T HE E FFECTS OF P ROFESSIONALIZATION ON THE
I NNOVATIVENESS OF F AMILY F IRMS
A N E XTENSION OF THE P ROFESSIONALIZATION C ONSTRUCT
MS C . THESIS
University of Groningen Faculty of Economics and Business
by
Gerben Wiersema – S2300311 Small Business & Entrepreneurship
Supervisor: dr. M.J. Brand Co-Assessor: dr. F. Noseleit
Date: 12-4-2018
Word Count: 11035
1
Abstract
In this thesis the effect of professionalization on innovativeness is researched in the family firm context. The literature suggests that negative relationships exists. Nonetheless, empirical evidence is lacking. Therefore, the effects of the professionalization dimensions; internal formalization (IF) and effective openness (EO), on innovativeness are researched. In addition, the professionalization construct explains little variance of family firm performance. It is therefore proposed to complement professionalization with a third dimension; innovation orientation (IO). The resource- based view and the dynamic capability view are used as theoretical foundation. The data of a questionnaire among 110 family firm CEOs is analysed. The study finds no significant relations between professionalization and innovativeness. Using regression analysis, this study finds that IO is significantly related to performance but does not increase the explanatory power when including the concept in an extended professionalization construct.
Keywords: Professionalization, Family Firms, Innovativeness, Innovation Orientation
2
Management Samenvatting
Zowel professionaliseren als innovatie zijn veelvoorkomend thema’s in onderzoek naar familiebedrijven. In de praktijk blijken dit uitdagende concepten te zijn voor familiebedrijven. Dit onderzoek gaat over professionalisering en de effecten hiervan op de innovatiekracht van familiebedrijven. In recentelijk onderzoek is professionalisering uitgewerkt tot een concept bestaande uit twee dimensies. De eerste dimensie gaat over het intern formaliseren van het bedrijf.
Hiermee wordt bedoeld dat het bedrijf op een formele manier wordt geleid en de invloed van de familiale kenmerken geminimaliseerd is. Dit gebeurt bijvoorbeeld door formele; trainingen, werving, doelstellingen en rapporten. De tweede dimensie gaat over effectieve openheid. Deze dimensie is vooral gericht op het openstellen van een bedrijf voor externe personen en ideeën in het management. Sommige academici verwachten dat professionalisatie negatieve effecten heeft op de innovatiekracht van bedrijven. De effecten van professionalisering op innovatie en bedrijfsprestaties zijn echter nog niet onderzocht en dit vormt het hoofddoel mijn onderzoek.
Een tweede doelstelling is om te onderzoeken of professionaliseren invloed heeft op het presteren van familiebedrijven. Dit hangt samen met de vraag of professionaliseren alleen uit de twee eerdergenoemde dimensies bestaat, of dat er nog een dimensie mist. Eerder onderzoek heeft namelijk nog niet kunnen aantonen dat deze dimensies tot betere prestaties leiden en het concept is nog niet af. Ik onderzoek of, als derde dimensie, innovatie oriëntatie belangrijk is om prestaties te kunnen verklaren. Innovatie oriëntatie meet de mate waarin een bedrijf gericht is op innovatie.
Dit uit zich bijvoorbeeld door; innovatie proactief te benaderen, continue te zoeken naar nieuwe technologieën, werknemers te stimuleren innovatief te werken of een duidelijke visie te hebben.
Voor dit onderzoek is een vragenlijst ingevuld door 110 familiebedrijven. Uit het analyseren van
de data blijkt dat geen significante effecten gevonden kunnen worden tussen professionaliseren en
innovatiekracht. Wel blijkt uit de resultaten dat innovatie oriëntatie een belangrijk rol speelt bij het
verklaren van prestaties. De belangrijkste management implicaties van het onderzoek zijn dat een
innovatie oriëntatie (Zie tabel 1, p15) een positieve invloed heeft op de prestaties van
familiebedrijven en dat professionalisering geen significant effect heeft op innovatiekracht. Hieruit
blijk dat bedrijven kunnen professionaliseren zonder dat dit negatieve effecten op de
innovatiekracht of de financiële prestaties van het bedrijf heeft.
3
1. Introduction
Family firms are often expected to professionalize and become more like non-family firms (Stewart
& Hitt, 2012). There is, however, still a lot of inconclusiveness about how to professionalize a firm.
In recent research, professionalization is developed as a multidimensional construct for family firms (Dekker, 2012). How this concept of professionalization relates to innovativeness is currently unknown. The relationship between professionalization and innovation is therefore the main focus of this thesis. Innovativeness in family firms has been increasingly studied by scholars (e.g. De Massis, Frattini, Pizzurno, & Cassia, 2015; Duran, Kammerlander, Van Essen, & Zellweger, 2016), as innovation is argued to be one of the key capabilities to be competitive in the 21-st century (Gudmundson, Hartman, & Tower, 1999). Because family firms are responsible for a significant share of economic growth and technological progress (Zahra, 2005; Zahra, Hayton, & Salvato, 2004), it is important to research for family firms specifically.
Family firms are viewed as a distinctive group in business research due to their familiness.
Familiness refers to unique resources and capabilities that result from family influence. Family firms face unique challenges because of their familiness (Chua, Chrisman, & Bergiel, 2009; Kraus, Craig, Dibrell, & Märk, 2012). One of these challenges originates from the point where the growth of the firm surpasses the capabilities of the family to manage this growth independently and effectively. Typically the firm then reaches the point where the informal and familial systems in place become inefficient (Flamholtz 1986; Suáre & Santana‐Martín, 2004). At this point, family firms often engage in the process of professionalization (Chua et al., 2009).
Professionalization can have different meanings in different contexts and on different levels
(Stewart & Hitt, 2012). On the individual level professionalization is a process that is about
continuous improving ones capabilities (Stewart & Hitt, 2012). Moreover, when it was initially
introduced in the business literature, professionalization was proposed to be something alternative
to the bureaucratization of firms (Stewart & Hitt, 2012). Bureaucratization in most cases has a
negative connotation, while professionalization is more positive. Hofer and Charan (1984) describe
the professionalized firm as having a functional organizational structure, based on a firm’s current
needs. Moreover, the authors state that in a professionalized organization; decision making is
4 delegated to lower level managers, formal systems for information are present, stable corporate strategies are defined, and the firm is not dependent on one individual (Hofer & Charan, 1984).
In the family firm context, professionalization also plays a role in supporting the family to efficiently control a growing firm (Dekker, 2012). Stewart & Hitt (2012) argue that professionalization should provide the family with the means to successfully manage and operate their growing firm, and become more like a non-family firm (Stewart & Hitt, 2012). For successful management, Dyer (1989), identified three ways in which a family firm can professionalize. These are; 1) professionalizing the owners of the founding family, 2) professionalizing the employees working for the firm or 3) bringing in outside professional managers (Dyer, 1989). To successfully operate a professional firm, the literature suggests organizational and operational tools such as formalization, standardization and systemization (e.g. Dekker 2012, Hall & Nordqvist, 2008).
Dekker (2012) developed and tested a multidimensional professionalization construct for family firms consisting of five factors: (1) financial control systems; (2) nonfamily involvement in governance systems; (3) human resource control systems; (4) decentralization of authority and (5) top level activeness. This resulted in two professionalization dimensions, effective openness (EO) and internal formalization (IF). The first is about the governing aspects of the family firm, while the latter is about the implementation of effective control systems (Dekker et al., 2013).
1.1 Problem identification
The development of a broad construct by Dekker is a meaningful contribution to the literature regarding professionalization in family firms. Hitherto, the construct had been operationalized by using only a limited number of variables. Some authors operationalize professionalization as a binary variable of having non-family managers working for the firm (e.g. Zhang & Ma, 2009). This is over simplified, as professionalization should be a multidimensional construct (Stewart & Hitt, 2012).
The construct introduced by Dekker is multidimensional and focusses on organizational
characteristics (Dekker, 2012). What it lacks are operational aspects, what it should have according
to Stewart and Hitt (2012). Moreover, the construct is of a static nature, and not continuous
.In
some papers it is suggested that it is something ongoing (Hwang and Powell, 2009; Stewart and
5 Hitt, 2012). As is the case for individuals, professionalization should entail continuous improvement of capabilities (Stewart & Hitt, 2012). The argument for this is that a firm cannot be suddenly professionalized and remain as such without devoting continuous attention to this. In the construct described by Dekker (2012), professionalization is more a status or label for a firm to attain by implementing certain aspects in an organization. In line with the dynamic capabilities view, the goal of professionalization should rather entail to be a capability that can generate sustained advantages for longer periods. Moreover, Dekker (2012) indicates that there are likely other dimensions to her construct of professionalization than established so far. In her research financial control and top-level activeness, two out of five factors, could not be significantly linked to performance (Dekker et al., 2013). In the research of Dekker only six percent of the performance variable was explained (Dekker et al., 2015).
The role of innovation is not considered in the development of the professionalization construct by Dekker (2012). That both concepts have not been linked is remarkable, as innovation management is argued to be a core challenge and critical for the survival of organizations (Kraus, Pohjola, &
Koponen, 2012; Mumford & Licuanan, 2004). According to Kraiczy, Hack and Kellermanns (2015), professionalization could even hamper innovation. Professionalization tools such as formalization and standardization are found to be negative for innovation (Damanpour, 1991;
Jansen, Van Den Bosch, & Volberda, 2006). Professionalizing therefore might make the organization rigid and less adaptable to changes in their environment. The effects of professionalization on firm innovation have not yet been empirically researched for the professionalization construct of Dekker (2012). Because professionalization is multi-dimensional, I will research the effects of Internal Formalization and Effective openness on the innovativeness of family firms. This results in the following research question and sub-questions:
Research Question 1: What is the effect of professionalization on the innovativeness of family firms?
Sub question 1a: What is the effect of Internal Formalization on Innovativeness?
Sub question 1b: What is the effect of Effective Openness on Innovativeness?
In addition to researching the effect of professionalization on innovativeness, this thesis has a
6 second goal. This goal is to research if there is ground for enhancing the professionalization construct with an additional dimension. There are several reasons to suspect that this is the case.
First of all, Dekker (2012) indicates that it is likely that some dimensions are still missing to her construct. Also, her construct only explained 6 percent of the performance variance (Dekker et al., 2015). Second, in the literature innovative orientation is suggested for families firms to lead to growth (Stenholm, Pukkinen, & Heinonen, 2016). This is related to professionalization, as an innovation orientation refers to providing the antecedents that lead to innovativeness (Siguaw, Simpson, & Enz, 2006). This has so far not been linked to professionalization. Third, several papers implicitly or explicitly regard professionalization as something continuous (e.g. Hofer & Charan, 1984; Hwang and Powell, 2009; Stewart and Hitt, 2012). In addition to this, the dynamic capability view suggests that capabilities enable managers to reconfigure resources (Eisenhardt & Martin, 2000), and therefore concepts should not be static.
This thesis researches if, by extending the professionalization construct with innovation orientation, a more profound relationship with family firm performance can be found. The second research question therefore is:
Research Question 2: Does an extended professionalization construct have better explanatory power of family firm performance compared to the professionalization construct introduced by Dekker (2012)?
To answer these questions this thesis provides a literature review on the firm level. The resource- based view and the dynamic capabilities view are used as theoretical foundation. Furthermore, the findings of a quantitative study among 110 Dutch family firms are presented.
1.2 Relevance and research gap
Dekker (2012) suggested that linkages with other academic fields such as innovation would
contribute to the construct of professionalization. Dekker et al. (2013) and Stewart and Hitt (2012)
addressed the need for more empirical research concerning professionalization. Likewise, some
authors advocated for more empirical research regarding innovation in the family firm literature
(De Massis et al., 2015; Duran et al., 2016). At last, the role of innovation in family firm research
has been found to be under-researched (De Massis, Frattini, Pizzurno, & Cassia, 2015; Kraus et al.,
7 2012). To the best of my knowledge there is no empirical evidence on the relationships between a multidimensional professionalization construct and innovativeness in the family firm context
1. From an academic perspective it is meaningful to research if, and how, professionalization and innovation are connected in the family firm context. Connecting the two can guide academics to further research the effects of professionalization on innovation. This thesis adds to the current literature by improving the measurement of the professionalization construct that was recently introduced by Dekker (2012). Furthermore, this research provides insights to the ongoing research on the innovativeness of family firms (Duran et al., 2016; Kraus, Pohjola, et al., 2012) . Innovation research in general as well as for family firms specific has been criticized to lack practical implications (Fuetsch & Suess-Reyes, 2017). A final contribution from this thesis therefore is that it tests the effect of innovation orientation on performance and finds a significant relationship for this.
2. Literature review
This section discusses where this thesis is situated in the existing literature. The context of family firm theory is described, and the development of the professionalization construct is reviewed.
Furthermore, the concepts of innovation orientation and innovativeness are elaborated on. Based on this, the conceptual model and the hypotheses are developed.
2.1 Family firms
Family firms represent roughly two thirds of the firms and therefore are an important group to study (Duran et al., 2016). Family firm definitions are abundant and saying what exactly a family firm is remains difficult. In a meta-analysis of over 250 articles Chua et al. (1999) suggest that family firms are defined based on; 1) the family having a majority share and 2) by the family shaping the directions of the firm. The family thus should be dominant, with a substantial share of ownership or decision-making rights. The family should have a long term vision in which the family pursues the option to maintain control (Chua et al., 1999). Although this definition provides the means to distinguish family firms, it does not yet explain why family firms are different from non-family firms.
1Diéguez‐Soto, J., Duréndez, A., García‐Pérez‐de‐Lema, D., & Ruiz‐Palomo, D. (2016) Research the effect of professionalism on innovation, however they take a binary approach to professionalism
8 2.1.2 Family Firm Theory
Family firms are a different and complex set of firms when compared to non-family firms. This is because they constantly have to find the right balance between three entities; the family, ownership and the business (Habbershon, Williams, & MacMillan, 2003). This results in different conflicts of interests between the entities, as some persons in the firm can belong to all three, while others only belong to one or two entities. In family firm research the relationships between family and business systems are essential to explain phenomena (Sharma, 2004). The constant interaction of the different entities results in unique resources and capabilities that family firms have compared to non-family firms (Kraus, Pohjola, et al., 2012; Pearson, Carr, & Shaw, 2008). This is referred to as familiness in the family firm literature (Habbershon et al., 2003; Stewart & Hitt, 2012).
Familiness is thus the encompassing term for the unique resources and capabilities of a family firm (Pearson, Carr, & Shaw, 2008). Resources are the specific assets that are present in the firm.
Examples of resources are an educated workforce, a wide network or a strong brand. When these resources are valuable, rare, in-imitable and non-substitutable (VRIN), they lead to sustained competitive advantages according to the resource-based view (Eisenhardt & Martin, 2000).
Capabilities are unique resources that are embedded in the organization and have the purpose to improve the productivity of other resources (Pearson et al., 2008). This is in line with the view that a professionalization construct should be continuous and aimed at improving a firm.
Examples of family firm capabilities are efficiently sharing information or strong collective goals (Pearson et al., 2008). The dynamic capabilities view complements the RBV with the argument that resources alone are not sufficient for creating sustained competitive advantages. A firm needs complementary capabilities that can be used by managers to integrate, build and reconfigure resources to adapt to their environment for sustainable competitive advantages (Eisenhardt &
Martin, 2000; Teece et al., 1997). Such complementary capabilities lead to innovativeness and
higher firm performance (Craig, Dibrell, & Garrett, 2014) In an article by Habbershon and
Williams (1999) a model is provided that depicts how family inputs lead from familiness to
capabilities and eventually to competitive advantages and performance (Fig 1). Herein, the authors
illustrate that strategy and process intervention should be used to “continuous reevaluated and
replenished” familiness and capabilities that come from this (for example in family- or
management meetings) (Habbershon & Williams, 1999). The figure shows that familiness can then
lead to competitive advantage and increased performance. Professionalization in this thesis is seen
9 as a capability between familiness and competitive advantage in the figure. This provides the underlying theory for the thesis and the development of an extended professionalization construct.
Figure 1: A Strategic Assessment of “Familiness” and Competitive Advantage. Source: Habbershon & Williams (1999)
The differences between family firms and non-family firms thus result from their familiness (Craig
& Moores, 2006a; Kellermanns, Eddleston, Sarathy, & Murphy, 2012). Examples of familiness are abundant. Duran et al. (2016), in a meta-analysis, find that family firm owners typically have a larger share of their wealth invested in the firm than owners of non-family firms. This influences their decision making and investment strategies (Duran et al., 2016). Furthermore, they argue that family firms have non-financial goals that are more valued than in non-family firms. Stewart &
Hitt (2012) provided a meta-analysis on the differences between family and non-family firms in order to explain what differentiates both groups. The authors find that differences exist in;
ownership, governance, returns, rewards, networks, leadership, careers and management (Stewart
& Hitt, 2012). Family firms typically are less formal, profit driven and performance based (Stewart
& Hitt, 2012). Moreover, family firms tend to focus more on staying competitive in the long term, compared to a short-term vision of non-family firms (Kraiczy et al., 2015). Overall, this results in that family firms are seen as weaker and that they are often urged to professionalize (Stewart &
Hitt, 2012).
2.2 Professionalization
Professionalization in general is often mentioned in relation to the life-cycle model (Flamholtz &
Randle, 2012; Zahra & Filatotchev, 2004). Flamholtz (1986), in his earlier work, described professionalization as a stage that follows after the entrepreneurial stage. Later on Flamholtz described it as a growth phase following expansion (Flamholtz & Randle, 2009). The bottom line here is, that at some point in the life stage model, firms will encounter problems that relate to size, complexity or age (Fiegener, Brown, Dreux, & Dennis, 2000; Yildirim-Öktem & Üsdiken, 2010).
During this the professionalization process will start. Flamholtz and Randle (2012) describe
10 reasons to professionalize that result from ‘growing pains’, such as increased complexity and size.
Besides that the firm becomes too complex to manage, reasons include; dealing with complex environments, dealing with institutional forces, accessing external finance, or increasing insights and expertise (Stewart & Hitt, 2012; Yildirim-Öktem & Üsdiken, 2010).
Chua, Chrisman and Sharma (1999) wrote at the end of the 20th century that the choice to professionalize is one of the two most important strategic decisions that a family firm faces. This is still relevant today, and professionalization remains important far after the choice to professionalise is made. PWC (2017) found in the US Family Business Survey that forty percent of the (US) family businesses indicated that professionalization will continue to pose a challenge in the upcoming years. Every family firm at some point has to professionalize (PWC, 2017). This is due to the fact that family firms face unique challenges when advancing in the life-cycle model compared to other firms (Chua et al., 1999; Fiegener et al., 2000).
Stewart and Hitt (2012) explain that the underlying reason for these challenges arise from familiness. The different characteristics were discussed in section 2.1.2. Moreover, the authors explain that familial threats, such as favouring family members and an informal approach, increases the unprofessional approach in family firms (Stewart and Hitt, 2012). It is therefore that a professionalization construct applied to the family firm context specifically is worthwhile. Dekker (2012) argues that in the family firm context professionalization should include aspects of adding non-family members, next to aspects such as; effectiveness and control, delegation of decision making and formal recruitment.
Over the years the literature has agreed on some aspects regarding what professionalization is for
family firms specifically. Professionalization in the family firm context is often associated with the
adding of non-family members to the management or the board of a firm (Dyer, 1989; Yildirim-
Öktem & Üsdiken, 2010; Zahra & Filatotchev, 2004). But this is not all, according to some authors
in the field. As discussed earlier, Dekker et al. (2013), Hall and Nordqvist (2008), and Stewart and
Hitt (2012) believe that the firm can be professionalized on an organizational level by
formalization. Moreover, professionalization is seen a continuous process, with the purpose of
improving (Stewart & Hitt, 2012). Professionalization is therefore about making less mistakes and
becoming better in what you do. Based on the literature on professionalization I believe that
professionalization should be a multidimensional construct that aims to continuously improve a
11 (family) firm, by professionalizing its management, organizational capabilities and strategic decision making. This includes the notions by Stewart and Hitt (2012) and Dekker (2012) that it should be multidimensional. It also adds a strategic aspect as described by Hofer and Charan (1984), who include stable corporate strategies in their definition of professionalization.
In this thesis, I therefore use the recent articles of Dekker et al. (2013;2015) and her doctoral thesis (Dekker, 2012) as principal and build on this construct of professionalization. I take this perspective as it is a multidimensional construct, that has been empirically tested. As discussed in the introduction, Dekker (2012) developed five factors to measure professionalization. For this Dekker (2012) researched the data of 532 Flemish family firms. From the literature Dekker identified 24 variables that measure the level of professionalization in a family firm. With these variables a factor analysis was conducted. The following factors were found in the analysis: (1) financial control systems; (2) nonfamily involvement in governance systems; (3) human resource control systems;
(4) decentralization of authority and (5) top level activeness. Appendix 2 provides a complete overview of the variables used by Dekker (2012).
Factor 1 and 3 measure the internal formalization of the firm. Internal formalization refers to implementing systems into the organization that allow for control (Dekker, 2012). The goal of this dimension is to make the organization more objective and data driven. The family aspects (i.e.
familiness) will have less influence on daily practices. In a professionalized firm; recruitment, training, rewards and control are based on facts and figures. The remaining three constructs (2, 4 and 5) measure the effective openness of the firm. ‘Openness’ in this refers to the capacity of the family firm to integrate external managers into their organization (Dekker, 2012). Effective openness is about the family allowing external managers to participate in the firm, in such a manner that the non-family and family members are enabled to effectively govern the firm (Dekker, 2012).
Top level activeness measures the frequency of meetings of the management and advisory boards and decentralization is focused on limiting the role of the CEO.
2.3 Innovation Orientation and Innovativeness 2.3.2 Innovation in family firms
The innovativeness of family firms has been an ambiguous resource topic (Duran et al., 2016).
Family firms have long been criticized to be conservative and less innovative than their non-family
counterparts (e.g. Habbershon et al., 2003; Kraus et al., 2012). However, recently the
12 innovativeness of family firms has been increasingly recognized and praised (Stenholm et al., 2016). In a meta-analysis of 108 primary articles, Duran et al. (2016) conclude that family firms in general have a higher innovation conversion rate. Meaning that family firms on average invest less in innovation but realize relatively higher output than non-family firms.
Agency theory, albeit not being the primary perspective of this research, offers explanations for the underinvestment that family firm literature observed. The first reason is that the owner’s wealth is in the firm and that therefore the owners are more careful when investing (Duran et al., 2016;
Miller, Le Breton-Miller, & Lester, 2011). Secondly, family firms are less likely to attract external capital to invest in innovation due to the fear of losing control of the firm (Duran et al., 2016). The higher innovation output that is now recognized in the literature, according to the agency theory, stems from less reliance on formal control and coordination mechanisms (Kraus, Pohjola, et al., 2012). This results for example in faster decision making for short and middle range decisions (Roessl, Fink, & Kraus, 2010). This indicates that more control and coordination mechanisms could hinder innovation.
The resource based view and dynamic capabilities theory offer similar explanations for higher innovation conversion in family firms. Craig, Dibrell and Garrett (2014) found that capabilities and resources that are unique to the family firm (i.e. the familiness) lead to the innovativeness of family firms. This happens trough three mechanisms according to the authors; family influence, family culture and flexible planning (Craig et al., 2014). The authors empirically validated that family influence is positive for the culture of the firm. This will enable the firm to respond better to changes, as they develop strategic and flexible capabilities, and this leads to innovativeness (Craig et al., 2014).
Family firms have characteristics that are directly linked to innovation. Family firms have the
willingness to pro-actively control the innovation process and the ability to develop a network in
the long term (Duran et al., 2016). In their meta-analysis, the authors find that family members are
more committed to the firm (Duran et al., 2016). This can, for example, be seen in long term
employment (Duran et al., 2016). Family firms, therefore benefit more from their human capital
than their non-family counterparts. In additon to this, Craig et al., (2014) describe that the typical
long tenures of family members contributes to higher tacit knowledge in the firm. De Massis,
Frattini and Lichtenthaler (2013), found that family firms are characterized by low formalization,
13 high decisional autonomy and being functionally organized. These are characteristics that are related to innovativeness (Damanpour, 1991; Jansen, Van Den Bosch, & Volberda, 2006). Overall, family firms thus have capabilities that seem to improve the innovativeness of the firm.
2.3.3. Innovativeness
Innovativeness is defined as: “ …the introduction of new processes, products, or ideas in the organization” (Hult, Hurley, & Knight, 2004). This means that innovativeness measures the outcome of the capability to develop new products or markets (Wang & Ahmed, 2004). Because this research is on the firm level, the outcomes of innovativeness need to reflect this as well (Stock
& Zacharias, 2011). This concepts thus concentrates on the outcomes of innovation across products, and measures what being innovative adds to an organization. Innovativeness has resemblances to other terms like innovation power or innovation outcomes. Innovativeness, however, has the advantage of using self-reporting measurements (Hurt, Joseph, & Cook, 1977).
This approach is useful as it fits to the article of Siguaw et al. (2006), where the innovation outcomes are measured as; form, type and rate. Innovativeness includes different types of innovations, such as radical and incremental and service or product innovations (Siguaw et al., 2006). When measuring innovation at an organizational level it is common to refer to innovativeness (Damanpour, 1991; Stock & Zacharias, 2011).
2.3.4. Innovation Orientation
Innovation orientation is one step before innovativeness. When taking a broad perspective, innovation orientation focusses on the tacit characteristics of a firm and refers to the intentions of the firm. Innovation orientation includes the antecedents of innovativeness (Hurley & Hult, 1998).
It is defined by (Siguaw et al., 2006) as:
“A multidimensional knowledge structure composed of a learning philosophy, strategic direction, and trans-functional beliefs that, in turn, guide and direct all organizational strategies and actions, including those embedded in the formal and informal systems, behaviours, competencies, and processes of the firm to promote innovative thinking and facilitate successful development, evolution, and execution of innovations” (p7)
An innovation orientation is embedded in the organizational culture and can be seen as consisting
of underlying success factors of innovativeness (Stock & Zacharias, 2011). Having a strong
14 innovation orientation includes that firms devote sufficient resources to areas of the firm that encourage the creation, development and implementation of innovations (Siguaw et al., 2006).
From the literature several characteristics repeatedly are mentioned for creating an organization that is directed towards innovation. Firms oriented to innovation are found to be proactively and constantly looking for new technologies and services (Wiklund & Shepherd, 2005). In addition, a strong vision towards innovation that is shared in different layers of the organization is often mentioned (E.g. PWC, 2017; Flören and Berent-Braun, 2017). High innovation orientation firms are likely to implement formal and informal policies, procedures and practices that empower employees to innovate and learn (Siguaw et al., 2006; Tidd & Bessant, 2013). Overall, the organizational culture is recognized to be focussed on innovation (Siguaw et al., 2006). In a large survey from PWC, a culture supportive of innovation is mentioned as the most important antecedent of innovation (PWC, 2017). This together constitutes an innovation orientation.
Stenholm et al. (2016), found that an innovation orientation is related positively to growth in family firms specifically, and not in non-family firms. The authors believe that this is because family firms are more focussed on long term growth. Because of this they focus on innovation and renewal, rather than short term profitability (Stenholm et al., 2016). This makes it an interesting concept to research for the purpose of this research.
Innovation orientation is often discussed in relation to a learning orientation (Calantone, Cavusgil,
& Zhao, 2002) or an entrepreneurial orientation (Wiklund & Shepherd, 2005). I chose to focus on innovation orientation as it fits most closely to the scope of the thesis. Moreover, innovation orientations is an extension of the dynamic capabilities theory, that concentrates on the creation of organizational competencies (Siguaw et al., 2006). I however take a broad approach to the construct and add risk taking and continuous learning to the concept. Risk taking and continuous learning are crucial for firm success and innovation (Calantone et al., 2002; Wiklund & Shepherd, 2005).
Taking Innovation Orientation construct completely apart from these characteristics would not provide a complete picture in my opinion. These factors are important for innovation and have been used in relation to innovativeness by other authors (e.g. Wang & Ahmed, 2004).
To summarize, innovation orientation provides many of the aspects that seem to be missing in the
professionalization construct of Dekker (2012). As Siguaw et al. (2006) state; innovation
orientation is a multidisciplinary construct that is not limited to narrow constructs, what is
15 essential for identifying and managing innovation patterns. The items that are identified in the literature are listed in table 1.
Table 1. Innovation Orientation Items
Item Relevant papers
Innovation Orientation
Resource allocation Siguaw et al. 2006, Wang & Ahmed, 2004
Shared vision Siguaw et al. 2006, PWC 2017, Flören & Berent-Braun, 2017 Outside-network Stewart & Hitt 2012, Flören & Berent-Braun, 2017 Innovative culture PWC, 2017, Hurley & Hult, 1998, Siguaw et al. 2006 Open-mindedness Calantone, Cavusgil & Zhao 2002, Wang & Ahmed 2004 Continuous improvement Wang & Ahmed 2004, Siguaw et al. 2006
Supportive culture Siguaw et al. 2006, PWC 2017 Entrepreneurial- and Learning- Orientation
Pro-activeness Wiklund & Shephard 2005, Wang & Ahmed 2004
Risk taking Wiklund & Shephard, 2005, Hurley & Hult 1998, Wang & Ahmed, 2004 Continuous learning Calantone, Cavusgil & Zhao 2002
2.4 Hypothesis development
2.4.1 The effects of professionalization on innovativeness
To research the effect of professionalization on the innovativeness of family firms both dimensions of Dekker (2012) are explored. Craig and Moores (2006) found that innovative family firms are less formal and decentralized. By professionalization these characteristics disappear. For example, the effects of being long-term oriented decrease when the firm adopts formal evaluation and budgeting systems. Duran et al., (2016) in a meta-analysis found that control mechanisms are negative for innovation. Kraiczy et al., (2015) state that control mechanisms, formalization and standardization might be negative for innovation. The flexibility of the firm will decrease when the formalization increases (Craig & Moores, 2006b) Another example is that rewards systems are found to have negative effects on innovation (Lawson & Samson, 2001). The dynamic capability theory suggests that formalization decreases the innovativeness of firms (Damanpour, 1991).
Therefore;
Hypothesis 1a: Internal formalization will have a negative effect on the innovativeness of
family firms.
16 Effective openness suggests that a firm has the capacity to allow external managers to participate in the firm in an effective manner (Dekker, 2012). Based on the literature review effective openness is expected to negatively relate to innovativeness. This is because familiness, according to recent literature, has positive effects on innovativeness. By effective openness the familiness of the firm will decrease. The first reason for this is that family members are more committed to the innovation outputs of their organization than external managers (Duran et al., 2016). A second reason for this is that family managers are recognized to develop long term relations with the stakeholders of the firm (Le Breton-Miller, Miller, & Bares, 2015). In order to develop healthy long term relations with their stakeholders, family firms are willing to invest in innovation (Le Breton-Miller et al., 2015). In addition to this, Gudmundson, Tower, and Hartman (2003) found that family ownership is positively associated with the introduction of new ideas and the implementation of innovation.
An underlying reason for this is that communicating between hierarchical levels decreases the flow of innovative ideas (Damanpour, 1991). In the literature review it was discussed that family firms benefit from long tenures of managers and high levels of decisional autonomy (e.g. Kraus, Pohjola, et al., 2012) and that family managers benefit from faster decision making. From a dynamic capabilities perspective, faster decision making and flexibility is known to be positive for innovation (Teece et al., 1997). Therefore, the hypothesis offered is;
Hypothesis 1b: Effective openness will have a negative effect on the innovativeness of family firms.
2.4.2 Innovation and performance
From the literature review it can be concluded that innovation orientation is a knowledge construct, and a precondition for an organization to be innovative. Having high IO means that an organization is able to effectively respond to changes in the market (Stenholm et al., 2016). In the literature there is agreement that innovation can be managed, as an integrated and continuous process, and that competitive success is dependent on this (Adams, Bessant, & Phelps, 2006; Craig & Moores, 2006b). This relates to continuous renewal, and as suggested by Habbershon and Williams (1999) this might lead to competitive advantages (see figure 1).
When innovation orientation is strong a firm has the right resources and capabilities in place that
are embedded in formal and informal processes (Siguaw et al., 2006). According to the DCV, these
capabilities are needed to reinvent the firm and stay competitive in the long term (David J Teece,
17 Rumelt, Dosi, & Winter, 1994). In firms with a high IO every member of the organization recognizes the importance to innovate and the attitude is favourable to innovative. It is therefore expected that innovation orientation will have a positive effect on performance. Thus,
Hypothesis 2: Innovation orientation is positively related to family firm performance
The last part of the research is to empirically test if an Extended Professionalization construct will have more explanatory power compared to the current professionalization construct. The current construct only explained around 6 percent of the variance of the performance variable (Dekker et al. 2015). For testing this hypothesis an extended professionalization construct that consists of three dimensions; Innovation orientation (IO), internal formalization (IF) and effective openness (EO) is researched.
The reasons for this are that there are likely more dimensions to the construct (Dekker, 2012), and innovation has not been taken into account so far. Moreover, some authors argued that the construct should be continuous (Hwang & Powell, 2009; Stewart & Hitt, 2012) and have a strategic nature (Hofer & Charan, 1984). By adding Innovation Orientation as a part of professionalization the construct becomes more operational and relevant, and more in line with the DCV. IO also adds to professionalization by providing a mean of introducing governance and strategic aspects as well as a continuous dimension. Having a strong IO helps to overcome the weaknesses family firms have in innovativeness. By this they become more like non-family firms, as is the purpose of professionalization according to Stewart and Hitt (2012). Family firms differ in; ownership, governance, returns, rewards, networks, leadership, careers and management. IO therefore will contribute to the part of professionalization that seems to be missing based on the literature. I therefore expect that this construct will have higher explanatory power.
The hypothesis offered is,
Hypothesis 3: An enhanced professionalization (with IO, IF & EO) construct will have higher explanatory power of firm performance compared to the current
professionalization construct
2.4.3 Conceptual Model
In the preceding sections the different concepts for this thesis are identified and defined. The first
part of the research focusses on the relationship between the professionalization dimensions and
18 innovativeness. The professionalization construct as developed by Dekker (2012) is used. I look at the two dimensions introduced by Dekker (2012) for hypothesis H1a and H1b. Innovativeness is used as defined by Hult et al. (2004) as the dependent variable for H1a and H11b. Throughout the thesis I have repeatedly referred to the concept of family firm performance, this is the dependent variable of hypothesis 2 and 3. Family firm performance in this thesis is viewed in terms of market growth, employee growth and financial performance. For the last part of the thesis the concept of innovation orientation is introduced (Siguaw et al., 2006). This is added as an extension and tested in hypothesis 3, shown in the dotted rectangle. The measurement of the concepts is discussed in the section 3.2.
Figure 2: Conceptual model
3. Method
3.1 Data Collection Method
A common approach for family firm research is to do convenience sampling as there are limited data bases regarding this subject readily available (Chua et al., 1999). A combination of several search strategies was therefore employed. The first search method was to subtract a sample of firms from the Bureau van Dijk database with a known website. I manually checked the websites to see if the firm was likely to be a family firm or not and if the website provided contact details.
Complementary to this I looked for family firms in my own network and the network of my
supervisor. Another method was searching on LexisNexis and Google for websites and news sites
with references to family firms. This resulted in a mailing list of 323, likely family owned,
businesses. To these firms a first mail and a reminder were sent. The mail asked if they were a
family firm, and if yes, if they wanted to fill out the questionnaire.
19 Because this was a small number an additional 2119 firms from the Bureau van Dijk database were mailed and asked if they were a family firm. To make a selection within the data-base, only firms with 5-250 employees were selected. This group most likely is in the process of professionalizing, and similar selection criteria are used in similar research (e.g. Dekker et al., 2013). All mails were send in a period of five weeks in December 2017 and January 2018. To these firms no remainder mail was send.
This resulted in a sample of 113 completed questionnaires. From this sample three respondents were deleted because they were either no family firm or they had less than five employees. This resulted in a total sample of 110 firms.
3.2 Measurements and scale development
In this section I will shortly describe how the different concepts are measured. For an extensive overview of the developed questionnaire see appendix 4.
3.2.1 Descriptive variables
For the descriptive part the questionnaire of Dekker (2012) is used as a basis for the questions.
Descriptive variables are asked on the firm level and the individual level. Three questions validate if a company can be classified as a family firm or not. The first question asks the respondent if 50%
of the shares are held by the family. The second question asked whether the respondent considers the firm to be a family firm and the last questions if there is intention to transfer the firm to the next generation. Firms should fulfil two out of these conditions to be considered as a family firm.
Another question asks if the CEO is a family member, and if yes, from what generation he/she is.
This is done as generational effects can influence the relationships between professionalization and innovation (e.g. Duran et al. 2016). Descriptive variables on the individual level are questioned regarding his/her level of education, age and tenure. At last the respondent is asked to identify the life-stage of the firm. This can be the; starting, growing, maturity or decline phase (Masurel & Van Montfort, 2006). Also, two questions related to professionalization were asked; if the professionalization is considered important (DV7) and if the firm is considered professionalized (DV8). These two variables are only used for descriptive purposes.
Professionalization (Prof)
For the professionalization part the questionnaire of Dekker (2012) is used. This is useful as it
allows to further test the construct of professionalization. Moreover, for the purpose of
20 generalizability this adds to the relevance of the outcomes. The questionnaire is however adapted in several ways
2. Questions are transformed from a binary question into 5-item Likert scales where possible to make the items of a continuous nature. The questions for Utility of Business plan (IF1) and Education Level (EO3) from Dekker (2012) are deleted as they are irrelevant for this research or because they were already deleted in the doctoral thesis of Dekker (2012). Formalized goals and objectives and Use of budgets, Budget Evaluation System and Firm Performance Evaluation from Dekker are merged into two questions to shorten the length of the questionnaire. The first two questions are merged to Formal budgeting (IF1) and the second two to Formal evaluation (IF2) (see appendix 2 & 3). These questions are the same factor in Dekker (2012) and therefore are assumed to measure the same variable. The non-Likert scale questions regarding effective openness are standardized. This is to account for different measurement formats and to improve the comparison process, this approach is similar to Smith et al. (1994). For the other items there were no problems with different measurements formats. The Cronbach’s alpha score for professionalization is 0.714. For this all items for IF and EO are included with the EO variables being standardized. The value of 0.714 is above the threshold of 0.7 (Tavakol & Dennick, 2011).
Innovation Orientation (IO)
In the second part of the survey several questions regarding the measurement of innovation orientation and innovativeness are asked. These questions are shuffled to reduce respondent bias.
A complete measurement of the concepts was not readily available and therefore questions from different sources were combined. In this thesis innovation orientation aims to measure the factors present that stimulate innovation. I therefore looked for existing questions that measure such factors. If a certain firm scores high for several of these factors, this gives an idea of the level of innovative orientation. A formative perspective is taken for this. I use questions that are related to the concepts that are identified in Table 1. Existing questionnaires from Wang and Ahmed (2004) and Vermeulen, de Jong and O'Schaughnessy (2003) are used to base the questions on.
This resulted in 10 5-item Likert scale questions related to innovation orientation. This is validated with a factor analysis (See appendix 5). This resulted in five factors, three with IO related items, one with performance and one with innovativeness items. In the factor analysis two items were deleted, open-mindedness and innovation culture, as their factor scores are below the threshold of
2From Dekker (2012) IF1 and EO3 deleted, IF 3 & 4 and IF 5 & 6 merged
21 0.5 (Hair et al., 1998). The Cronbach’s alpha of all IO items is 0.671, which is close to the 0.7 threshold (Tavakol & Dennick, 2011).
Innovativeness (INN)
Innovativeness aims to measure the outcomes of innovation. The questions are based on Calantone et al. (2002) and Keskin (2006). I looked for questions that relate to the speed, type and frequency of innovation as discussed by Siguaw et al. (2006). To reduce the respondent-bias the respondent is asked to compare innovativeness to competitors and take a customer perspective. This resulted in five Likert scale items (appendix 3). One additional question is asked that takes into account a 5 year period to control for the effect that investments for innovation often only pay out after several years (question INN06) (e.g. Craig & Moores, 2006a). In the factor analysis (appendix 5), innovation turnover, innovation flexibility and innovation incremental (reversed) are deleted because they are under the 0.5 threshold of (Hair et al. 1998). The Cronbach’s alpha of the remaining items is 0.777.
Family Firm performance (Perf)
Family Firm performance is measured based on three questions. Two subjective questions ask the respondent to compare their financial success and market share to their competitors. Comparing with competitors is important according to Wiklund and Shepherd (2005). Dess and Robinson (1984) state that subjective measurement by the top management team can be considered adequate, especially for privately held firms. However objective measures are preferred. The last question is therefore of an objective nature and measures the percentage of growth or decline of employees.
This is based on the CBS (2016) method of measuring firm growth and has been previously used in research (e.g. Klomp & Van Leeuwen, 2001). For performance the alpha is 0.613. This does not increase by deleting other indicator variables. Scores of 0.6-0.7 can be seen as questionable (George & Mallery, 2003). For exploratory research this is however considered sufficient
3. Also a Cronbach’s alpha is not very meaningful for items with a small number of indicators (Tavakol &
Dennick, 2011).
3.2.2 Control variables
Questions regarding firm size, firm age and industry are asked for the purpose of control (Dekker, 2012; Kraiczy et al., 2015). The pre-coded industry control variable is based on De Jong and Vermeulen (2006). For Age and Size, the natural log is taken to reduce the skewness. Firm age is
3 Also factor analysis shows that performance variables are closely related (see appendix 5).
22 also used as a selection variable. This is to be able to exclude firms that are younger than 5 years from the sample, as these firms are unlikely to already be professionalizing.
3.3 Plan of analysis
After cleaning and preparing the data set the first step is to do an analysis of the descriptive variables. The second step is to look for multi-collinearity in a correlations matrix (Hair et al., 1998). Three dummy variables for manufacturing, service and wholesale/retail are added. This is based on the industry question. The industries are divided on manufacturing, service and wholesale/retail and other, as is used as in Dekker (2012). The ‘other’ category is not used as control variable. A first test employed is the Harman’s one-factor test, to test if common method bias is present. Six factors explained 64.48% of the variance. The first factor explains 27% of the variance, indicating that common method bias is not problematic for this study (Podsakoff &
Organ, 1986).
For professionalization the unweighted average of all Internal Formalization and Effective Openness items are used. For Innovation Orientation the unweighted average is calculated for all items, except for Open-mindedness and innovation culture. This is based on the factor analysis performed (Appendix 5) For Innovativeness the unweighted average is calculated for all indicators except Innovation Incremental, Innovation Flexibility and Innovation Turnover. For Performance the unweighted average is calculated for all three items. As the scores can now be seen as interval data hierarchical regression analysis is possible.
The next step is to test for the assumptions that have to be fulfilled to do regression analysis. All
constructs are tested for non-normality or heteroscedasticity. Testing for non-normality resulted in
that the outliers for LogSize, LogAge and Professionalization are deleted. All constructs were found
to be normally distributed with no problems of heteroscedasticity. For all variables the highest
Skewness is 0,322 and the highest Kurtosis value is -0,329. This meets the requirements of Hair et
al. (1998). A last step is to calculate the Professionalization Extended variable. For this the
unweighted mean of the dimensions (Internal Formalization, Effective Openness and Innovation
Orientation) is calculated. The mean is taken, as for every firm professionalization is different
(Stewart & Hitt, 2012). This means that a firm can get a high score on, for example, two dimensions
and still be considered professionalized.
23
4. Results
4.1 Descriptive Variables
The descriptive statistics are depicted in table 2
4. The characteristics of the individuals in the data set show that the CEO was on average 48 years old and 17 years in charge of the firm. The firm- level descriptive indicators illustrate that on average the firms employed 101 (>15 hours) people.
After deleting the outliers identified by SPSS, the largest firm in the sample had 1100 employees, with a standard deviation of 134. The firms on average are 74 years old, with the oldest firm being 234 years old. When comparing this to the results of Dekker (2012) it can be concluded that the sample in this thesis contains larger and older firms. The average size is 101 employees compared to 29 and the firm age is 73 years compared to 27. The age and the tenure of the CEO’s are quite similar to Dekker (2012). The age is 48 years compared to 47, and the tenure was 17 years compared to 16. The descriptive statistics further show that 94,5 percent of the firms agreed or fully agreed that professionalization in their organization is important. Moreover, 90.9 percent of the family firms indicated that they regarded themselves as being professionalized. No significant differences in mean were found for the early and late respondents by using a T-test for a split sample. For performance, innovativeness and professionalization, all significance levels are above 0.50.
Table 2. Descriptive variables CEO Age
(years)
CEO Tenure (years)
Firm Size (employees)
Firm Age (years)
Mean 48 17 101 73
Std. Deviation 10 12 134 36
Minimum 27 1 5 18
Maximum 77 50 1100 234
4 Extreme values (outliers) have been removed for Firm Size and Firm Age
24 For the industry the descriptive statistics are summarized in table 3. The Manufacturing category is represented the most (40%). This is also seen in other family firm research and the Manufacturing sector is known to have a relatively large share of family firms (CBS 2017; Flören, 2017).
Table 3. Descriptive statistics industry
Manufacturing Service Wholesale Other Total
Frequency 44 15 29 22 110
% 40 13,6 26,4 20 100
Descriptive variables regarding the performance indicators are summarized in Table 4. For this the sample was split in low and high professionalized firms. For this the sample is split in two equal groups based on the professionalization score. Table 4 shows the means of both groups. Family firms that were more professionalized were more innovative and had higher performance. There was no significant difference between the mean age of both groups. For size there was a significant and large difference.
The correlation matrix (Table 5) shows the correlations of the most important variables used in the regression analysis. It can be concluded that there is a high correlation between INN and IO (0,568).
These two items are however not tested in the same models. Furthermore, there is a high correlation between Size and Internal formalization (0,491) and Size and Effective Openness (0,491). This can be interpreted as the larger the firm is, the more the firm is professionalized. None of the correlations are above 0,6, indicating that multi-collinearity is not a problem (Hair et al., 1998).
Table 4. Group Statistics (N51-55 per group)
High Sig Low Sig
Prof 0,51 *** -0,48 ***
Perf 3,72 * 3,56 *
IO 4,08 *** 3,81 ***
INN 3,60 ** 3,32 **
LogSize 4,64 *** 3,58 ***
LogAge 4,18 4,18
a. Significance levels ***0,01, **0,05, *0,1
25 Table 5. Correlation Matrix
Mean
Std.
Deviation Perf INN IO IF EO LogSize LogAge Manufacturing
Wholesale
& Retail
Perf 3,63 0,52
INN 3,46 0,63 ,201*
IO 3,93 0,41 ,344** ,452**
IF 3,50 0,78 ,325** ,216* ,515**
EO 0,50 0,40 ,202* 0,16 ,248** ,458**
LogSize 4,07 1,10 ,355** ,306** ,333** ,491** ,453**
LogAge 4,17 0,51 -0,11 -,245* -0,04 0,04 -0,17 0,00
Manufacturing 0,40 0,49 0,02 -0,01 0,01 ,236* 0,13 ,224* 0,07 Wholesale &
Retail
0,26 0,44 0,07 -0,14 -0,16 -0,08 -0,14 -0,18 0,00 -,489**
Service 0,14 0,34 -0,05 0,00 0,11 -0,08 0,00 0,11 -0,16 -,324** -,238*
*. Correlation is significant at the 0.05 level (2-tailed).
**. Correlation is significant at the 0.01 level (2-tailed).
4.2 Regression Analysis
For further analysis hierarchical regression is performed. The dependent variable for the first models is innovativeness. First the control variables will be included in model 1, and thereafter the main effect variables are included in Model 2 (Table 6). For this the ratio of 15 subjects per predictor requirement is fulfilled (Field, 2000). The Cook’s test found no influential data points that have a value higher than 1 (Field, 2000). For the missing values the Pairwise method is used as this minimalizes the loss of information as is the case by Listwise deletion (Hair et al., 1998). In Pairwise deletion all available data is used (Hair et al., 1998). For model 1 and 2 all variance inflation factor (VIF) values are below 3 (with the highest being 1,936), meaning that multicollinearity is not present. The Durbin-Watson of the first regression analysis is 1,861, what is close to the value of 2. This indicates that autocorrelation is not a problem for this regression.
The total sample for this was 101, as some missing values were deleted.
26 Table 6: Hierarchical Regression Analysis Model 1 & 2 (N=101)
Dependent Variable
Innovativeness Innovativeness
Model 1 Model 2
Std. Beta P-value Std. Beta P-value Control Variables
Constant 0.00 0.00
LogSize 0.35 0.00 0.32 0.01
LogAge -0.26 0.01 -0.28 0.00
Manufacturing -0.27 0.03 -0.30 0.02
Service -0.23 0.05 -0.23 0.05
Wholesale & Retail -0.27 0.03 -0.29 0.02
Independent Variables
Internal Formalization
0.15 0.19
Effective Openness -0.10 0.36
R2 0.21 0.23
Adjusted R2 0.17 0.17
F-Value 5.07 0.00 3.91 0.00
a. Regression coefficients are reported as standardized β
For Internal Formalization the standardized beta was 0.15. From the regression analysis hypothesis 1a cannot be accepted at the 0.05 significance level (p=0.19). The sign of Internal Formalization was positive, indicating that a positive effect is likely to exist between Internal Formalization and innovativeness. There is a non-significant negative effect of Effective Openness on the innovativeness of family firms with a standardized beta of -0.10. at a 0.05 significance level (p=0.358). There is no support for hypothesis 1b.
As the results of the linear regressions are not significant additional analysis is required. I therefore
test for nonlinear effects and the moderation effects for the relationships of Internal Formalization
and Effective Openness with the dependent variable innovativeness. First, I test for non-linear
effects by testing the squared values of IF and EO. From this there was no increase in adjusted R-
squared results. The results of the test for non-linearity show an unstandardized beta of 0.005
27 (p=0.96) for IF and an unstandardized beta of -0.046 (p=0.623) for EO. Second, for testing the moderation effect the interaction term is tested for both constructs. This means IF*EO with standardized values is included in the model. For this test a non-significant result is found. The result is a standardized beta of 0.055 (p=0.560) for the interaction term of IF and EO.
To test hypotheses 2 and 3, additional regression analyses are performed (table 7 & 8). In this model performance is taken as the dependent variable. Model 3 is the base model for Model 4 and 5. The Durbin-Watson of model 3 and 4 is 1,818, so there is no auto-correlation. The highest VIF value is 1,932, meaning that there are no indicators of collinearity problems (Hair et al., 1998).
Table 7: Hierarchical Regression Analysis Model 3 & 4 (N=101)
Dependent Variable
Performance Performance
Model 3 Model 4
Std. Beta P-value Std. Beta P-value Control Variables
Constant 0.00 0.00
LogSize 0.39 0.00 0.30 0.00
LogAge -0.12 0.20 -0.11 0.22
Manufacturing -0.05 0.73 0.00 0.99
Service -0.11 0.35 -0.10 0.38
Wholesale & Retail 0.10 0.43 0.15 0.21
Independent Variables
Innovation Orientation
0.27 0.01
R2 0.17 0.23
Adjusted R2 0.12 0.18
F-Value 3.80 0.00 4.73 0.00
The results of model 4 show a positive and significant relationship between Innovation Orientation
and Performance with a standardized beta of 0.27. Hypothesis 2 can thus be supported at the 0.01
significance level. This means that having an Innovation Orientation does lead to higher
performance for family firms.
28 Table 8: Hierarchical Regression Analysis Model 3 & 5 (N=101)
Dependent Variable
