and work characteristics play a key role in promoting employee engagement in innovative behaviours (Hernaus et al., 2019; Parker, 2014; Shalley et al., 2000; Yuan and Woodman, 2010; Shin et al., 2017). When employees perceive that desired work outcomes demand innovation, they are more likely to engage in innovative behaviour (Frèour et al., 2024; Yuan and Woodman, 2010, 2021; Shin et al., 2017). These features are closely associated with knowledge work, in which a high level of autonomy, challenging tasks and conditions conducive to displaying innovative behaviour are present. The motivational factors (Amo, 2006; Amar, 2004), creative self-efficacy (Newman et al., 2018; Norena-Chavez and Guevara, 2020) and individual problem-solving style (Scott and Bruce, 1994) have been identified, in general, as antecedents of employee innovativeness. However, given the specific nature of KWs, our understanding is limited about the factors that promote their innovativeness. To address this research gap, this study builds on entrepreneurship research and focuses on KWs’ individual attributes, such as internal locus of control (ILC) and social competence in terms of enhancing their innovativeness. Highlighting individual attributes is appropriate because KWs conduct their professions relatively independently according to their dispositions (e.g. Kelloway and Barling, 2000; Kubo and Saka, 2002; Jacobs, 2017); that is, their work is entrepreneurial by nature.

Entrepreneurial research (e.g. Mueller and Thomas, 2001) that employs Rotter’s (1966) locus of control theory suggests that individuals with a strong ILC believe that they, rather than external circumstances, are responsible for the outcomes they experience. Therefore, they deliberately place their attention on their own efforts and abilities and prioritise these factors over external conditions when striving to achieve positive personal results (Ng et al., 2006). Hamzah and Othman (2023) found that ILC positively related with entrepreneurial competency, in which innovativeness is one dimension of competency. The study by Xu et al. (2020) demonstrated that ILC has both direct and indirect positive effects on innovative behaviour among Y-generation employees (born after 1980). Wijbenga and Van Witteloostuijn (2007) found that ILC is positively related to entrepreneurs’ engagement in innovative strategies. While ILC is associated with entrepreneurial activities (Krueger et al., 2000; Hamzah and Othman, 2023), including innovativeness (e.g. Xu et al., 2020; Hamzah and Othman, 2023), its role in enhancing KW innovativeness remains unexplored.

However, innovativeness does not occur in isolation—it thrives through collaboration with others (e.g. Axtell et al., 2000; Björklund et al., 2023) within social exchange relationships (e.g. Agarwal, 2014; Shalley et al., 2004) and networks (Cangialosi et al., 2021). In prior research, social competence (a set of social skills through which individuals interact effectively with others) is associated with entrepreneurial behaviour (Baron and Markman, 2003; Lans et al., 2015, 2016), building and maintaining social networks (Baron and Tang, 2009) and small-firm learning and performance (Lans et al., 2016).

Specifically, Baron and Markman (2003) found that a higher level of social perception (the ability to perceive other persons accurately) related to greater entrepreneurial success. Lans et al. (2016) found that owner-managers’ social competence significantly fostered small-firm performance. Additionally, Yun and Lee (2017) found that social skills moderating between knowledge sharing and R&D personnel’s work performance. Scholars also acknowledge that by employing social competence, entrepreneurs develop their social capital (Kreiser et al., 2013; Baron and Markman, 2003; Baron and Tang, 2009; Lans et al., 2016). Overall, there is a need to investigate social competence in the context of knowledge work because relatedness, need for social connections and belonging, is one of the key drivers of meaningful work (e.g. Ryan and Deci, 2000) and entrepreneurship (Hamzah and Othman, 2023).

Given this background, this study addresses the research question “what is the role of the internal locus of control and social competence in knowledge worker innovativeness?” We employed partial least squares structural equation modelling (PLS-SEM) and related procedures to test the proposed research models using data collected from 415 KWs in the fields of technology and architecture.

This study makes a novel contribution to the discussion of the individual attributes of KW innovativeness. First, by building on entrepreneurial research (e.g. Mueller and Thomas, 2001; Baron and Markman, 2003) and confirming the positive relationship between ILC and KW innovativeness, the study broadens the discussion to include individual attributes beyond mere motivational factors. Second, by examining the role of social competence in KW innovativeness and demonstrating social perception as a moderator in the relationship between ILC and KW innovativeness, this study deepens the understanding of the attributes that drive KW innovativeness. Additionally, our findings contribute to the literature on individual innovativeness by building on the work of Janssen (2000) and De Jong and Den Hartog (2008). Moreover, the contributions to the management of KW innovativeness (e.g. Tsai, 2018; Amabile and Pratt, 2016) are discussed in the section on managerial implications.

To investigate the individual attributes of KW innovativeness, we built on prior research on innovative work behaviour (e.g. De Jong and Den Hartog, 2008; Janssen, 2000), ILC (Rotter, 1966; Mueller and Thomas, 2001; Krueger et al., 2000; Hamzah and Othman, 2023) and social competence (Baron and Markman, 2000, 2003). We considered Amabile’s (1983, 1988) componential model of creativity and innovation in organisations, which explained innovativeness in terms of both individual factors (e.g. expertise, creativity skills and motivations) and contextual factors. In line with this approach, other scholars have focused either on motivational or contextual factors or both when explaining employee innovativeness.

Research has suggested that the personal traits of ILC and innovativeness are associated with entrepreneurship (Hsiao et al., 2016) and entrepreneurial potential (Mueller and Thomas, 2001). Individuals with a high ILC believe that their work outcomes are primarily shaped by their own actions (Rotter, 1966; Chen et al., 2016). Consequently, having a strong ILC is considered essential for engaging in independent entrepreneurial action. Innovativeness, in turn, refers to an individual’s ability to generate novel ideas and implement them in practice (De Jong and Den Hartog, 2008; Janssen, 2000). For example, Yuan and Woodman (2010) suggested that individuals engage in innovative behaviour when they expect that it will improve their work performance and when they expect risks and gains to their professional images.

Social competence, which is understood as a set of skills that enables individuals to interact effectively with others (Baron and Markman, 2000), includes the ability to perceive and interpret other people and to adapt to new social situations (Baron, 2000), such as when seeking partners to support and implement ideas. As innovativeness requires interaction with others (Axtell et al., 2000; Björklund et al., 2023), social competence is crucial for fostering innovativeness among KWs. For example, Yun and Lee (2017) found that social skills enhanced the relationship between knowledge sharing and work performance among R&D personnel.

A KW is someone “ … who knows more about his or her job than anyone else in the organization” (Drucker, 1999, p. 19) and whose primary responsibilities involve tasks related to the acquisition, generation and distribution of knowledge. KWs can be labelled also as knowledge professionals (Mazmanian et al., 2013) as they perform autonomous, complex and knowledge-based tasks within their professions by processing knowledge and organising their work accordingly (Edgar et al., 2015; Humphrey et al., 2007; Kubo and Saka, 2002; Davenport et al., 2002; Jacobs, 2017; Robinson and Marino, 2015). Thus, KWs can relatively independently organise their work according to their personal dispositions. In line with this view, Kelloway and Barling (2000) emphasised the view that knowledge work amounts to discretionary behaviour within organisations and comprises knowledge-related behaviours. Such a behaviour relies on individual’s dispositions to translate their abilities into action (Perkins et al., 1993), which are shaped by personal attitudes, values, emotions, interests and sense of identity (Prawat, 1989).

KW innovativeness can be understood, in line with De Jong and Den Hartog (2008), as the innovative work behaviour of a KW. As such, it refers to an individual’s ways of working by continually exploring opportunities and generating, promoting and implementing novel ideas (e.g. Janssen, 2000; De Jong and Den Hartog, 2008). In this regard, Scott and Bruce (1994) understood innovativeness as an individual’s discretionary behaviours.

The locus of control theory (Rotter, 1966) involves both the external and internal locus of control (ILC). ILC refers to an individual’s conviction that they are active agents and that their life outcomes are primarily determined by their own actions, decisions and abilities, whereas individuals driven by an external locus of control tend to view themselves as passive agents. Individuals with a strong ILC perceive themselves as having personal responsibility and autonomy in shaping their future (Chen et al., 2016; Rotter, 1966). They are proactive agents who shape their behaviour and maintain control over most situations they face.

Therefore, ILC is associated with independent work, such as self-perceived employability (Aybas and Kirbaslar, 2014), entrepreneurial intentions (Arkorful and Hilton, 2022), entrepreneurship (Boone et al., 1996; Lewin and Stephens, 1994). ILC, as an individual’s belief that their outcomes at work are mainly determined by their own actions (Rotter, 1966; Chen et al., 2016), is different from the concept of self-efficacy, which is an individual’s belief in their own ability to execute the necessary actions to perform desired tasks or goals (Bandura, 1982). Thus, ILC is seen as a necessary condition for acting (Mueller and Thomas, 2001), and it can be associated with the central characteristic of autonomous knowledge work. Additionally, individuals with strong ILC tend to be intrinsically motivated (Reeve et al., 1987). In this regard, ILC is an interesting but little-studied concept in the context of autonomous knowledge work.

Social competence, which comprises a set of skills through which individuals interact with others (e.g. Baron and Tang, 2009; Notari et al., 2014), is associated with many positive outcomes in organisations (Baron and Tang, 2009; Belliveau et al., 1995), the work performance of R&D personnel (Yun and Lee, 2017) and entrepreneurial success (Baron, 2000; Baron and Markman, 2000; Baron and Tang, 2009; Shane and Venkataraman, 2000). For example, Baron (2000) claimed that entrepreneurs’ strong social competence, referring to their ability to perceive and interpret other people and to adapt new social situations, predicts their success.

Social competence is further described as an overarching concept that encompasses the collective impact of multiple social skills (Spence et al., 1999; Baron and Markman, 2003). Previous research (e.g. Baron and Markman, 2003) presented a five-dimensional social competence concept comprising social perception, impression management (e.g. making a favourable first impression), persuasiveness, social adaptability and expressiveness. “Social perception” refers to the accuracy with which an individual perceives others, such as their traits, intentions and motives (e.g. Zebrowitz and Collins, 1997; Baron and Markman, 2003). “Impression management” comprises a wide range of techniques for inducing positive reactions in others (e.g. Bolino et al., 2008). “Persuasiveness” refers to an individual’s ability to convince others to adopt his/her point of view in face-to-face encounters (e.g. Shavitt and Brock, 1994), while “social adaptability” describes an individual’s ability to adapt to or feel comfortable in a wide range of social situations. Finally, “expressiveness” refers to an individual’s ability to clearly express emotions and feelings to generate enthusiasm in others (Baron and Markman, 2000, 2003).

With specific regard to knowledge work, this study focuses on the three dimensions of social competence: social perception, social adaptability and self-promotion. “Social perception” is associated with various positive outcomes, such as favourable performance evaluations by supervisors, (Hochwarter et al., 2007) and positive ratings by interviewers (Kacmar et al., 2003). Individuals who perceive other people accurately can select suitable collaboration partners or employees and are more likely to succeed in negotiations. “Social adaptability” refers to the ability to adapt individual behaviour to social situations. An individual who feels comfortable in a wide range of social situations has more opportunities for collaboration and is able to maintain her or his functionality even in adverse circumstances (Baron and Markman, 2003; Baron, 2000). “Self-promotion” refers to the ability to present a positive image of oneself and one’s accomplishments to others, which constitutes impression management (Baron and Tang, 2009; Baron and Markman, 2003). It involves strategically presenting oneself to others to build a positive image, gain recognition or advance personal or professional interests. Self-promotion can take various forms, such as highlighting accomplishments, skills or positive attributes through conversations, resumes or other promotional channels. It is essentially a means of advocating for oneself to create a favourable impression and attract opportunities.

In the main effect model, the relationship between ILC and KW innovativeness is discussed. ILC refers to the degree to which individuals perceive themselves as active agents with strong personal responsibility and autonomy in shaping their actions and the future (e.g. Zhao et al., 2023; Chen and Silverthorne, 2008; Chen et al., 2016; Rotter, 1966; Mueller and Thomas, 2001). In earlier research, ILC was associated with independent work, such as self-perceived employability (Aybas and Kirbaslar, 2014), entrepreneurial intentions (Arkorful and Hilton, 2022) and entrepreneurship (e.g. Boone et al., 1996).

However, its role in individual innovativeness remains surprisingly unexplored. In contrast, individual innovativeness in organisations is often explained with motivational (e.g. Newman et al., 2018; Norena-Chavez and Guevara, 2020; Scott and Bruce, 1994) and contextual factors (e.g. Chen et al., 2016; Haider et al., 2023; Ononye, 2023; Muchiri et al., 2020). Yet, knowledge work has entrepreneurial features: it is multifaceted (Jarrahi et al., 2023; Baer, 2003; Kelloway and Barling, 2000), with KWs independently organising their tasks and taking responsibility for various knowledge-related activities (Edgar et al., 2015; Humphrey et al., 2007; Jacobs, 2017). Therefore, it is appropriate to draw on entrepreneurial research, where ILC is used to explain the behaviour and performance of entrepreneurs based on the view that a strong ILC increases the likelihood of successful entrepreneurship (e.g. Miller and Toulouse, 1986).

Innovativeness is an intentional behaviour that demonstrates novelty and is dependent on KWs’ personal dispositions and behaviour (Kelloway and Barling, 2000) while conducting their work. KWs who demonstrate high ILC have a strong belief that they are active agents and bear responsibility for their own actions and performance (e.g. Zhao et al., 2023), and which is likely to encourage them to engage in innovative behaviours. On this basis, we posit the following hypothesis.

The relationship between an independent and dependent variable is often influenced by a moderator variable, which can affect both the direction and/or strength of the relationship (Baron and Kenny, 1986). In this study, we employed three social competence dimensions as moderator variables because KWs likely differentiate in terms of their personal dispositions and social competencies to interact with other people. For example, Baron (2000) suggested that individuals who possess well-developed social competencies likely experience favourable outcomes because they can interpret and control social interactions effectively to achieve their motives and goals (Yun and Lee, 2017; Baron and Tang, 2009; Notari et al., 2014). In the following section, we discuss our hypotheses in more detail.

Based on the prior research, we claim that the skill to perceive and read other people’s traits, intentions and motives accurately (Zebrowitz and Collins, 1997; Baron and Markman, 2003), such as truthfulness of intentions, can influence the relationship between ILC and KW innovativeness. This aligns with findings that revealed that social perception influences human behaviour (e.g. Ferguson and Bargh, 2004). For example, social perception might play an important role when a KW interacts with others to seek collaboration partners and support for their novel ideas. A KW’s high level of social perception likely strengthens the relationship between ILC and KW innovativeness. In other words, KWs who rely on their abilities, actions, decision (ILC) to be innovative are likely to demonstrate even greater innovativeness when their social perception is high. In line, social skills moderated the relationship between knowledge sharing and performance among R&D personnel (Yun and Lee, 2017).

In contrast, when a KW’s level of social perception is low, the relationship between ILC and KW innovativeness is weaker. It is likely that KWs with a low level of social perception face difficulties in interpreting other people’s motives. Thus, they often prefer to remain alone in terms of developing and implementing their ideas, which likely leads to lower innovativeness. On this basis, we posit the following hypothesis.

The skill to adapt and feel comfortable in a wide range of social situations (Baron and Markman, 2003; Baron and Tang, 2009; Cooper and Hetherington, 2005), including rapid changes (e.g. Baron and Markman, 2003; Mackey et al., 2013), can be beneficial for KW innovativeness. Social adaptability is also seen as a self-regulatory skill through which individuals use resources to address emerging demands (Van den Tooren and De Jonge, 2010). As individuals differ in terms of their adaptability (Chan, 2000; Mackey et al., 2013), we propose that the KW’s ability to adapt to various social situations is likely to strengthen the relationship between ILC and KW innovativeness. That is, KWs with strong social adaptability are ready to engage in and maintain functioning in collaborations and even in adverse and fast-changing circumstances (Baron and Markman, 2003; Mackey et al., 2013). This is likely to strengthen the relationship between ILC and KW innovativeness. In contrast, KWs with a low level of social adaptability may avoid unfamiliar social situations, preferring to stay in “safe” social contexts, thus losing the possibility of the enhancement effect of social relationships and the coincidental drivers of innovativeness. Therefore, we propose that when a KW’s level of social adaptability is high, the relationship between ILC and KW innovativeness is strongest, whereas with a low level of social adaptability, the relationship between ILC and KW innovativeness is weakest.

In the realm of self-promotion, social competence involves what is known as impression management (e.g. Bolino et al., 2014; Bolino et al., 2016). This entails employing a variety of techniques to elicit positive reactions from others. In scholarly discussions, the concept of impression management is tied to an individual’s skill in shaping, upholding or safeguarding the perception that others hold of them (Rosenfeld et al., 1995; Bolino et al., 2014).

According to earlier research, the fact that individuals care about their images may influence their willingness to engage in proactive behaviours (e.g. Grant et al., 2009; Bolino et al., 2016). That is, KWs high in self-promotion aim to build and maintain a positive image of themselves and their accomplishments in the minds of others. In some cases, this may reduce their interest in engaging in novel experiments of idea generation and implementation. Specifically, while experiments are inherently involved in innovativeness (e.g. De Jong and Den Hartog, 2008), some KWs may see failure as damaging to their image.

In contrast, KWs low in self-promotion do not care about promoting their images; when self-promotion is low, it likely strengthens the relationship between ILC and KW innovativeness (the relationship between ILC and KW innovativeness is strongest). KWs who make no effort to maintain their images for others likely rely on their own competencies, consider themselves capable and courageous to demonstrate novel ideas and experiments. When a KW’s self-promotion is strong, however, it weakens the relationship between ILC and KW innovativeness as the effort to self-promote may direct attention to maintaining an image and away from actual innovative work.

To empirically examine the relationships between KWs’ ILC and innovativeness and to investigate the moderating effect of social perceptiveness, social adaptability and self-promotion on this relationship, we adopted a purposive sampling approach while testing the external validity of the research. Purposive sampling involves selecting units of analysis, such as individuals, groups of individuals or organisations with a well-defined, specific purpose of answering research questions (Etikan et al., 2016).

We surveyed individual members of the Academic Engineers and Architects trade union (TEK) in Finland. We chose the trade union as the members are all university graduates in technology and architecture who hold roles as experts or occupy leadership positions within organisations, thereby making them a representative sample of knowledge workers (Rüdiger and McVerry, 2007). The individuals were representative of the members of TEK and comprised KWs of different ages, genders and technical fields. Our data were gathered through a web-based survey that garnered 423 responses out of 4,940 distributed questionnaires. (An 8.6% response rate is typical for similar web-based surveys that are distributed by TEK to their members.) The effective sample size was 415 responses.

The survey was in the interest of the members of TEK as there was ongoing discussion at the time regarding the changing nature of knowledge work in technology. The respondents were invited to answer the survey through their monthly newsletter. In the survey, the dependent, independent and demographic variables were organised in themes. The anonymity of responding was highlighted, and the respondents were rewarded with the possibility of taking part in a lottery for a gift voucher of their choice.

Among the 415 respondents, the gender distribution was 26.7% female (N = 111) and 73.3% male (N = 304), a ratio that closely mirrored the composition of Finnish university graduates in technology and architecture. In terms of educational attainment, 78.8% held a master’s degree (N = 327), 15.4% possessed a doctoral degree (N = 64), 3.1% held a bachelor’s degree (N = 13) and 2.7% had other educational backgrounds (N = 11). The respondents’ ages were distributed as follows: 12.8% (N = 53) were aged 29 or below, 20.1% (N = 84) were aged 30–39, 26.5% (N = 110) were aged 40–49, 21.4% (N = 89) were 50–59 and 19.0% (N = 79) were 60 years old or older.

We took the following steps to reduce the risk of potential common method bias (CMB) stemming from self-reported data (e.g. Minbaeva et al., 2012). To reduce any risk of the respondents answering to align with the expectations of others, the survey design and administration explicitly assured confidentiality (Minbaeva et al., 2012). Further, we used clear wording and understandable terminology to keep the survey concise (MacKenzie and Podsakoff, 2012; Podsakoff et al., 2024). We tested the survey with 16 knowledge workers (11 male and 5 female) to improve the scale items. Additionally, the contact person in charge of research at TEK assessed the survey’s wording and terminology (MacKenzie and Podsakoff, 2012). Based on their feedback, we made necessary adjustments to improve the survey’s quality. The fact that the survey asked experienced respondents to assess concrete constructs further reduced the possibility of CMB (see MacKenzie and Podsakoff, 2012). Moreover, the anchoring of the scales varied (i.e. it was different for ILC, innovativeness and social competence), which helped reduce the possibility of CMB (Podsakoff et al., 2003; MacKenzie and Podsakoff, 2012).

Moreover, Harman’s single-factor test revealed five distinct factors that collectively explained 60.84% of the total variance. Notably, the primary unrotated factor accounted for 32.74% of the variance, indicating that no single factor predominated. Finally, analysis using PLS-SEM demonstrated strong discriminant validity, thus providing reassurance regarding any CMB concerns (Ahammad et al., 2017). Overall, these assessments collectively suggested that CMB was unlikely to be a significant issue.

We used multi-item measures with a 7-point Likert scale for the constructs based on the measures in the existing literature (see  Appendix 1).

Dependent variable: KW innovativeness was measured using four items from the innovative behaviour scale by Janssen (2000) and one idea generation item from the scale of creativity (Tierney et al., 1999). This aligned with the scale of innovativeness comprising dimensions of idea generation, idea promotion and implementation (De Jong and Den Hartog, 2008; Oldham and Cummings, 1996; Janssen, 2000). As our survey measured both creativity and innovativeness, it was important to avoid repetition of the items of idea generation.

Independent variable: We measured ILC with the five-item scale by Chen et al. (1998). One item was removed due to low factor loading.

Moderation variables: Social competence was measured with the self-perceived social competence scale presented by Baron and Markman (2003), Baron and Tang (2009) and Bolino and Turnley (1999). The scale involved three dimensions of social competence: social perceptiveness (5 items), social adaptability (4 items) and self-promotion (4 items). One item was removed from the social adaptability scale due to a low factor loading.

Control variables: We employed the demographic variables relating to age, gender and education as control variables.

We employed partial least squares structural equation modelling (PLS-SEM) for our analysis (e.g. Ringle et al., 2015; Ringle et al., 2024) in line with the established guidelines (e.g. Hair et al., 2014). First, we assessed the reliability and validity of the measurement models. Next, we applied the structural model to test our hypotheses. To examine the moderation effects, we analysed both the direct relationships between the variables and the interaction term, following recommendations from the literature (e.g. Baron and Kenny, 1986).

While PLS-SEM and covariance-based structural equation modelling (CB-SEM) are considered complementary methods (e.g. Jöreskog and Wold, 1982), scholars recommend using PLS-SEM when the study’s goal is to predict key target constructs or to identify key drivers rather than focusing on theory testing (Hair et al., 2011). Additionally, PLS-SEM is particularly favoured for exploratory research models (e.g. Hair et al., 2011).

Table 1 presents the descriptive statistics and the correlation analysis for the actual analyses and testing of the hypotheses. We first assessed the reliability and validity of the measurement model and then used the structural model to test our hypotheses.

To test the measurement model, we assessed internal consistency and discriminant validity, both of which showed good validity and reliability for the operationalisation of the concepts under examination. The construct reliabilities (CR values) of all our constructs were above the recommended threshold of 0.7 (Bagozzi et al., 1991). Also, the factor loadings of the items were high enough and statistically significant, which indicated that all items were related to their specific constructs, thereby verifying the relationship posited among the indicators and constructs ( Appendix 1). As shown in Table 1, the average variance extracted (AVE) values exceeded the cut-off point of 0.5 (e.g. Fornell and Larcker, 1981). Finally, the tests of discriminant validity (Table II) showed that each construct’s squared AVE was greater than the correlation between the other constructs. As Table 3 displays, the heterotrait-monotrait ratios (HTMT) for all pairs of constructs varied between 0.388–0.590 and undercut the threshold value (0.90). These results indicated that the constructs in the model differed from each other. See  Appendix 1 for the factor loadings, CRs and AVEs.

Additionally, collinearity analysis indicated no presence of multicollinearity among the indicators, as all the variance inflation factor (VIF) values ( Appendix 2) were below the benchmark threshold of 5.0 (e.g. Hair et al., 2016).

The results indicated that ILC (β = 0.296; p = 0.000; f² = 0.11), self-promotion (β = 0.26; p = 0.000; f² = 0.060), social adaptability (β = 0.112; p = 0.046; f² = 0.01) and social perception (β = 0.139; p = 0.008; f² = 0.02) positively related to KW innovativeness. The main effect model explained 36.0% (R² = 0.360) of the variance in KW innovativeness. Additionally, the control variables of age (β = 0.112; p = 0.012) and education (β = 0.089; p = 0.029) were significantly and positively associated with KW innovativeness, while gender showed no significant relationship (β = 0.155; p = 0.108) (see Figure 1).

As proposed in hypothesis 2, the interaction term social perception x ILC significantly related to KW innovativeness (β = 0.108; p = 0.045). In contrast, hypothesis 3 and hypothesis 4 were rejected (Table 4). Figure 2 displays the moderating effect graph. The effect size, f² = 0.02, for the relationships between the interaction term of social perception x ILC and innovativeness was small (Cohen, 1988). The Q² value, which measured the model’s predictive relevance for innovativeness, was 0.31 (>0), which indicated that the model had predicative relevance. Of the control variables, age (β = 0.118; p = 0.007) and education (β = 0.089; p = 0.031) were significantly related to KW innovativeness. Overall, the model was able to explain 37.4% of the variance in KW innovativeness.

Amid the increasing prominence of KWs in modern organisations (Berg and Gmyrek, 2023) and considering the distinct nature of knowledge work (Jarrahi et al., 2023; Baer, 2003; Kelloway and Barling, 2000) and its pivotal role in driving organisational performance (Seth and Lee, 2017; Zhang-Zhang et al., 2022), understanding the individual factors that foster innovativeness among KWs is of critical importance. While prior research has explored motivational (Newman et al., 2018; Norena-Chavez and Guevara, 2020) and contextual (Chen et al., 2016; Haider et al., 2023) drivers of innovativeness, as well as key work characteristics (e.g. Hernaus et al., 2019; Holman et al., 2012; Yuan and Woodman, 2010; Shin et al., 2017), the individual attributes of KW innovativeness remain underexplored.

This study fills this gap by shifting the focus toward the individual dispositions and skills that shape innovativeness at work.

In exploring the relationship between ILC and KW innovativeness, we examined the conditions under which this relationship is strongest or weakest. To achieve this, we introduced three dimensions of social competence as boundary conditions that influence KW innovativeness. First, our findings confirmed that ILC significantly and positively related with KW innovativeness. KWs with a strong ILC view themselves as responsible for their innovativeness and demonstrate it through their intentional actions.

Second, we found that when KWs’ social perception is high, the relationship between ILC and KW innovativeness was strongest, whereas with low social perception the relationship was weakest. Thus, by honing their social perception skills, KWs can effectively identify potential collaborators who can provide support for their ideas and initiatives. This ability enables them to discern individuals who can assist in turning ideas into practical solutions. In sum, high social perception helps KWs to network and collaborate to enhance their innovativeness. In contrast, low social perception skill likely prevents a KW from seeking support and collaboration, which is likely to weaken the relationship between ILC and KW innovativeness.

Regarding the demographic variables, our findings showed that age and education positively influenced KW innovativeness. This suggests that greater experience and higher levels of education enhance innovativeness.

This study offers contributions to the existing theory and research on individual innovativeness. While previous research has focused on employee innovativeness in organisations (e.g. Scott and Bruce, 1994; Chen et al., 2016), this study recognises that KWs represent a specific type of employee as their work is highly autonomous and challenging, and it inherently demands innovativeness. By drawing on research around ILC (e.g. Mueller and Thomas, 2001) and social competence (Baron and Markman, 2003), which are commonly used to explain entrepreneurial behaviour and performance, this study extends the understanding of the individual attributes of KW innovativeness. As an implication to the model of creativity and innovation in organisations (Amabile, 1988), our study broadens the individual factors to consider not only creativity skills and motivational factors but also attributes like ILC and social competencies, which were overlooked in earlier research.

While social relationships are recognised as key drivers of individual innovative behaviour (e.g. Axtell et al., 2000; Björklund et al., 2023; Agarwal, 2014), this study broadens the understanding of KW innovativeness with the consideration of social competence (Baron and Markman, 2003). Specifically, it shows that social perception serves as a boundary condition in the relationship between ILC and innovativeness.

To conclude this study broadens the understanding of the individual determinants of KW innovativeness by providing insights from entrepreneurship research. Shifting focus to individual attributes is relevant, as knowledge work relies on the personal dispositions and skills of KWs. Thus, our study moves beyond the contextual factors examined in previous research (e.g. Dul et al., 2011; Nisula et al., 2022). In doing so, it also enriches the broader literature on individual innovativeness, including the works of Janssen (2000) and De Jong and Den Hartog (2008).

With respect to managerial implications, this study provides the following implications for organisations. Considering that knowledge work heavily relies on individual competencies, capabilities and inclinations, effective management of KWs entails acknowledging their individual factors. While KWs with high ILC engage in innovative behaviours, KWs with low ILC may need greater encouragement and support. First, organisations could offer tailored employee support and development opportunities to address the unique needs of individual KWs. For KWs with low ILC, human resource management (HRM) practitioners could offer additional encouragement by providing support structures, opportunities for competence and skill building and mentorship programs (e.g. Jayasingam and Yong, 2013).

Second, organisations would benefit from recruiting KWs with the specific capabilities of ILC and social competence. Therefore, we call for development in recruitment and selection practices to identify individuals with personals traits such as ILC and social competence, especially within innovative industries involving complex and continuous changes. Third, by offering frequent opportunities and room for social interaction, organisations can stimulate the development of KWs’ social skills.

Finally, in terms of implications on the individual level, our research suggests that KWs consciously cultivate not only their personal resources and skills but also work practices and structures that enable them to fully harness their capabilities in pursuing their competencies and autonomy. We believe that adapting the organisation’s innovation management to the evolving needs of KWs—can enhance their innovativeness. For example, organisations could offer options such as flexible working hours or remote work, depending on the preferences of the KWs. Such self-customised work practices and flexibility are likely to enhance their ILC, which can foster their innovativeness at work.

With respect to policy implications, this study highlights the individual attributes of KWs that are crucial for fostering their innovativeness. KWs represent a growing (e.g. Berg and Gmyrek, 2023) and critical group of employees for organisational functioning and performance (e.g. Seth and Lee, 2017; Zhang-Zhang et al., 2022). Given the rapid and profound shifts in the nature of work—such as the rise of platform work for highly skilled KWs (Nemkova et al., 2019; Rolandsson et al., 2020), hybrid work (Jarrahi et al., 2023), self-employment (Jarrahi et al., 2023) and the expansion of the digital labour market (Dorschel, 2022; Graham et al., 2017) that reshape the way knowledge work is carried out (Katz and Krueger, 2019)—policymakers need to acknowledge these societal changes. They should develop strategies, regulations and guidelines to support the education and development of KWs to ensure they are prepared to meet the diverse and evolving demands of modern work life.

Although this study offers novel contributions, it is not without theoretical and empirical limitations. First, while this study identified a significant and positive moderation effect of the interaction term social perception x ILC for KW innovativeness, the effect size was relatively weak. The fact that the model was tested with a specific and limited group of KWs may explain the observed effect size. Therefore, future studies could test the model with a wider variety of KW groups. Additionally, future research could explore social skills across different types of knowledge work and further clarify their role by examining a broader range of social skills. Future studies should also investigate both the direct effects of social skills and their mediating effects on KW innovativeness. Because social relationships are key to individual innovativeness (e.g. Axtell et al., 2000), the social skills of KWs warrant more attention as a driver of innovation. The ability to interpret and leverage social relationships and collaboration can significantly enhance personal achievements, such as innovativeness.

Second, this study relied on self-reported measure of variables, which increases the likelihood of common method bias (CMB) (e.g. Podsakoff et al., 2003). Self-reported measure is relevant as KWs are in the best position to assess their innovative behaviour (Shalley et al., 2000; Conway and Lance, 2010) and social competence. Despite the various methods available to reduce potential CMB, researchers should focus on study design and adopt strategies that diversify the assessment of self-reported data, such as using longitudinal studies to capture data over time.

Third limitation arises from our sample. Our sample’s demographic makeup predominantly consisted of males from the technology and architecture sectors. The observed positive correlation between gender and KW innovativeness may therefore be skewed due to the high proportion of male participants (73.3%). Relatedly, the study is limited by the relatively narrow focus of our sample, which primarily included highly educated professionals in the fields of technology and architecture, all of whom worked as experienced professionals in traditional organisational settings. Considering the diverse nature of modern knowledge work and the rise of hybrid work models, future research should explore a wider variety of KWs, including self-employed professionals, hybrid workers and entrepreneurs from a broader range of disciplines and job roles. We also recommend that future studies examine knowledge work in diverse contexts, such as digital platforms, temporary projects and in technology- or AI-mediated environments, while taking into account a range of cultural, economic, political and social factors. These studies could offer valuable insights as the innovativeness of KWs may differ from those in more traditional organisational roles.

Finally, as our data were obtained from KWs in Finland, the results may not be generalisable to other countries with significantly different economic, political, cultural and social contexts. That is, Finland has a relatively strong legal system as well as strong negotiation power within labour unions; thus, the study results should be somewhat comparable with other Western countries with similar legal systems and unions. On another note, the concept of social perception could also be somewhat culturally specific, which is something that future studies should consider. Therefore, it would be valuable to assess the findings of this study in other types of cultures, such as those with different social proximity, referring, for example, to the social norms and traits in business relationships.

Funding: This research was funded by the Finnish Funding Agency for Technology and Innovation (Tekes): 2923/31/2014, decision date 11.11.2014.