Sponsorship in the Dutch SME sector

MSc Thesis

MSc Business Administration - Small Business & Entrepreneurship University of Groningen

Sponsorship in the Dutch SME sector

Identification of sponsorship activity

and SME influencers of sponsorship motives.

January 2019

Student: Jaap Toering (s2528347) j.f.toering@student.rug.nl

Wordcount: 18,000

TABLE OF CONTENTS (1) PREFACE

INTRODUCTION

RESEARCH FRAMEWORK 1.1 Sponsorship

1.1.1 A brief history of the sponsorship market 1.1.2 The art of sponsorship in a nutshell

1.1.3 Scientific attention to corporate sponsorship 1.1.4 Objectives and results of sponsorship

1.2 Corporate Social Responsibility

1.2.1 CSR developments and trends

1.2.2 Sponsorship: a perfect example of CSR 1.2.3 CSR motives

1.3 The SME sector

1.3.1 The challenge of researching small firms 1.3.2 Specific SME characteristics influencing CSR

1.4 Hypotheses development 1.4.1 Resource poverty 1.4.2 Owner-manager influence 1.4.3 Business locality 1.5 Research scope RESEARCH METHOD 2.1 Data collection 2.1.1 Independent variables 2.1.2 Dependent variables 2.1.3 Firm-specific characteristics 2.2 Data analysis

2.2.1 Preparing the dataset 2.2.2 Aggregating variables

2.2.3 Illuminating sponsorship activity

2.2.4 The effects of SME specific characteristics 2.2.5 The influence of firm-specific characteristic

2.3 Quality criteria

TABLE OF CONTENTS (2)

RESEARCH FINDINGS 3.1 Preparing the dataset 3.2 Aggregating variables 3.2.1 Variability testing 3.2.2 Reliability testing 3.3 Descriptive statistics 3.3.1 Data sample 3.3.2 Correlations 3.3.3 Sponsorship activity

3.4 Influencers of sponsorship activity

3.4.2 Effects of firm-specific characteristics

3.5 SME characteristics

3.5.1 Direct effects of SME characteristics 3.5.2 Moderating effect of resource poverty

DISCUSSION 5.1 Implications

5.2 Limitations and future research CONCLUSION

ACKNOWLEDGEMENT REFERENCES

APPENDIXES

Appendix 1 - Research objectives Appendix 2 - Survey structure Appendix 3 - Contact list

A. Preparing dataset B. Aggregating variables

C. Factor analyses & reliability test results C. Descriptive statistics

PREFACE

INTRODUCTION

In the year 2018, 70 percent of all small Dutch firms sponsored a yearly contribution to at least one local sports or culturally related entity. Together, they have generated a total value of €1 billion worth of sponsorship value (Panteia, 2019). Significantly more than the government, which invested a total value of €126 million in sports and another €802 million in the cultural sector. Unsurprisingly, perhaps, in a year of economic prosperity and in a decade in which Corporate Social Responsibility (CSR) strategies have become the rule rather than the exception. Still, the small and medium-sized enterprise (SME) sector appears to be needed now more than ever. Together it constitutes for 99% of all businesses in the EU, 66% of all employment and 50% total added value (Lepoutre & Heene, 2005). These numbers keep increasing as shifts in economic structures force large firms to scale down or to exchange places with small firms, better able to cope with rapidly changing business environments (Webb & Carter, 2001).

On the sponsorship sector, these trends have their impacts too. Sponsorship literature, however, primarily focuses on large firms, as it is the large firms that draw most media attention (Spais & Johnston, 2014). Unfortunate, given the accumulated social and economical contribution small firms provide to society (Fuller, 2003) and the potential of sponsorship as marketing instrument for small firms (Gardner & Shuman, 1988). Therewithal, the fast majority of literature emphasizes the objectives and outcomes of sponsorship. No attention seems to be granted to firms’ internal motive to sponsor (Slåtten _{​et al.,}​2016). This research therefore contributes to the sponsorship literature by initiating a stream of literature to fill part of the void that currently exists regarding sponsorship motives and SME literature.

Based on related CSR literature, a framework was constructed that hypothesizes stimulating effects of three SME characteristics (resource poverty, business locality and owner-manager influence) on three distinct sponsorship motives (philanthropic, proactive commercial and responsive commercial). Information was derived from a sample size of 69 micro, small and medium-sized firms, who participated in a survey. Simultaneously, this research serves another objective: to illuminate sponsorship activity, by deriving any additional relevant information regarding sponsorship activity, and its relation to sponsorship motives, and firm-specific and SME characteristics.

RESEARCH FRAMEWORK 1.1 Sponsorship

Sponsorship is a concept well-known to many. Yet, opinions and definitions regarding its true roots, application and implementation tend to be contradicting. An introduction of its heritage, the academic history and a decisive determination of the research framework is therefore required, in order to understand the scope of this study. This is the goal of chapter one. It first introduces developments and trends of sponsorship. Additionally, it explains how sponsorship collaborations come about and introduces the main actors and various types of sponsorship occurrences. Furthermore, chapter one introduces the scientific attention drawn to the research domain, particularly discussing its definition, relevance and the literature gaps that exist in the field of sponsorship motives and sponsorship research within the SME sector. An overview of the research scope, the variables and objective is depicted in appendix 1.

1.1.1 A brief history of the sponsorship market

Ancient works of the Greek philosopher Xenophon prove the origin of sponsorship to date back at least two millenia. Xenophon explained how in many of the ancient Hellenic civilizations, sports, culture and arts played considerable roles in society. In order to remain such cultural events accessible for all societal layers, wealthy aristocrats regularly funded individual athletes, actors or events out of philanthropic motives. In return, they received appreciation, popularity and fame from their communities. For the largest portion of time since then, sponsorship endeavours have barely developed in terms of occurrence frequency and financial value.

The revival of the Olympic Games in 1896 appeared to be the starting signal for an accelerated development of the global sponsorship market (Kissoudi, 2005). Particularly the great sponsorship success of the Games’ 1936 edition in Berlin initiated a period of significant growth. For that growth, two important drivers have been responsible. First, increasingly larger events have required increasingly larger budgets, boosting the urge for sponsors. Second, as the events grew in size, they started to attract larger groups of spectators, increasing their commercial potential for investors (McKelvey & Grady, 2008).

Over time, sponsorship has become increasingly commercial. In fact, corporate sponsorship has become the world’s fastest growing marketing communication strategy (Meenaghan, 2001; Bal _{​et al.,}​2010; Groza _{​et al.}​, 2012; Delia & Armstrong, 2015). This is represented by the market figures, which indicated accumulated annual expenditures in sports, arts, entertainment, community causes and events of approximately U.S. $50 billion in the year 2012 (IEG, 2012). For many firms, sponsorship can no longer be unseen from their marketing strategies.

1.1.2 The art of sponsorship in a nutshell

In every sponsorship agreement, there are at least two parties involved: the sponsor (those who provides support) on the one hand and the sponsee (those who receive support) on the other. The type of support may vary from either financial contributions to any sort of supply of goods, services, knowledge, or access to relations. Although the sponsor-role can be fulfilled by any type of entity, in practise it is mostly firms that engage in sponsorship endeavors. Still, this does not hold per se and therefore there is a distinct term referring to those situations in which the sponsor-role is fulfilled by firms: _{​corporate sponsorship. Note that corporate sponsorship distinguishes itself} from corporate philanthropy, donations and patronage, by its commercial attitude (Polonsky & Speed, 2001).

In practise, the sponsees tend to be entities, organizations or events that stand-alone would not be profitable. Therefore, they regularly rely on support from sponsors and/or donors. Although the sponsees do not per se generate financial value from their operations, they do play an important role for society according to many. This is one of the reasons there are sponsors willing to invest. Examples of sponsees can be found particularly in the sports, arts, cultural and music sectors, of which the sports sector holds the eminent largest share (Madill & O’Reilly, 2010; Plewa & Quester, 2011).

Although there are only two parties directly involved in sponsorship agreements, sponsorships are aimed to affect a wider variety of publics. Those publics include external audiences (e.g. current and potential consumers of a sponsor’s products or services, financial institutions and community leaders; Gardner & Shuman, 1988) as well as internal audiences (e.g. employees and partners; Khan & Burton, 2014) and particularly those that have strong liking for the sponsee (Crimmins & Horn, 1996). The positive influence a sponsorship collaboration might have on those third parties, provides another reason why firms invest in sponsees.

1.1.3 Scientific attention to corporate sponsorship

Europe’s cultural variety has led to a wide range of attitudes towards sponsorship and its role in society. Resultantly, ambiguity exists among scholars regarding the theoretical definition of sponsorship. The absence of one commonly accepted definition of sponsorship is extensively addressed in all of the three comprehensive reviews.

One of the researchers, Björn Walliser (2003), makes a bid at defining sponsorship worth mentioning, based on all literature published before the year 2003. He finds that regardless the many contradicting perspectives, there are two basic elements to which most scholar seem to agree. Based on these findings, he formulates the following definition (p.8-9), which will be applied in this study.

‘Sponsorship (1) is based on an exchange between sponsor and sponsee, and

(2) pursues marketing (communication) objectives by exploiting the association between the two.’

This definition comprises two important elements. First, it encompasses the two parties involved in any sponsorship agreement, as described above. Second, it clearly distinguishes sponsorship from corporate philanthropy, donations or patronage in a sense that sponsorship must always involve at least some extent of commercial motives (Polonsky & Speed, 2001). In other words: in sponsoring, the sponsee will always be expected to provide at least some sort of compensation that contributes to the sponsors strategic marketing objectives. On the contrary, Walliser’s definition purposely leaves room for wide interpretations of sponsorships as it does not include any other restrictions. According to the definition, for example, sponsorship might include any sort of sponsorship (in finance, goods, services, knowledge, or relations) is included. Applying Walliser’s definition of sponsorship enables this study to remain a wide scope, serving its explorative purpose.

1.1.4 Objectives and results of sponsorship

Regardless their differences, sponsorships and advertising can be applied complementary. Additional advertisement expenditure spent to promote the sponsorship agreement beyond the boundaries of the initial agreement is referred to as _{​sponsorship-linked marketing or ​sponsorship leverage (Cornwell, 1995). Some scholars argue that the} effectiveness of sponsorships is only optimized when fully embedded in integrated marketing strategies (Quester and Thompson, 2001) and leveraged successfully (Khan & Burton, 2016). The benefits of sponsorship leverage are, however, restricted (Meenaghan & Shipley, 1999). Over exaggeration of a firm’s social affection might conflict with the sponsorship’s true philanthropic nature and might therefore result in critical attitudes towards the sponsorship agreement and eventually will affect the sponsor’s reputation negatively, rather than positively.

There are many more examples of positive results derived from sponsorship as a strategic instrument. Besides increased brand awareness, firm image, goodwill and recall recognition of the sponsor (Rifon et al. ​, 2004), amplified brand loyalty (Sirgy _{​et al.,}​2008), for example, and customers’ intentions to purchase sponsor’s products (Ngan _​et al.,​2011) have been recognized. Also, the general positive impacts on return on investment (Harvey, 2001) through additional achievement of media exposure (Crompton, 2004), and even enhanced attitudes towards the sponsee (Bhat, 2012) have been identified in the literature. Other research (Gardner & Shuman, 1988) underlines the importance of quantification of expected result. The theory explains that assessment procedures may be challenging, as sponsorships may involve hidden costs and hard to measure benefits, but are crucial in the estimation of sponsorship effectiveness as a strategic tool.

1.1.5 Sponsorship motives

Although there has been a significant body of literature focussed on the customer perspective of corporate sponsorship, sponsorship from the firm perspective appears to be neglected, say Slåtten _{​et al.}​(2017). In fact, they claim to be the first to investigate or identify the main factors that motivate firms to engage in sponsorship, which leaves the research domain of sponsorship motives relatively untouched. Unfortunately, their qualitative approach and findings retrieved from eight Norwegian semi-large firms, do not sufficiently represent the research scope and approach of this study. The research conducted by Slåtten _{​et al.}​(2017) does, however, provide some leads that serve as departure point for this study. For example, the authors identify two “fundamental pairs of contrastive orientations related to sponsorship motives” (p.143): _​internal ​versus _{​external ​motives} ​and _{​opportunistic} ​versus altruistic motives​, Once assembled, the two orientations form a construct of four sponsorship motivational categories, dubbed the _{​Sponsorship Motive Matrix} ​(SMM).

motivation and commitment to the firm among employees, cooperative partners, stakeholders and others” (p.148). And, even though the internal focus of sponsorship might indeed by highly relevant (also illustrated by Khan & Burton, 2016), the examples of both internal and external focus as presented by Slåtten _{​et al. seem to relate more to} internal and external _{​objectives rather than to internal and external ​motives} ​of sponsorship. This leaves the opportunistic versus altruistic dimension as the only classification of sponsorship motives. Still, a more thorough understanding of sponsorship motives is required. Since it appears that the sponsorship research domain abstains when it comes to motives, it might therefore be wise to broaden the research perspective.

1.2 Corporate Social Responsibility

In order to gain deeper understanding of the concept of _{​sponsorship - and particularly of its niches regarding} sponsorship motives​and SME sponsorship - ​it is useful to place sponsorship in broader perspective. There has been a number of scholars that did likewise and recognized sponsorship as fraction or example of Corporate Social Responsibility ​(CSR). Since the research domain of CSR shares significant overlap regarding both practical implications and theoretical development with sponsorship domain, it provides handhold on the vacancies found in sponsorship literature. This chapter therefore introduces CSR’s origin and developments, explains its relationship to sponsorship, and emphasizes particularly its amalgamation of altruistic and opportunistic nature.

1.2.1 CSR developments and trends

The topic of Corporate Social Responsibility drew first academic attention in the late 1950s, early 60s. It was at that period when the many countries recovered from the damage done by WWII. Enterprises took on sizes that had never been seen before. Subsequently, their impact on society increased significantly. The first CSR works (e.g. that of Davis, 1960; Eells and Walton, 1961) responded to that increasinging impact. They questioned the responsibilities of businesses, which comprised at that time only earning profits. Their efforts induced instalment of legislation and several governmental bodies in the 70s, which primarily served the interests of employee protection. Later, other stakeholders such as the community and environment followed. In the successive decades, increased attention of business press, governments, ventures and political leaders towards firms’ ethical behavior have stimulated interests of the public in (non)ethical business practises (Campbell, 2007). Under increasing transparency and media attention ever since, firms have increasingly been forced to comply to jurisdiction and ethical standards.

The commercialization trend has motivated some authors to incorporate the commercial aspect of CSR in their definition (i.e. Kolk & Pinkse, 2006) and others to obelize philanthropy in its totality (i.e. Moon, 2001; Rollinson, 2002), arguing that altruism in the end always comes down to self-preservation. However, the discussion whether or not philanthropy is driven by self-interests or not, does not deny that in principle there are non-commercial motives that drive CSR engagement. What this discussion does prove, is that the perceptions of CSR develop over time and strongly depend on subjectivity. Resultantly, no consensus can be found on one true definition of CSR until today.

1.2.2 Sponsorship: a perfect example of CSR

In response to the absence of one true definition, Archie B. Carroll (1991) has published a renowned work, called The pyramid of CSR, in which he attempts to explain CSR by distincting four layers or firms’ responsibilities: economic, legal, ethical, and_{​philanthropic. Carroll’s theory argues that all firms have the economic responsibilities} of providing goods and services to the community primary to any other responsibility. Equally important to these economic responsibilities come the firm's’ responsibility to obey the law, which embodies the basic notions of fair operations. Third come the firm’s ethical responsibilities to perform ‘those activities and practices that are expected by societal members even though they are not codified into law’ (p.41) and fourth come the firms’ philanthropic responsibilities which reflect those actions that are not expected in any moral/ethical sense, but do contribute to society’s expectations of being good corporate citizens. In an ideal world, all firms comply to all corporate responsibilities. In practice, the pyramid is build bottom-up. Firms comply first to the first two layers and only later they add the third and fourth (Carroll, 1991).

In Carroll’s theory, the term philanthropy refers to those social activities that are performed discretional/voluntary, meaning they were not requested by society, fully disregarding whether or not the gestures are compensated. This terminology might be entangling, given the definition provided before, since Carroll includes contributions of money, facilities or employee time to humanitarian causes as arts, education and the community among the term corporate philanthropy. Both sponsorship and corporate donorship are thus included in the fourth layer and Carroll’s definition of philanthropic responsibility..

1.2.3 CSR motives

Keeping in mind that CSR literature might not necessarily dictate sponsorship literature, but does provide insight, it seems wise to consult the CSR research domain in order to better understand the various sponsorship motives. Fortunately, in the CSR literature, more has been published regarding corporate motives. A two-step of classifications can be identified: the first separates philanthropic from commercial motives, the second subdivides the commercial motive in either responsive or proactive commercialism.

Philanthropic versus commercial motives

The first fundamental orientations among CSR motives is similar to that identified by Slåtten _{​et al. (2017) and} separates _{​philanthropic} ​from _{​commercial motives}​. This orientation consists of purely altruistic driven motives on one end of the spectrum, and utterly commercial incentives on the other. Placed in chronological perspective, CSR is rooted in primarily altruistic motives. The commercial motives arose only later and increased over time. Firms engaging in CSR are never driven by either one of the extremes and will always be motivated by an amalgamation of both.

Responsive versus proactive commercialism

The commercial motive, at its turn, can be subdivided into two main streams. Hemingway and Maclagan (2004) define these two commercial motive streams as being either _{​responsive}​and_​proactive​social responsible behavior_​. The responsive motive refers to those social responsible behavior exhibited in response to external pressure - or threats (Baron, 2001) - on the firm’s reputation. Reputational threats, such as media attention due to integrity, ethics, or environment related violations, could potentially damage the firm’s share values, its market share and thus its profitability. Corporate social responsible behavior in response to that, could prevent such harm (Graafland _{​et al}​., 2004). Responsively motivated CSR strategies thus comprise those actions taken to prevent damage that would not have been taken in absence of external pressure.

1.3 The SME sector

Visibility from outside the organizations appears to be a primary motive for responsible behaviour (Bowen, 2000). In this line of reasoning, it would be those who experience large scale public attention who engage most in socially responsible behaviour, including sponsorship. At first sight, these would be large firms, as they generally face the largest groups of stakeholders. Conventional views on CSR therefore suggest that small firms engage in CSR to a lesser extent than large firms. Besides, small firms are ought to cope with resource constraints that prohibit them from responsible behaviour. More recently, however, contradicting perspectives have found other characteristics that drive socially responsible behaviour among small firms, leaving the research field in ambiguity (Lepoutre & Heene, 2006). This paragraph explains the ambiguity by providing a brief summary of sponsorship and CSR literature in the SME sector, elaborating extensively on those SME characteristics found to influence social responsible behaviour.

1.3.1 The challenge of researching small firms

Sponsorships are proven to have potential value for the promotions mix of small companies, as - for small firms in particular - they can be a cost-effective alternative for expensive traditional marketing practices (Gardner & Shuman, 1988). Yet, only few small firms seem to engage in sponsorships. Why this is the case, is uncertain. Research on sponsorship in the SME sector is extremely limited, say Khan and Burton (2016), as ‘the prominence of large entities means that sponsorships by large organisations attract most media and research attention’ (p.25). Dreadful, argue Webb and Carter (2001), as shifts in the industrial infrastructure decrease the number of large firms, increasing the relevance of the SME sector. The role of SME research also increases as large firms search for organizational structures better able to cope with rapidly changing business environments.

In their study _{​Sponsorship Activities in the Small Firm Sector,} ​Webb and Carter (2001, p.170) add that ‘the promotional activities employed by large firms operating in geographically dispersed markets, serving a large customer base and selling directly to the consumer are inappropriate for the majority of small firms. Effective communications for most small firms entails direct communication with a target market that is small in number, narrowly based and spatially restricted’. The issue that small firms cannot be regarded as ‘little big firms’ has been recognized before (by i.a. Dandridge, 1979; Welsh & White, 1981). The SME sector characterizes itself by facing other internal and external dynamics that influence their decision-making processes differently compared to large firms. And, even within the ‘SME sector’ variety is so wide and dispersed, that investigating the SME sector requires a distinct research approach (Tilley, 2000)

First, the SME sector may not be ignored as it constitutes 99% of all businesses, 66% of all employment and 50% total added value in the EU. Second, large firms are becoming increasingly entrepreneurial, so research on small, more entrepreneurial firms may yield valuable insights for large firms too. Third, CSR strategies designed for large firms may not work for small firms as they face a number of specific characteristics, which have an impact on small businesses’ CSR practices.

1.3.2 Specific SME characteristics influencing CSR

The fact that there are SME-wide characteristics that influence CSR practises within small firms, appears to be reasonably accepted. However, opinions regarding how those characteristics influence the CSR practices seem contradicting. Some praise small firms for being socially responsible by nature, by providing welfare through creating jobs, stimulating economic growth and creating innovation, particularly in local communities. Others argue that small firms are limited by resource constraints in such an extent, that they cannot afford to invest in CSR.

Jan Lepoutre and Aimé Heene (2006), contribute significantly to the research field of CSR within SMEs by critically reviewing all prior available theoretical and empirical contributions on the size-social responsibility relationship among small businesses. In their study, the authors recognize the fragmented research field and attempt to develop a coherent theory on Small Firm Social Responsibility (SFSR). In doing so, they have classified the contingent factors of small business behaviour into four dimensions: issue related, personal, organizational and context characteristics. Throughout these four dimensions, three commonly returning factors influencing SBSR can be identified, being: resource poverty, business owner-manager influence​, and external stakeholders pressures​. According to the authors, those are the three factors that are imperative in the role CSR plays within a small firm.

Disunity prevails when it comes to the influence of _{​resource poverty on CSR practises. Two streams can be} identified. The conventional stream considers CSR to be an additional business practise and thus a cost burden. Since small firms face resource constraints more than large firms (Nooteboom, 1994), small firms are ought to invest significantly less in CSR practises than large firms. The second, more recent stream of literature argues that the same resource constraints in fact stimulate small firms to engage in CSR through enhanced innovativeness and responsiveness (Jenkins, 2006). Different opinions seem to depend primarily on how CSR practises are perceived: once perceived as additional act of kindness, CSR might be a cost-burden. When perceived as essential fraction of business practises, CSR might form a potential source of competitive advantage (Tilley, 2003).

Besides the work of Lepoutre and Heene, studies by professor Heledd Jenkins (2006; 2009) provide large contribution to the research field of SBSR. The goal of her 2006 study, for example, was to progress understanding of both the limitations on and opportunities for CSR in SMEs through the exploration of exemplary characteristics in the study companies. According to Jenkins ‘there are difficulties associated with CSR in SMEs and resources will always be limited. However, rather than seeing difficulties as a barrier, they could be approached as a challenge that needs to be overcome through innovation’ (p. 252). Jenkins’ perspective thus constitutes to the statement made earlier, which suggests the role of SME characteristics depend on how CSR is perceived. Consequently, Jenkins lists a set of SME characteristics that can aid the adaptation of CSR. Two underlying SME factors seem imperative: the relatively high degree _{​of creativity/innovativeness}​ and _{​owner-manager influence.}

SME characteristics in sponsorship literature

The influence of SME characteristics on sponsorship motives has not been addressed explicitly in the sponsorship literature. The SME sector has not been studied extensively either. Still, there is some sponsorship research that does seem to recognize similar characteristics unconsciously. Khan and Burton (2016), for example, investigate the internal effects of sponsorship on small firms. According to their study, large firms might have more resources to better leverage sponsorships both internally and externally, but small firms enjoy enhanced internal communication due to their relatively low size and quicker communication.

Russo and Perini (2010) relate to the external stakeholder pressure. They find that SMEs have a more direct connection with stakeholders in the direct local community, as they tend to operate more locally. Resultantly, SME employees appear to have stronger affiliations with the sponsored entities, as they are more closely related. Koch (2013) adds that since having a good reputation is of paramount importance to competitiveness, small businesses would naturally engage in local CSR practises, such as sponsorships, so that customers can put a face to the business. Gardner and Shuman (1988) also find that small firms that sponsor locally may find they can counter misgivings and foster goodwill, in order to decrease community hostility towards a sponsor’s past actions, or help counter negative publicity.

1.4 Hypotheses development

1.4.1 Resource poverty

The role of resource constraints on CSR appears to be ambiguous. Some argue that the lack of resources restrict small firms from investing in CSR activities, whereas others provocate that the limited accessibility of resources forces small firms to think and act more innovative than large firms (Jenkins, 2006). The role resource poverty plays on CSR investment depends on how CSR is perceived. Once perceived as being an additional act of kindness, resource constraints might hinder CSR engagement. In this sense, resource poverty would restrict firms from investing in sponsorships too. The absence of resources, on the contrary, might stimulate innovative thinking. And, since sponsorship has the perk of being a very effective and a potentially inexpensive marketing tool when leveraged right, sponsorships might be a source of competitive advantage. In that sense, resource constraints might positively affect the proactive commercial motives of sponsorship. From the latter statement, the following hypothesis can be derived:

H1: Resource poverty has a positive effect on the proactive commercial motive to sponsor.

Arguing from the conventional perspective and the resource-based view (Barney, 1991), resource poverty will also show negative impact on the relationship between sponsorship motives and actual sponsorship activity. In this way it argues that even though firms have all the incentives to sponsor, their actual sponsorship activity will be limited by the limits of their resources. Therefore, the following hypothesis can be formulated:

H5: Resource poverty has a negative moderating effect on the relationship between sponsorship motives and sponsorship activity.

This hypothesis is labeled ‘hypothesis 5’, as it will be tested separately from the other hypotheses, since it is the only hypothesis that incorporates a moderating effect.

Innovativeness

The theory of Jenkins (2006) explains that, forced by resource constraints and stimulated by effective communications, small firms appear to be more creative and innovative in their business practices than large firms. The high degree of innovativeness would imply small firms to be more creative and thus proactively driven in their sponsorship endeavors. This assumption underlines sponsorship theory presented by Gardner and Shuman (1988) that argues sponsorship to be a creative and cost-effective alternative for traditional marketing channels, especially useful for small firms. However, as the high degree of innovativeness is a direct effect of resource constraints, it is assumed that it does not have an effect on the sponsorship motives. Instead, it is the underlying characteristic of resource poverty that affects the proactive commercial sponsorship motive. This, at its turn, is already reflected in the first hypothesis. Therefore, innovativeness is not included in the hypotheses as a separate variable.

1.4.2 Owner-manager influence

The smaller the company, the larger the influence of its owner. Not only on formal strategy and decision-making; also on business culture. The more influence the owner-manager has on its firms, the more its personal norms and values influence business practices. In the case of CSR, the business-owner’s personal affection with particular entities or topics such as the environment, or charities therefore affect CSR endeavors. In the context of sponsorship, personal values of owner-managers might result from personal experiences, affection, relationships, or direct contact with sponsorship-dependent entities. Therefore, the higher the owner’s influence on its business, the more likely it is that the firm is philanthropically motivated to engage in sponsorship. The following hypothesis can be formulated.

H2: The degree of owner-manager influence has a positive effect on the philanthropic motive to sponsor. The influence of an owner’s personality traits also affects the entrepreneurial mindset of the firm. The more influence an owner-manager has in its business, the more likely it is that he can pursue creative marketing strategies. This research therefore expects the effect of business-owner influence to also stimulates the firm’s proactive commercial sponsorship motive. The following hypothesis was formulated accordingly.

H3: The degree of owner-manager influence has a positive effect on the proactive commercial motive to sponsor. 1.4.3 Business locality

When it comes to the sponsorship context, results were found specifically regarding small firms facing local stakeholders: small businesses operate significantly more local than large firms (Russo & Perini, 2010) and therefore a good local reputation is of imperative important (Koch, 2013). Those that deem their local reputation important, will subsequently invest more in there local reputation. And, since local sponsorship is an effective way of improving local reputation and to counter misgivings and foster goodwill and decrease hostility towards a sponsor’s past actions (Gardner & Shuman, 1988), small firms might feel they have to engage more in sponsorship agreements than large firms. Based on that assumption, the fourth hypothesis assumes a positive effect of business locality on the responsive sponsorship motive.

H4: The degree of business locality has a positive effect on the responsive commercial motive to sponsor.

1.5 Research scope

Figure 1: Visualization of the research scope versus that of current sponsorship research.

Figure 2: Visualization of the variables and their expected influences on sponsorship motives and activity.

RESEARCH METHOD

The objective of this study is twofold. First, it investigates the effect of three SME characteristics on three predefined sponsorship motives derived from CSR literature. Second, it attempts to illuminate any information on sponsorship activity in the Dutch SME sector. The second objective also includes the effects of various firm-specific characteristics on the sponsorship motives. Objective 1 constitutes the core of this analysis. For that a theory testing approach, as proposed by Van Aken, Berends and Van der Bij (2012), is applied. Van Aken _{​et al.}​suggest a five step approach: (1) identification of definitions and the literature gap, (2) generation of conceptual model and hypotheses, (3) data collection, (4) data analysis, and (5) interpretation of results with the comparison to hypotheses, reaching conclusions and implications. The first two steps have already been taken. This methodology section proceeds by explaining how the data collection is executed, how the data analysis will be conducted, and how the study’s quality is ensured.

2.1 Data collection

Data required to achieve the objectives is gathered through a survey, distributed among a total of 1079 Dutch SMEs, selected through theoretical sampling. This research includes firms only that comply to the SME definition as determined by the European Commision (2003). It includes only firms consisting of less than 250 employees, which have a turnover of a maximum of €50 million. Additionally, all firms that are not owner-managed are excluded (Spence, 1999).

The data collection process consists of two phases. In the first round of data collection, recipients were retrieved from a secondary database that has access to public data provided by the Dutch Chamber of Commerce, called Orbis. Orbis also provided the firm’s contact details. The recipients were contacted through email. The survey was sent to a total of 972 firms, which were selected based on firm size in terms of employees and revenue. Within a timeframe of four weeks, 22 recipients responded (2.3% response rate). Orbis’ data appeared not to fully up-to-date: three firms exceeded the revenue limits and another one consisted of more than 250 employees. Those firms were excluded, leaving a total of 18 respondents.

and asked to participate in the survey. The survey was thereafter sent to a total of 82 firms, of which 62 completed the survey (75.6% response rate). The total sample size amounts 70 respondents (N=70).

Besides increased reliability through reduced risk of respondents biases, the theoretical sampling applied in both rounds of data collection grants the advantage of increasing the variety of respondents. The Orbis dataset differentiates itself as - in general - its firms are larger and vary from a more dispersed range of industries. Additionally, it includes firms that do not engage in sponsorships. This is something which provides interesting insights, but could not be achieved by the method applied in the second data collection round. Contrarily, the second round of data collection grants the opportunity to target firms that have a strong pronounced opinion regarding sponsorship. Additionally, it enables the researcher to contact firms that sponsors specific sectors, which again allows for purposely stimulating the variety of participants.

Survey

The survey was prepared following Emans’ (2004) interview guide and constructed and spread by means of Qualtrics. First, four underlying concepts were identified from theory: _{​SME characteristics, firm characteristics,} sponsorship activity and_{​sponsorship motives}​. Based on the four concepts, a total of 15 variables were derived. From the variables, a list of in total 19 questions was formulated. The vast majority of questions aims to retrieve interval data on a 1-100 scale. The remaining questions generate ratio and categorical data. An extensive overview of the four concepts and all associated variables is depicted below in figure 4, below. The following sections illustrate how the questions related to the independent, dependent and control variables were formulated.

2.1.1 Independent variables

Three independent variables were included in the analysis, derived from theory proposed by Lepoutre and Heene (2006), who initially proposed resource poverty, business owner-manager influence and external stakeholder pressures as imperative determinants of CSR behavior within SMEs. The latter variable was transferred to a new variable, labeled _{​business locality}​, based on Gardner and Shuman’s (1987) theory, in order to fit the sponsorship context. Figure 5 displays which questions were formulated to determine the independent variables.

Figure 5: Fraction of the survey guide, illustrating how the variables and questions measuring SME characteristics have been formulated.

2.1.2 Dependent variables

Coherent to two superior concepts, a total of six dependent variables were measured. The first concept analyses sponsorship activity​, in order to answer the first objective of this research: to uncover the current state of sponsorship among Dutch SMEs. Sponsorship activity is measured through three variables: ​sponsorship investment, including the number of sponsees and the firm’s total sponsorship value, ​sponsorship type,​including the sponsee’s sector and organizational type, and​sponsorship engagement,​reflecting the degree to which the sponsoring firm is actively concerned with its sponsees. An overview is provided below in figure 6.

In the initial stage of data analysis, several variables are recoded into new variables, among which sponsorship investment relative to firm revenue. This new variable is also included as indicator of sponsorship investment. The third concept,_{​sponsorship motives, is aimed at achieving the second objective of this thesis and is broken-down into} three dependent variables to do so: _{​philanthropy, proactive commercialism}​and_{​responsive commercialism, derived} from CSR literature, including Hemingway and Maclagan (2004) and Hoevenagel (2004). How these three variables are measured is depicted in figure 7 below.

Figure 7: Fraction of the survey guide, illustrating how the variables and questions measuring sponsorship motives have been formulated.

2.1.3 Firm-specific characteristics

A total of six firm-specific characteristics is included, retrieved from contributions of Spence (1999), Webb and Carter (2001), Lepoutre and Heene (2006) and Khan and Burton (2016). The control variables measure firm-specific characteristics:​firm age, industry sector, ownership, firm size, firm life cycle​and​customer type.​The latter refers to the percentage of B2C customers related to the percentage of B2B customers. Questions related to the firm-specific characteristics are depicted in figure 8.

2.2 Data analysis

The survey gathers quantitative data, analyzed by means of SPSS. The analysis section of the study is structured into two steps, in coherence with the the sub objectives of this study. However, before the analyses can be performed, the dataset must be prepared. It includes screening the data for missing or erroneous data and labelling, recoding and merging variables. Every section of paragraph 2 of the methodology sections covers one of the steps and explains how they are conducted. An extensive overview of every step can be found in the analysis logbook (appendix 2).

2.2.1 Preparing the dataset

Before any analyses can be performed, the dataset will be prepared. After the results are downloaded from Qualtrics into SPSS. First, variables will be named, labels will be added, and new variables will be created or recoded when necessary. Then, several cases will be excluded, when not meeting the requirement criteria as determined in the data collection section above. Thereafter, the dataset will be scanned for potential erroneous data or outliers. When needed, additional variables will be added once the open entry boxes (otherwise, namely: ... ),, indicate additional widely supported answers.

2.2.2 Aggregating variables

Every variable is measured in 4-8 questions. Before the variables can be applied in analyses, the questions must be aggregated. Before this can be done, however, the questions are to be tested for unidimensionality (to verify no other constructs are measured by the questions) and reliability (to determine the degree in which the questions measure the construct that was planned to be measured). For the unidimensionality test of each variable, a Principal Component Factor Analysis (factor analysis) will be applied. Before the factor analysis can be performed, however, a Bartlett’s test of Sphericity and a Kaiser-Meyer-Olkin Measure of Sampling Adequacy (KMO) are performed to test whether the samples are appropriate and adequate for factor analyses. Once the the factor analyses are performed, a reliability test is performed, applying a Cronbach’s alpha threshold of >0.6, preferably >0.7 (Nunnally, 1978).

In order to test hypothesis 5, another round of aggregation is required to combine the three sponsorship motives into one overarching variable: _{​sponsorship motive}​. This is a formative construct, as the three underlying concepts measure different antecedents of sponsorship motives. Therefore, no reliability test is required. In aggregating the sponsorship motives, all underlying variables will be weighted equally.

2.2.3 Illuminating sponsorship activity

The facts and figures are derived from the descriptive statistics and visualize sponsorship behaviour regarding sponsorship value, type, sector, number of collaborations, etc. An analyzation of their results visualizes what is currently happening in the sponsorship market. Figure 9 visualizes the research scope of step 1.

The second step concerns effects of external factors on sponsorship activity (figure 10). It includes the exploration for potential correlations between sponsorship motives, SME characteristics and firm-specific characteristics and the various sponsorship activity variables. The latter step will be challenging, as it deals with multiple types of variables (figure 11). Multiple types of analyses are therefore required: a regression analysis is performed to test regressions between the sponsorship motives and sponsorship activity and engagement. To better understand the effects of firm size, age and industry, a matrix is constructed manually to assess their relations to the various sponsorship motives.

2.2.4 The effects of SME specific characteristics

The second part of the data analysis section is dedicated to hypotheses testing. In total, five hypotheses are tested. Hypotheses 1 through 4 all hypothesize correlations between SME characteristics and sponsorship motives (figure 12). To test these estimated correlations, results from the Pearson correlation matrix are consulted and four regression analyses are performed. The fifth hypothesis predicts a negative moderating effect of resource poverty on the positive relationship between sponsorship motives and sponsorship activities that was tested in the regression analysis. A moderator analysis is applied to test hypothesis 5. The scope of hypothesis 5 is visualized in figure 13.

2.2.5 The influence of firm-specific characteristic

Figure 9: Scope of objective 2.1: Analyzing facts and figures regarding sponsorship activity and coherent variables.

Figure 10: Scope of objective 2.2: Discovering what motivates firms to engage in sponsorship activity?

Figure 11: Objective 1.2 deals with multiple variable types and therefore requires various analyses.

Figure 12 Visualisation of hypotheses h1 through h4.

2.3 Quality criteria

In order to draw reliable and credible conclusions that apply for a wide public, the whole research process, including literature study, data collection, data analysis, conclusions and discussions must meet particular quality criteria. Researchers Van Aken et al. (2012) list controllability, reliability and validity to be the three most important research quality criteria. Given the importance of quality for this research, a separate section is specifically dedicated to the quality standards. This section explains how these three criteria are warranted in this study.

2.3.1 Controllability

The term controllability implies that your research can be replicated by other researchers. Controllability of this study is secured primarily through the extensive description of the data collection and data analysis processes in the previous methodology sections, in the research findings section and in its associated appendixes, which includes among others: the survey structure (appendix 2), contactlist (appendix 3) and the data analysis logbook (appendix 4A). Contact databases are kept confidential for privacy reasons. Additionally, utmost emphasis was put on proper referencing in the literature sections.

2.3.2 Reliability

Four potential sources of biases are listed by Van Aken et al. (2012). Those biases might obstruct reliability and hinder a research in its independency of its particular characteristics: the researcher, the instruments, the respondents and the situation. In order to decrease researcher related biases, this study has followed the five research process steps as proposed by Aken et al. (2012) and followed every step closely. Two rounds of data collection were performed in order to create a database that consists of a wide variety of respondents, in terms of firm size, age and industry. This attempts to decrease risk of respondents biases. Additionally, the second round of data collection includes firms that sponsor in various sectors. More than just sports, which appears to be dominating. In order to reduce situation-biases, attempts are made to include firms from more regions than Groningen. Although the majority of firms originates from the north of The Netherlands and primarily Groningen, firms from more provinces, including - but not restricting - Friesland, Drenthe, Overijssel and Utrecht, are included.

2.3.3 Validity

RESEARCH FINDINGS

The research findings cover the analyses performed in answer to the research objectives and present their results. The chapter is separated into five paragraphs, in accordance to the research steps as presented in the research method. First, an explanation is provided on how the dataset was prepared and how the survey questions have been aggregated to construct the variables. Thereafter, descriptive statistics are discussed in order to grasp understanding of the dataset under scrutiny. The descriptive statistics also serve in the first step of solving the first research objective, of which further analyses are discussed in the third paragraph. Subsequently, findings regarding the second research objective, including answers to the hypotheses are elaborated.

3.1 Preparing the dataset

Before any analysis can be performed, the dataset must first by prepared. The preparation process includes the selection of cases, preparation of variables and the computing and aggregation of new variables. As the aggregation process requires extensive analyses, it is covered separately, in paragraph 3.2. The first three steps are discussed independently. An extensive overview of all steps and coherent information can be found in the data analysis logbook, attached in appendix 4A.

The raw dataset as derived from Qualtrics, contained a total of 125 registered responses. A significant number of the respondents, however, did not complete the full questionnaire. Therefore, all surveys with a progress of <80 percent were excluded. A total of 76 responses remained. Of those responses, all firms that did not meet the qualification criteria were excluded: three firms of more than 50 employees and one firm that exceeded €50 million revenue. Furthermore, two outliers (sponsorship expenditure of >50% of revenue) and one test case were excluded as they detrimentally influence the sample, leaving a residual of N=69 responses. Any additional information provided by Qualtrics, including contact names, email addresses, response time and coordinates were excluded.

All variables have been named and labeled. Data types have been updated and decimals were reduced to 0. For all variables, the measure were indicated properly. Additionally, values were added to the variables or translated from Dutch to English, in cases of ordinal and nominal data.

3.2 Aggregating variables

This research contains eight interval variables that are all measured in a range of 4-8 questions (see paragraph 2.1 for details). In order to perform the analyses required in later stages, all underlying questions must be aggregated into their superior variable. Before that, tests must be performed to identify whether the questions measure the same underlying construct (and no other constructs) and to what degree they measure what was intended to measure. For that, variability and reliability tests are performed. Results of each analysis are discussed separately.

3.2.1 Variability testing

The variability test consists of three stages: the factor analyses and two checks performed in advance to test whether or not the sample is appropriate and adequate for factor analyses: the Bartlett’s Test of Sphericity and the Kaiser-Meyer-Olkin Measure of Sampling Adequacy (KMO). Three questions were first converted to new variables as their results purposely measured the exact opposite of the intended outcomes. For example: the question ‘to what extent would you describe your firm as a start-up’, measures exactly the opposite of its overarching variable ‘firm life cycle’. Outcomes of both tests are depicted in table 1.

The approximate of Chi-square differs among the variables ranging from 48.1 to 227.32. The degrees of freedom (ranging 6-28) differ, depending on the number of questions involved. All variables are tested significant at 0.001 level of significance, implying the variables are appropriate for factor analyses. The KMO threshold was set at 0.5. All but one variable (firm life cycle: 0.384) exceed the threshold. For firm life cycle, it implies there is weak correlation among the questions. The number of questions (4) might be the cause of this problem. According to the theory, a factor analysis is not an appropriate technique to perform. Just to be sure, the variable is still included in further analyses. For the rest of the variables, Factor Analysis is considered as an appropriate technique for further analysis..

Principal Component Factor Analysis

Table 2: Results of KMO Measure of Adequacy and Bartlett’s Test of Sphericity.

Table 3: Most relevant conclusions of the variability and reliability tests.

Three questions are excluded: Related variable

- Indicate the degree of your firm's ambition to grow. Firm life cycle - I could invest more in sponsorship. Sponsorship engagement - I sponsor in response to bad publicity. Responsive commercial motive

For the residual five variables, all of their questions remain included: - Resource poverty

- Business locality - Owner-manager influence - Philanthropic motive - Proactive commercial motive

Table 4: Results of the Reliability test (Cronbach’s α) of all variables including no. questions.

Variable Type Reliability No. questions

Firm life cycle CV 0.519 3

Resource poverty IV 0.686 5

Business locality IV 0.766 5

Owner-manager influence IV 0.879 5

Sponsorship engagement DV 0.609 4

Philanthropic motive DV 0.730 8

3.2.2 Reliability testing

After the PCA, all potential valid scenarios of every construct were tested for reliability based on Cronbach’s Alpha. Outcomes of α >0.9 are considered excellent, 0.8-0.9 scores are considered good, 0.8-0.7 are considered acceptable, and 0.7-0.6 scores questionable. Results are depicted in the same appendix (4B). The appendix also includes an extensive overview of every step taken. The most important conclusions are listed in table 3.

Based on the outcomes of both the PCA and the reliability test in combination with the literature relevance, all eight variables have been aggregated. Table 4 displays an overview of the eight variables, their reliability scores and the total number of questions included. As can be derived from the table, four out of eight variables score either acceptable or good (green), two variables score questionable (yellow) and another two variables score just below the threshold (red). Unsurprisingly, based on their KMO score, firm life cycle is one of the variables that scores below the threshold. However, the limited reliability scores of firm life cycle and sponsorship engagement do not pose serious reliability threats, as they play limited roles in the further analyses. That of resource poverty and the responsive commercial motive, however, are considerable limitations in terms of reliability. Still, given the fact that the variables score slightly above and only slightly below the threshold, they will be conserved.

3.3 Descriptive statistics

Interpreting the descriptive statistics serves as first step in the analysis, providing insights on the dataset. The following section therefore provides an overview of the most relevant insights derived from the descriptives. First, it discusses statistics regarding the data sample and included firms. Then, it elaborates on correlations between the interval and ratio variables. Third, it dives deeper into the descriptives regarding sponsorship activity. A more extensive overview of all descriptives is added in appendix 5.

3.3.1 Data sample

The survey was directed at the person responsible for sponsorship agreements. In most cases this was the director, the owner, or the owner-director (68.6%). In other cases it were sales, marketing, finance or office managers (12.9%). In 18.5 percent of the cases, it was a secretary or general employee in knowledge of the sponsorship strategies who was able to fill out the survey.

Table 5: Classification of firms based on European Commission (2003) guidelines.

Figure 14: Distribution of firms based on firm age.

3.3.2 Correlations

Table 6 also provides a correlation matrix, which includes an overview of correlations among all ratio and interval variables. Correlations tested at a significance level <0.05 can be recognized by their light green colour. Correlations tested at a <0.01 significance level are indicated in dark green. The most relevant findings are discussed below, discussed per concept.

SME characteristics

The degree of B2C sales negatively correlates (-0.307*) with the degree of business locality, which implies that local businesses tend to do more business with businesses rather than consumers, or that B2B firms tend to operate more locally. This is interesting as firms that score high on business locality (apparently mostly B2B firms) invest significantly more (0.299*) in sponsorship in terms of value relative of their revenue. B2C firms, on the contrary, appear to invest more in sponsorship in terms of number and value.

Sponsorship motives

Regarding sponsorship motives, the strong intercorrelation among the three motives (0.404, 0.405 and 0.389) stands out at first sight. This might indicate an underlying concept influencing all three variables. Potentially, this could be a general sponsorship motive. A principal component matrix (table 7) including the three variables confirms the presence of an underlying factor.

Table 7: Results of a factor analysis proving an underlying general sponsorship motive.

When it comes to correlations among the SME characteristics and the sponsorship motives, the proactive sponsorship motive appears to be most strongly correlated with both resource poverty and business locality, implying that the more a firm operates locally and faces resource constraints, the more it is proactive commercially driven when it comes to sponsorships. The proactive commercial sponsorship motive does not correlate with the degree of owner-manager influence. The responsive commercial sponsorship correlates most strongly with the degree of business locality (0.229) and owner-manager influence (-0.177). However, there is slightly insufficient evidence (sig. = 0.058 and 0.146, respectively) to support these claims. No correlation between resource poverty and the responsive motive can be recognized. Lastly, the philanthropic motive, does show a small potential correlation with business locality, but again this correlation is not significant.

On the output side, stronger and more significant correlations can be recognized: all sponsorship motives correlate strongly to the degree of sponsorship engagement. This implies that the higher a firm’s motive to sponsor, the more engaged it will be to its sponsorship collaborations. Particularly the proactive (0.648**) and the philanthropic (0.514**) motive relate strongly to sponsorship engagement. The responsive commercial motive indicates a strong, but relatively lower score (0.338**). Furthermore, significant correlations between the philanthropic motive and sponsor value (0.258*), the proactive motive and sponsor value/revenue (0.347*) and the responsive motive and no. sponsorships (0.285*) can be recognized. Additionally, two SME characteristics appear to have direct effect: business locality correlates positively to sponsorship value (0.299*) and the owner-manager influence negatively correlates to the number of sponsorships (-0.272*).

Sponsorship activity

Besides their correlations to the sponsorship motives, the sponsorship activity variables correlate strongly among each other: the number of sponsorships increases the total sponsorship value and the sponsorship value/revenue ratio correlates significantly to sponsorship engagement. The latter two variables appear to be proper indicators of the degree in which firms value their sponsorship agreements. Although the absolute value of sponsorship investments and the number of sponsees increase significantly as firm size expands, the relative sponsorship value decreases. Similar holds for older firms, which generally sponsor more different organizations, but sponsor relatively less in terms of revenue share.

3.3.3 Sponsorship activity

In order to gain even better understanding of sponsorship activity in the Dutch SME sector, a more thorough look at the descriptives is required. This paragraph therefore dives deeper into the sponsorship activity indicators and their correlations as detected in paragraph 3.4.2. Also, it investigates the effects of industries that could not be included in the correlation matrix, since it is measures categorical data type.

Sponsorship activity indicators

The numbers regarding the student and charity sectors, however, might potentially obstruct reliability of further analyses as the high degree of students per capita in Groningen might not be representative for the rest of The Netherlands and charity does not comply to the terms as defined in the literature study. Charity is a perfect example of the CSR branche of corporate giving/donations, but unfortunately does not relate to corporate sponsorship. Eliminating the concerning cases from the dataset, however, would reduce the sample size in such an extent that it would harm reliability even further. The cases will therefore remain included.

Since the sports sector is represented most significantly, it might not be surprising that associations hold for the largest share of sponsored entities (58.9%). Foundations and events represent another 20.3 and 16.3 percent. The residual value (4.5%) is invested in other types of organizations, including firms, non-profit organizations (such as theaters, but also research centers) and professional sports teams.

Table 8: Table containing annual figures of revenue, sponsorship value, no. sponsees and related ratios.

3.4 Influencers of sponsorship activity

3.4.1 Effects of SME characteristics and sponsorship motives

In the regression analysis, the three sponsorship motives are included first separately and then together as aggregated variable. The SME characteristics are included separately. Additionally, the effect of percentage B2C sales isincluded. Firm size and age included as categories in the following analysis. The results of the regression analysis are presented in tables 9.1, 9.2 and 9.3 and discussed below.

Table 9.1-9,3: Results of the regression analyses, including independent variables (sponsorship motives, SME characteristics & the degree of B2C sales) and dependent variables (sponsorship engagement & sponsorship value / revenue).

More positive results can be recognized when sponsorship engagement acts as dependent variables. All sponsorship motives appear to have a positive and significant effect (table 1). The proactive commercial motive explains most of its variance (R²=0.420, p<0.001) and 63.7 percent of its slope (β=0.637, p<0.001). The philanthropic motive explains 26.5 percent of the variance (R²=0.265, p<0.001) and predicts 56 percent of the slope (β=0.560, p<0.001). The prediction and effect of the responsive commercial motive is less and less significant (R²=0.114, p<0.001; β=.399, p<0.05). The aggregated sponsorship motive explains 65.7 percent of the variance and 0.957 of the slope (R²=0.657, p<0.001; β=.957, p<0.001). The low difference in R² between the separate and grouped sponsorship motives indicate the relatively large overlap among the various motives. Also, the relatively low significance levels of the effects of responsive motives underlines the relative weakness which was also found in the PCA and reliability test. The firm-specific and SME characteristics do not show any significant effects on sponsorship engagement. Sponsorship engagement thus appears to be the best measure for sponsorship activity, compared to the other sponsorship activity indicators. It is strongly affected by all three sponsor motives. Sponsorship value / revenue indicates only one relation to the sponsorship motives, tested at lower significance.

3.4.2 Effects of firm-specific characteristics

It would be interesting to understand what type of firms endeavor in which sponsorship behavior. Yet, no analysis has been performed so far on the influencers of sponsorship characteristics, including the sponsorship type, sponsee type and sponsee sector. Therefore, this paragraph investigates the effects of firm size and age and industry on the sponsorship characteristics. For that, a table including descriptive statistics was made manually (table 10). To understand the influence of size and age, both variables are grouped based on the EU guidelines and sample size.

The matrix shows that the older and larger a firm grows, the relatively less it spends on sports, the more it invests in music and other sectors. The arts and culture sector seem to fluctuate at each others expense; only a slight increase in expenditure can be recognized between micro and small-sized firms. Young firms appear to spend less of their financial resources on sponsorships and choose to provide services instead. This appears not to depend on firm size. Additionally, young firms sponsor particularly associations and the sports sector. As they grow older, they give more to foundations, charities and music (events).