Culture Matters

(1)

Neshe Tuna

Culture Matters

Analysis of Culture in Sweden and

Finland and Its Influence on Innovation and

Job Performance

Business Administration

Bachelor’s Thesis

15 ECTS

(2)

(3)

Abstract

The present paper aimed to study the culture in Sweden and Finland, by analyzing two research centers of one multinational organization in each location. Hofstede’s cultural framework was used as a benchmark for the research. Further, the study investigated the significant impact that culture has on innovation and job performance. To achieve the purpose, quantitative approach was adopted and a self-competition questionnaire was distributed to the employees of both research centers. Two hypotheses were tested concerning with the relationship between any of Hofstede’s cultural dimensions and innovation, respectively job performance. After a regression analysis was conducted, two models were created which described the impact of the cultural constructs. All of Hofstede’s initial four dimensions indicated a relationship with innovation, while only individualism and uncertainty avoidance showed to be significant in predicting job performance. An unexpected result was discovered in the individualism index, which has drastically shifted, putting Finland in the collectivist societies, with Sweden following the same trend. The study’s results should be able to contribute to the better understanding of culture’s influence on innovation and job performance in multinational organizations and help practitioners by demonstrating the importance of considering culture’s effect.

Keywords: culture, Hofstede, innovation, job performance, cultural dimensions, R&D

(4)

1. Introduction

This chapter presents the background of the study and its purpose. Further, the scope and limitations of the research are discussed, and an overview of the whole thesis is presented.

1.1. Background

Culture has been a field of study for many different disciplines – from Anthropology, Sociology, and Psychology to Economics, Management, and Political science. Culture has been defined by many scholars (e.g. Hofstede 2010; Schwartz & Davis 1981; Kluckhohn and Kroeber 1952; cited in Adler 1997), but being an abstract phenomenon, it is difficult to have a unified concept. Nevertheless, Geert Hofstede’s research into culture gives a simplified and clear description of the construct that has been used as a benchmark for numerous empirical studies on the topic (Kirkman et al. 2006).

The past several decades organizational culture in particular has been increasingly lucrative field of study from managerial point of view (Hofstede 2001). A strong culture can foster organization’s capacity to thrive (Groysberg et al. 2018). Even though organizations can be seen as separate bodies that have their own corporate culture (Mercadal 2014), they are still consisted of individuals and groups that had their own believes, values, principles, and ideologies prior to joining the work place in question (Hofstede et al. 2010). Thus, it could be argued that corporate culture is also fluid and could be shaped to and from its entities’ behaviour and needs.

Scholars have been studying culture’s impact on organizational performance (e.g. Lee & Yu 2004; Ambos & Schlegelmilch 2008; Leach-López 2013; Nazarian et al. 2017) and their researches indicate that a strong culture can affect performance in a positive direction, usually in financial way. However, their studies do not indicate performance on individual level. The present study aims to address the gaps in previous research by examining the potential relation between culture and self-perceived job performance.

(8)

Wilson 2012). Ambos & Schlegelmilch (2008) have empirically proved that certain societies with specific cultural profiles have natural advantage when it comes to innovation. This paper aims to contribute to the existing literature by examining national culture on organizational level and its connection to innovation.

Hofstede’s research gives cultural profiles to more than 40 countries, clustering some of them (Hofstede 2001). The present study is focusing on two of them – Sweden and Finland. Hofstede (2001) puts Nordic countries, together with the Netherlands, in the same cluster when it comes to values and culture, with few differences between Finland and Sweden. Nevertheless, a more recent study of Brodbeck et al. (2000) of the cultural variation of leadership across European countries revealed that even if Finland and Sweden belong to the same cluster, there are some specific differences between the managerial type in both countries. For example, team integration and collaboration are more valued in Finland than in Sweden, while in Sweden a more humane orientation is what is considered to be an important quality for a good leader (Brodbeck et al. 2000). Thus, it is easy to speculate that comparing both countries in even smaller scope will reveal bigger differences that will call for different types of management and cultural construct. After all, there are some specific dissimilarities between both countries. For example, even if Finland is part of the Nordic countries, it is not considered to be part of the Scandinavian culture as Sweden, Norway, and Denmark are. Furthermore, both countries come from different linguistic groups. And finally, Finland is considered to be more homogenous country with low immigration rates, compared to Sweden (Eurostat 2018).

1.2. Purpose

Being able to understand and shape the culture in given organization can bring benefits not only to the organization itself in form of better revenues and more innovation (Ambos & Schlegelmilch 2008), but also to the staff who will feel more contented, motivated, and engaged at the workplace (Groysberg et al. 2018).

(9)

cultural dimensions are considered as a benchmark for the analysis. According to Hofstede’s country scores, Finland and Sweden are part of the same cluster, which in general should mean that both research centres should have little to no difference, with biggest gaps in the masculinity and uncertainty avoidance dimensions (Hofstede 2001). Understanding the cultural variation can help practitioners to predict and address potential problems in cross-cultural interactions.

Furthermore, the study will evaluate the relationships between Hofstede’s dimensions, job performance and innovation, aiming to examine the hypotheses that there is a significant relationship between any of the cultural dimensions and job performance, alternatively innovation. Given that both entities are part of the same company with the same practices and that they have the same function, it can be argued that differences in the culture could be an important determinant in the difference in the levels of job performance and innovation. The study’s results should be able to contribute to the better comprehending of culture’s influence on innovation and job performance in multinational organizations.

1.3. Scope and Limitation

Thousands of scholars have used Hofstede’s cultural dimension in their researches in diverse managerial fields (Kirkman et al. 2006). Majority of them used Hofstede’s framework comparing countries that are distinctly different in cultural aspect in order to increase the variance (Kirkman et al. 2006). It is easy to see and compare differences when they are obvious, but as argued above, even though Finland and Sweden should not exhibit big differences, according to Hofstede’s (2001) research, there are still some distinctions that can lead to cross-cultural problems.

(10)

Being limited to one company, it could be assumed that the conclusions made from this study are applicable to the specific organization and therefore might not work for another multi-national corporation. Another consideration to be made is that the reliability of the study should be tested in a future instance. Nevertheless, the results of this study could be used from different practitioners to understand the influence of culture on job performance and innovation. The results could be used as a frame for making small improvements which can lead to tangible changes.

1.4. Disposition

(11)

2. Literature Review

This chapter focuses on the theoretical background for the study and review of the literature on the topic is presented. The main purpose of this chapter is to critically examine the literature on culture and its connection to innovation and job performance in order to achieve the aim of the study.

2.1. Culture and Organizations

Although there is a vast amount of academic literature on the topic, there is no simple and uniform definition of the word “culture”. Many scholars have constructed definitions of the term, such as the broader anthropological explanation of Kluckhohn and Kroeber (1952; cited in Adler 1997) or the more organizational definition of Schwartz & Davis (1981). For the purpose of this study Hofstede’s (2010, pp. 6) definition will be used and culture will be referred to “as the collective programming of the mind that distinguishes the members of one group or category of people from others”. Hofstede’s definition of “culture” refers to national cultures, but we can define it the same at organizational level as well (Hofstede 2001). After all, organizations are made from people, therefore we can assume that organizational culture is “the collective programming of the mind that distinguishes the members of one organization from others” (Hofstede et al. 2010, p. 344). If culture is adjusted with the personal values and needs of the organizations members, it can support the organization’s capacity to thrive (Groysberg et al. 2018).

(12)

Figure 1: Levels of Culture

As it can be observed from the figure, values represent the deepest level of culture, while symbols are the most superficial manifestation of culture (Hofstede et al. 2010). The core of culture is values and they are defined as broad preferences for certain state of affairs over others and are connected with strong feelings, such as abnormal versus normal or dangerous versus safe (Hofstede et al. 2010).

Rituals, heroes and symbols are positioned together under the concept of practices. Practices are visible for observation but their meaning is invisible for the observer since their cultural meaning lies in the way they are interpreted by the group members (Hofstede et al. 2010). Rituals are shared activities that are essentially irrelevant to the achievement of goals but are still carried out for their own sake (Hofstede 2001). The next “onion layer” in the cultural diagram is heroes. These are persons, real or imaginary, who have characteristics that are highly valued in the society (Hofstede 2001). The last, and most superficial level, is symbols. Those can be either words and pictures, or objects that have a special meaning to those who are part of the group (Hofstede 2001).

Rituals Heroes Symbols

(13)

Managers and their co-workers are all part of national societies so if we want to be able to understand them, we need to understand their societies first (Hofstede et al. 2010). Nevertheless, at the organizational level, cultural differences manifest themselves mainly in practices and not so much in values (Hofstede 2001).

2.2. Hofstede’s Cultural Framework

Hofstede’s cultural dimensions have been used as a basis for measure of culture in many studies since their creation during the last century. His research shows how values in the workplace are influenced by culture (Hofstede Insights 2018a). The famous IBM study, which was used as a basis for the creation of Hofstede’s framework, is known to many scholars who study culture (Kirkman et al. 2006). Hofstede (2001) used data collected from two attitude surveys in the multinational company. The surveys were conducted in the span of almost ten years (Hofstede 2001). Data was gathered from 72 countries and more than 116 000 questionnaires were collected (Hofstede 2001). The input led to the creation of four initial cultural dimensions – power distance index (PDI), individualism and collectivism (IDV), masculinity and femininity (MAS) and uncertainty avoidance index (UAI) (Hofstede et al. 2010). The country scores are measured in a scale from zero to one hundred.

2.2.1.Power Distance Index (PDI)

This dimension is used to describe how the group handles inequality. In societies with low PDI score, employees do not rely upon their bosses and they would prefer to be consulted and included in the decision making. On the other side, in countries with high PDI score, subordinates depend on their managers. Here, however, such dependence can be either expected and even accepted, or rejected. (Hofstede et al. 2010)

(14)

expect to be consulted and managers are called on first name basis (Hofstede Insights 2018b).

2.2.2. Individualism and Collectivism (IDV)

This dimension has to do with people’s understanding and self-describing as a collective “we” or individual “I” (Hofstede et al. 2010). The higher the score of a country is the more individualistic it is considered to be and contrariwise – the lower the score is, the more collectivist a society is. In collective countries group members have bigger emotional dependence of the organization and vice versa (Hofstede 2001). Some of the individualistic characters according to Hofstede et. al (2010) are personal time, freedom and challenge, while in the collective side are attributed importance of training, physical conditions and skills. The level of IDV is expected to influence the employees’ intentions for following the organizational rules, which means that a more moral involvement can be assumed in countries which are collective and more calculative involvement is to be expected in individualistic societies (Hofstede 2001). Etzioni (1975) describes calculative involvement as a low intensity positive or negative orientation to the organization, while moral involvement is high intensity positive orientation. Individualism and collectivism is negatively correlated with uncertainty avoidance index and power distance index (Hofstede 2001).

(15)

Figure 2: Cultural dimensions

2.2.3. Masculinity and Femininity (MAS)

On the high end of this dimension masculinity traits are found and on the lower side is the feminine traits, meaning that a country with low score is to be considered a feminine one. The more masculine a certain society is the more its members prefer and aim for recognition, earnings, advancement and challenge (Hofstede et al. 2010). On the femininity side of the dimension are attributed qualities as cooperation, security, manager friendly environment and physical conditions (Hofstede 2001).

Being in the same cluster, both Finland and Sweden are considered to be Feminine societies but here it can be clearly seen that there is a bigger gap in the scores (see figure 2). The countries with lowest ranking on masculinity i.e. most feminine societies are Sweden, Norway, the Netherlands and Denmark with Finland coming close but not in the top ten (Hofstede et al. 2010). A good manager in such society should be supportive and aim for consensus, and decisions should be made through involvement of all participants (Hofstede Insights 2018b). In feminine societies conflicts are resolved by compromise and negotiation, and incentives for employees could be flexibility, well-being and free time (Hofstede Insights 2018b).

0 10 20 30 40 50 60 70 80 90 100

(16)

2.2.4. Uncertainty Avoidance Index (UAI)

As the name of the index suggests, this dimension describes the degree to which the members feel uncomfortable with uncertainty. High scores on this dimension indicate that people in the society fear ambiguity and try to avoid it at any cost (Hofstede 2001). The fundamental issue here is the need for written and unwritten rules and predictability, and if such is not present stress in the members of the organization can manifest (Hofstede et al. 2010).

UAI is the other index that shows bigger difference between Sweden and Finland. Finland’s score on the uncertainty avoidance index is 59 (see figure 1) which is high and means that the society has a preference for avoiding uncertainty (Hofstede Insights 2018b). This means that people favour rules and even have emotional need for them (even if they seem not to work), time is money, working hard and being busy is the norm, punctuality and precision are highly valued and security is major reason for motivation thus leading to decreasing of innovations (Hofstede Insights 2018b). Sweden per contra has a low score on the UAI (see figure 2), which means that people in the society have low preference for avoiding ambiguity (Hofstede Insights 2018b). Having low UAI means that people don’t want more rules than the necessary and if the rules don’t work they should be altered, schedules are flexible, hardworking is not the norm and not at any cost, punctuality and precision are not natural and deviation from the norm is easily tolerated and thus leading to more innovations (Hofstede Insights 2018b).

2.2.5.Criticism of Hofstede’s Framework

Even though Hofstede’s work has been validated by different studies (Shane 1993, 1995; Kirkman et al. 2006; Ambos & Schlegelmilch 2008) there has been also a considerable amount of criticism towards his work (Sivakumar & Nakata 2001; Spector 2001). In spite of the criticism of Hofstede’s work, it has still been favoured from many scholars, because of its clarity and resonance with managers (Kirkman et al. 2006).

(17)

a restricted sample by involving only one organization (Sivakumar & Nakata 2001).

Hofstede (2001) addresses some of the criticism himself. About the oversimplifying of the model and making it only five dimensions, Hofstede (2001) argues that additional dimensions should be statistically and conceptually independent from the five original ones and those should be validated. On the topic of the reliability of his work and the fact that the IBM research data is obsolete, Hofstede (2001) claims that his dimensions are believed to have centuries-old roots and therefore should be stable over time. Hofstede (2001) points out that his dimensions have been replicated and validated several times after the original study.

Nevertheless, his research is considered to be one of the most comprehensive studies of culture and has been object of many validity and reliability checks (Shane 1995; Hofstede 2001; Kirkman et al. 2006).

2.3. Culture and Innovation

Innovation constitutes of the successful implementation of solutions to problems, ideas, services and products, and bringing them to the market (Harper 2013). Innovation has become a main challenge for organizations who need to adapt to the fast pace of environmental changes in order to keep their competitiveness on the market (Porter & Stern 2001). Scholars have already proved empirically that there is a connection between innovation and one or more of Hofstede’s cultural dimensions (e.g. Shane 1992; Shane 1993; Shane 1995; Taylor & Wilson 2012).

(18)

Shane (1995) did a more broad research focusing only on the connection between uncertainty avoidance and innovation, which confirmed that countries with low uncertainty avoidance have higher levels of organizational innovation.

Ambos & Schlegelmilch (2008) argue that a more feminine society that prioritizes soft skills is more fitting for a R&D capability augmenting laboratories which in general focus on more radical innovations. However, for exploiting laboratories, which focus on more accumulative innovations, more masculine traits are to be favoured (Ambos & Schlegelmilch 2008). Ambos & Schlegelmilch’s (2008) study shows that societies that have high scores on power distance, masculinity and uncertainty avoidance but score low on individualism have natural advantage when it comes to accumulative innovations.

2.4. Culture and Job Performance

Every single organization in the world, no matter of its culture, depends on the performance of people (Hofstede 2001). Defying and studying individual performance has been a topic for research for the past decades (Sonnentag et al. 2008). Job performance can be defined as the actions of the individual which are relevant to the goals of the organization (Schmitt & Borman 1993). Performance should be distinguished from effectiveness, efficiency or productivity (Schmitt & Borman 1993). Effectiveness measures financial values of sales and productivity is the ratio of effectiveness in relation to the cost of the outcome (Sonnentag et al. 2008).

(19)

In her study of the influence of culture on job performance of mid-level managers Leach-López (2013) discovered that uncertainty avoidance and individualism have a significant correlation with job performance. An inverse relationship between UAI and job performance indicates that managers with lower UAI show higher job performance (Leach-López 2013). An interesting and unexpected discovery is found in the relationship between IDV and performance (Leach-López 2013). The relationship between the two variables was proven to be negative which indicates that group work leads to better performance, as opposed to individual efforts (Leach-López 2013).

(20)

3. Research Methodology

This chapter focuses on the methodology that was chosen for the research. It aims to explain the theoretical basis behind the methodology to be used. The chapter also deals with criteria for the validity and reliability of the study and the motivations for adopting the specific research approach.

3.1. Research Strategy

Business research does not happen in a bubble, it exists in the context of the social science disciplines, such as psychology, sociology, anthropology, and economics (Bryman & Bell 2015). There are two approaches to a study which explain the link between research and theory, namely inductive and deductive. The present study adopted the deductive approach, since it is more suitable for the research’s purpose. A deductive research is concerned with deducing hypothesis, based on existing theory and subjecting those hypothesis to empirical examination (Bryman & Bell 2015). The deductive approach was selected since it gives the possibility to explain the correlation between concepts and variables, such as culture, innovation, and performance, by using a quantitative research strategy (Bryman & Bell 2015).

Quantitative research is a research strategy that focuses on quantification of data and analysis (Bryman & Bell 2015). Quantitative studies are usually deductive and objective in their nature (Saunders et al. 2009). Given the purpose of the research and the phenomena that has been studied, a quantitative strategy was adopted, since it gives the researcher the option to test hypothesis and correlations between variables.

3.1.1.Epistemological considerations

(21)

al. 2009). On the other side, interpretivism focuses on subjectivism and advocates for the differentiating between people and object of natural sciences (Bryman & Bell 2015). To achieve the aim of the paper the study has adopted positivist approach. It was chosen because it allows the researcher to stay independent of the observed phenomena and give objective results. The positivist approach also allows for quantitative measuring of the hypothesis and thus perfectly allies with the aim of this study.

3.1.2. Ontological considerations

Bryman & Bell (2015) argue that ontology deals with the nature of social entities, describing two positions – objectivism and constructionism – and linking the differences between them with culture and organization. Objectivism takes social phenomena as being independent from actors (Bryman & Bell 2015). Constructionism, on the other side, implies that social phenomena and their meanings are constantly being revised and changed through social interaction (Bryman & Bell 2015). While objectivism sees the organization as a tangible object, that has its own rules and exist in his own reality that is external to its members, constructionism adopts a completely different view of the organization and culture (Bryman & Bell 2015). Bryman & Bell (2015) argue that by adopting an constructionist position the researcher accepts that actors are involved in constructing and changing the reality in organizations and cultures, not vice versa. Since the paper is exploring a social phenomenon and the main premise of the study is that the organization’s members have impact on the company, a constructionist position was adopted. Even if in general an objective approach is preferred for quantitative studies (Bryman & Bell 2015), constructionism was chosen since it incorporates better with the study’s purpose.

3.2. Research Design

(22)

(Saunders et al. 2009). Descriptive research suits better for the study’s purposes as it is quantitative by its nature so that the information collected can be statistically connected to a population (Saunders et al. 2009). Since the present study focuses on details and description of phenomena, and its aim is to give prediction, a descriptive research was found appropriate. However, while descriptive research only describes phenomena, explanatory research goes beyond and tries to find and explain correlations between variables (Saunders et al. 2009). Therefore, it was also found relevant for the study’s purpose. Since the present study has twofold research purposes – first, to describe and evaluate the cultural dimensions of the organizations, and second to search for correlations and attempt to explain the cause between Hofstede’s dimensions and innovation/job performance – the following paper adopted both descriptive and explanatory concepts.

The research instrument used was a self-completion questionnaire. This instrument was selected over structured interviews because it is quicker to administer, removes the effects of the interviewer, and is more convenient for the respondents (Bryman & Bell 2015).

3.3. Data Collection

There are two types of data that a researcher can use for completion of a study, namely primary and secondary data (Saunders et al. 2009). Primary data is data which is gather specifically for the study’s purpose, while secondary data is data that was already assembled for other reasons that the present study (Saunders et al. 2009).

(23)

To best apprehend the cultural dimensions, the newest version of Hofstede’s own questionnaire (VSM 2013) was selected. The survey was developed and updated specially for assessing and comparing cultures. The culture was measured through Hofstede’s four original constructs – power distance (PDI), individualism (IDV), masculinity (MAS), and uncertainty avoidance (UAI).

Concerning the variables of innovation and job performance, part of the QPS Nordic (Dallner et al. 2007) questionnaire was used. This questionnaire was selected, since it was developed specially for Nordic countries. Questions that describe innovative climate and self-perceived job performance were selected from the survey.

The last part of the questionnaire contained demographical questions, such as gender, age, nationality, and educational level. Questions on occupation, position, and department of employment were included as well. Those were selected with consideration to more easily compare the variables, but also in order to be able to remove variation because of job positions and/or nationality, or at least be able to explain variation due to the abovementioned factors. For example, an administrative employee might not have a lot of space for innovation, while a researcher in the patent division will be affected from innovative climate.

(24)

Figure 3: Sample of survey questions

3.3.1. Sample Selection and Response Rate

One of the most difficult decisions is to select the sample size (Bryman & Bell 2015). Nevertheless, Hofstede (2001) gives recommendation of a minimum sample size of 20. Hofstede’s recommendation has been taken in consideration and more than 20 respondents were chosen to participate in the survey. The total sample size was 142. Response rate was almost 50 percent (49%). 70 questionnaires were answered. A slightly lower response rate was expected since a significant percent of the employees in the research centre in Finland were believed not to possess significant English skills. Nevertheless, Field (2013) recommends that for each predictor in the model, the minimum cases needed are 10. With four to five predictors, 40-50 cases should be sufficient for a regression analysis. In order to increase the response rate, prior to the survey an e-mail through the internal communication channels of both research centres was sent to the employees that encouraged the staff to answer

How often, in your experience, are subordinates afraid to contradict their boss? o Never

o Seldom o Sometimes o Usually o Always

Are workers encouraged to think of ways to do things better at your workplace? o Very seldom or never

o Rather seldom o Sometimes o Rather often

(25)

in order to increase the response rate, which proved to be successful and a new spike of responses was observed.

Since one of the main aims of the study is to examine culture in the whole organization, informants from all the departments of both research centres were invited to participate in the survey. However, in order to be able to more accurately analyse the construct of innovation, employees who are part of the administration departments were excluded when a correlation test was done.

3.4. Data Analysis

At the beginning of the analysis process descriptive statistics were used to check for irregularities in data, such as outliers. The first part of the study aims to compare the scores of Sweden and Finland and in order to accomplish this task, Hofstede’s framework and formulas were used to calculate and compare the scores on each one of the cultural dimensions. To reduce the risk of human error, computing of the means of the responses was done using SPSS statistics.

Following, the second part of the analysis, continued with hypothesis testing. Hypotheses were constructed concerning the relationship between the four cultural dimensions and job performance and innovation. The following hypotheses were proposed:

(26)

Figure 4: Hypotheses

A principle component analysis (PCA) was run for each set of the constructs. Since the items for the specific constructs on the questionnaire are same as the original surveys they are adopted from, and no changes or translations have been made, the factor analysis was done only on those items that describe a certain construct. PCA was run setwise, considering that the items were subjected to many validity and reliability checks, leading to stable questionnaires that do not need additional factor analysis. Another argument for choosing this method for data reduction over taking the means of the items and combining them together for computing a new variable is that the latter method takes the assumption that all the questions have the same weight and significance. Since the present study does not, and as shown from the factor loadings, PCA was seen as the most appropriate method for data reduction.

The next step was checking for correlations and testing the proposed hypotheses through creation of models. All cultural dimensions were taken in as predictors and as dependent variables were chosen innovation, and job performance. Based on that two models were proposed:

Model 1:

y = β0 + β1x1 + β2x2 + β3x3 + β4x4 + Ԑ

(27)

y = innovation x1 = power distance x2 = individualism x3 = masculinity x4 = uncertainty avoidance Model 2: y = β0 + β1x1 + β2x2 + β3x3 + β4x4 + Ԑ where: y = job performance x1 = power distance x2 = individualism x3 = masculinity x4 = uncertainty avoidance

Data analysis was completed with testing the final models for assumption violations.

3.5. Quality Criteria

(28)

3.5.1. Validity

Internal validity tests if the measures used are really measuring the intended constructs (Saunders et al. 2009). Internal validity is usually checked by checking for evidence that supports the answers found in the questionnaire (Saunders et al. 2009). Since the questionnaire used in this study was an already established one, with strong and repeatedly tested result, it was assumed that the validity criterion is achieved.

3.5.2.Reliability

As mentioned above, reliability is concerned with the consistency of a measure (Bryman & Bell 2015). Factors that are involved when assessing if a concept is reliable or not are stability and internal reliability (Bryman & Bell 2015). According to Hofstede & Minkov (2013) the reliability of the VSM questionnaire should be taken for granted. The Cronbach’s alphas for the original four dimensions are shown to be above the threshold of 0.7 (Hofstede & Minkov 2013). The internal reliability of the sections that were taken from QPS Nordic cannot be proven a priori and for those two constructs (innovation and job performance) Cronbach’s alphas were tested. In order to check for the stability, an often used method is test-retest, in which the questionnaire needs to be administered twice to respondents (Bryman & Bell 2015). This reliability criterion was not tested, since the time frame of the study was insufficient for administrating two identical questionnaires at two different time points.

3.6. Ethical Considerations

(29)

3.7. Limitations of the Study

To address some of the main criticisms of studies that use Hofstede’s research as a benchmark for their studies, the present paper has adopted and analyses the four initial dimensions. Further, countries were not chosen by random, and the similarity of the countries’ scores was chosen on purpose. However, this can lead to difficulties in analysing and reporting data.

(30)

4. Empirical Findings

This chapter presents the results of the primary data analysis. Descriptive statistics, reliability checks, and assumption validations are reported, as well as the outcomes of the hypothesis testing and regression analysis. The chapter also focuses on calculation on the country scores of the four constructs.

4.1. Measures

A summary of the demographics of the respondents is presented in table 1. The majority of the respondents were female (62.3%). The age distribution was rather equal in percentage, starting from 30 and up to 59 years. More than half of the respondents reported holding a higher university degree (67.1%). There was an almost equal return rate in both countries, with Sweden having 52.9% of the total responses.

Table 1: Demographic profile

Age N % 20-24 1 1.4 25-29 4 5.7 30-34 12 17.1 35-39 12 17.1 40-49 18 25.7 50-59 19 27.1 60 or over 4 5.7 Gender N % Female 43 62.3 Male 25 36.2 Other 1 1.4 Formal Education N %

Secondary school/ vocational school (10-12 years) 4 5.7 College degree (13-16 years) 19 27.1 Higher university degree (more than 16 years) 47 67.1

Country N %

(31)

(32)

Table 2: Descriptive statistics, factor loadings, and Cronbach's alphas

Table 3 shows the correlations between the constructs. Pearson’s correlation shows significant negative relationship between job performance and individualism, and job performance and uncertainty avoidance. There is

Construct Cronbach’s α Item Factor loading Mean Std. deviation Job Performance 0.831 JPR1 0.844 3.94 0.726 JPR2 0.804 3.71 0.805 JPR3 0.853 3.82 0.748 JPR4 0.759 4.11 0.640 Innovation 0.770 INV1 0.837 3.75 0.799 INV2 0.796 3.96 0.818 INV3 0.850 3.49 0.837 Power Distance 0.842* PDI1 0.616 1.68 0.657 PDI2 -0.477 2.10 0.756 PDI3 0.394 2.74 0.874 PDI4 0.717 2.44 0.870 Individualism 0.770* IDV1 0.578 1.76 0.715 IDV2 0.694 1.88 0.636 IDV3 0.563 1.69 0.580 IDV4 0.592 2.84 0.803 Masculinity 0.760* MAS1 0.836 1.95 0.623 MAS2 0.730 1.78 0.649 MAS3 0.628 2.49 0.812 MAS4 0.378 2.48 0.709 Uncertainty Avoidance 0.715* UAI1 -0.708 3.11 0.704 UAI2 0.790 2.21 0.795 UAI3 0.483 1.65 0.644 UAI4 0.526 2.32 0.931

(33)

also significant negative relationship between innovation and power distance, and innovation and uncertainty avoidance.

The full model, including all the culture constructs as independent variables was tested first. The overall model showed to be significant (F=6.966, p<0.01), and all the independent variables showed also significance (see table 4), which leads to the conclusion that the full model is the most appropriate one. Thus, Hypothesis 1 is fully supported. Furthermore, the R2

value of 0.349 indicates that nearly 40% of the variation is explained by the predictors.

Hypothesis 1 is concerned with the relationships between the four culture dimensions and innovation. The correlation coefficients that are presented in table 3 show that only two of the cultural dimensions have correlation with innovation, however the regression analysis shows that all four constructs have significant relationships with innovation. All dimensions, except masculinity show an inverse relationship with the dependent variable.

Multicollinearity was tested through VIF and tolerance statistics, showing no collinearity in the data. VIF<10 with average VIF no substantially higher than 1. Tolerance statistics were well over 0.2, with average tolerance

JPR INV PDI IDV MAS UAI Gender Age Education

Job performance (JPR) 1

Innovation (INV) 0.549** 1

Power Distance (PDI) -0.180 -0.309* 1

Individualism (IDV) -0.422** -0.225 0.017 1

Masculinity (MAS) -0.246 0.119 -0.081 0.503** 1

Uncertainty Avoidance (UAI) -0.528** -0.371** 0.070 0.002 -0.004 1

Gender -0.078 -0.208 0.122 0.134 0.169 -0.005 1

Age 0.081 0.099 -0.203 0.066 0.132 -0.003 0.200 1

Education 0.007 0.194 -0.250* 0.050 0.090 -0.093 0.215 -0.148 1

**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).

(34)

~0.9. Residual plot, histogram and normal probability plot showed no that the assumptions for homoscedasticity, linearity, and normality were not broken.

Table 4: Hypothesis testing

The full model, including all the culture constructs as independent variables was tested first. The overall model was significant (F=13.612, p<0.01), but not all of the variables showed significance. Masculinity and power distance were proven to be insignificant. After that, a stepwise backword method was adopted, leading to the exclusion of those two variables and creating a reduced model, consisting only of two predictors – individualism and uncertainty avoidance (see table 5). Thus Hypothesis 2 is partially supported. Furthermore, the R2_{value of 0.487 indicates that almost}

50% of the variation is explained by the independent variables.

Hypothesis 2 is concerned with the relationships between the culture and innovation. The correlation coefficients that are presented in table 3 indicate that individualism and uncertainty avoidance have correlation with job performance. The same results are observed through the regression analysis. Both predictors show inverse relationship with the dependent variable.

Multicollinearity was tested through VIF and tolerance statistics, showing no collinearity in the data. VIF<10 with average VIF no substantially higher than 1. Tolerance statistics were well over 0.2, with average tolerance around 1. Histogram, normal probability plot, and plot of the residuals showed

(35)

Table 5: Hypothesis testing

4.4. Hofstede’s Cultural Dimensions

After computing the means of the variables in SPSS, for all the dimensions in respective country, the following scores which can be seen in the chart in figure 5 were found.

Hypothesis 2 F p R2 _ΔR2

Model 1 (full model) 13.612 0.000 0.521

Coefficients B p Standardised β Constant -0.058 0.578 Power Distance -0.197 0.066 -0.185 Individualism -0.554 0.000 -0.509 Masculinity 0.006 0.958 0.006 Uncertainty Avoidance -0.603 0.000 -0.584 Step 2 F p R2 _ΔR2 Model 2 18.511 0.000 0.521 0.000 Coefficients B p Standardised β Constant -0.058 0.575 Power Distance -0.197 0.063 -0.185 Individualism -0.550 0.000 -0.505 Uncertainty Avoidance -0.603 0.000 -0.584 Step 3 F p R2 _ΔR2

Model 3 (final model) 24.720 0.000 0.487 -0.034

Coefficients B p Standardised

β

Constant -0.023 0.824

(36)

Figure 5: Country scores

The differences between Hofstede’s reported scores, which were used as a reference scores and the scores from the present study are displayed in figure 6. Hofstede’s reported scores are marked with dark yellow and dark blue. The computed scores from this study are marked with light yellow and light blue. The biggest shifts, when comparing to the reference values can be observed in individualism and masculinity indexes. Power distance and uncertainty avoidance show no drastic deviation from the reference values.

0 10 20 30 40 50 60 70 80 90 100

(37)

(38)

5. Discussion

The following chapter focuses on analysis of the results of the study. An overall review and comparison of the culture scores is presented. Further, the regression models and their applications are discussed. Potential limitations of the study and results are also considered in the present chapter.

5.1. Cultural Dimensions

The results show no significant gap between the countries in the power distance index. When using Hofstede’s reported scores as a reference, it could be assumed that there has not been a significant change in this dimension. With low scores on this dimension, employees are expecting decentralized power with causal relationship to managers (Hofstede Insights 2018b).

In the individualism index a more drastic change and difference can be observed between the countries. The reference scores put both Sweden and Finland in the individualistic category. According to the study’s results, Finland is to be considered a collectivist society, while Sweden has a medium score of 53 which puts the country in neither category. Nevertheless, the results show a tendency towards collectivist society for Sweden as well. In collectivist societies people are part of groups to which they are loyal, in exchange for the group’s protection (Hofstede Insights 2018b). In such societies employee relationships are resembling family relationships and promotions are not done entirely on merit (Hofstede Insights 2018b). The present development of the scores is intriguing, since Northern countries (and high-income countries in general) have reputation for being highly individualistic societies. Such shift could lead to internal problems, if company’s managers and policies do not adapt to employees’ culture.

(39)

more flexibility and free time, as well as causal relationship with managers (Hofstede Insight 2018b). Manager friendly environment incorporates well with both countries’ low scores on the power distance index.

An interesting result can be observed in the uncertainty avoidance index. There is almost no shift from the reference scores. However, this does indicate that there is a significant difference between the countries, making Sweden a country with low uncertainty avoidance and Finland with high. As one of the characteristics of uncertainty avoidance is preference and emotional need for rules (Hofstede Insights 2018b), the difference in the cultures should be taken into consideration from managers both on local and international levels.

According to the study’s results all four dimensions have relationship with innovation, with majority of them showing a negative one. Only masculinity exhibits positive correlation. The coefficient of determination indicates that the model explains 35% of the variability of the response data. Such percentage should be seen as relevant, since there are many other factors that are not included in the model, which could have impact on innovation, such as job performance.

(40)

According to the regression analysis uncertainty avoidance and individualism have highest impact on innovation. Power distance and masculinity have a smaller one. Sweden has a neutral score on individualism and low score on uncertainty avoidance, indicating that the country should have a natural advantage in innovation. The score on power distance is also small, which should only enhance the innovation levels. The only dimension that does not fit the model is masculinity which should be higher or increased so that innovation levels rise. Finland has low score on the individualism index but scores put the country high up in the uncertainty avoidance index. Being one of the main predictors, steps in decreasing individualism should be taken in order to increase innovation. As with Sweden, Finland has low score both in power distance and masculinity, thus the same recommendations apply.

An interesting result can be observed in comparing the correlation matrix with the regression. The correlation analysis showed no significant relationship between innovation and individualism; and innovation and masculinity. Nevertheless, the regression analysis showed that all of the four constructs have a significant relationship with innovation. It could be speculated that abovementioned two dimensions have part in predicting innovation, but only when they interact with other cultural constructs.

The study’s findings show a significant inverse relationship in two of the four cultural dimensions, that is individualism and uncertainty avoidance. According to the coefficient of determination the model explains 49% of the variability of the data. Given that there are many factors that contribute to job performance, which are not included in the model, such percentage should be seen as relevant.

(41)

(42)

6. Conclusion

This chapter gives a summarised conclusion of the research and also points out its limitations and gives recommendations for further research.

The study’s findings indicate that overall the cultural profile of the countries has not changed drastically, when using Hofstede’s scores as a reference. There is, however, an intriguing shift in the individualism dimension. As opposed to common knowledge and expectations, as well as Hofstede’s (2001) reports, that Nordic countries are highly individualistic, the present study found that Finland has deviated and became a collectivist society, with Sweden seemingly following the same path. Such change in values could be an important factor in many aspects – from organizational level to psychological health – not only for job performance and innovation, and, therefore, requires further research.

The present study showed that there is a significant relationship between more than one of the cultural constructs and innovation, and, respectively, job performance. The correlation matrix showed another intriguing significant correlation, that is between innovation and job performance. This relationship has not been tested. If included in the models, it could lead to better explanation of the dependent variables. The influence of job performance on innovation could be an important factor in understanding and increasing innovation levels in companies and could be a topic for further research, and also for development of the present models.

The present study’s results can be used by multinational companies in decisions as to where to place R&D centres or how to coordinate company policies so that job performance and innovation levels rise. Another possible implication that can be used by managers in local levels is for better understanding of cultural differences and using the knowledge in inter-cultural interactions, and when coordinating international projects between those two countries.

(43)

(44)

Bibliography

Adler, N. J. (1997). International dimensions of organizational behavior. Cincinnati, Ohio : South-Western College Pub, cop.

Ambos, B. & Schlegelmilch, B. B. (2008). Innovation in multinational firms: Does cultural fit enhance performance? Management International Review, 48(2), 189. doi:10.1007/s11575-008-0011-2. Brodbeck, F. C., Frese, M., Akerblom, S., Audia, G., Bakacsi, G., Bendova,

H., Bodega, D., Bodur, M., Booth, S., Brenk, K., Castel, P., Hartog, D. D., Donnelly-Cox, G., Gratchev, M. V., Holmberg, I., Jarmuz, S., Jesuino, J. C., Jorbenadse, R., Kabasakal, H. E. & Keating, M. (2000). Cultural variation of leadership prototypes across 22 European countries. Journal of Occupational & Organizational Psychology, 73(1), 1–29.

Bryman, A. & Bell, E. (2015). Business research methods. Oxford : Oxford Univ. Press.

Dallner, M., Lindström, K., Knardahl, S. & Örhede, E. (2007). QPSNordic - the General Nordic Questionnaire for Psychological and Social Factors at Work.

Etzioni, A. (1975). A comparative analysis of complex organizations : on power, involvement, and their correlates. New York: Free P.

Eurostat (2018). Migration and migrant population statistics - Statistics Explained. http://ec.europa.eu/eurostat/statistics-explained/index.php/Migration_and_migrant_population_statistics [2018-02-15].

Field, A. (2013). Discovering statistics using IBM SPSS statistics : and sex and drugs and rock “n” roll. Los Angeles : London : Sage.

(45)

Harper, G. D. J., MSc (2013). Innovation. Salem Press Encyclopedia.

Hofstede, G. H. (2001). Culture’s consequences : comparing values, behaviors, institutions, and organizations across nations. Thousand Oaks, Calif. : Sage.

Hofstede, G. H., Hofstede, G. J. & Minkov, M. (2010). Cultures and organizations : software of the mind : intercultural cooperation and its importance for survival. New York : McGraw-Hill.

Hofstede, G. & Minkov, M. VALUES SURVEY MODULE 2013 MANUAL, 17.

Hofstede Insights (2018a). National Culture. Hofstede Insights. https://www.hofstede-insights.com/models/national-culture/ [2018-03-6].

Hofstede Insights (2018b). Country Comparison. https://www.hofstede-insights.com/country-comparison/ [2018-02-9].

Kirkman, B. L., Lowe, K. B. & Gibson, C. B. (2006). A quarter century of Culture’s Consequences: a review of empirical research incorporating Hofstede’s cultural values framework. Journal of International Business Studies, 37(3), 285–320. doi:10.1057/palgrave.jibs.8400202.

Leach-López, M. A. (2013). Moderating Effect of Hofstede’s Cultural Values on the Locus of Control/Job Performance Relationship of Managers in USA, Mexico, South Korea and Hong Kong. Journal of Business Strategies, 30(1), 1–20.

Mercadal, T. (2014). Organizational culture. Salem Press Encyclopedia. Nazarian, A., Atkinson, P. & Foroudi, P. (2017). Discussion paper:

Influence of national culture and balanced organizational culture on the hotel industry’s performance. International Journal of Hospitality Management, 63 22–32. doi:10.1016/j.ijhm.2017.01.003. Porter, M. E. & Stern, S. (2001). Innovation: Location Matters. (cover

(46)

Saunders, M., Lewis, P. & Thornhill, A. (2009). Research methods for business students. Harlow : Financial Times Prentice Hall.

Schmitt, N. & Borman, W. C. (1993). Personnel selection in organizations. San Francisco: Jossey-Bass.

Schwartz, H. & Davis, S. M. (1981). Matching Corporate Culture and Business Strategy. Organizational Dynamics, 10(1), 30–48.

Shane, S. (1992). Why do some societies invent more than others? Journal of Business Venturing, 7(1), 29–46. doi:10.1016/0883-9026(92)90033-N.

Shane, S. (1993). Cultural influences on national rates of innovation. Journal of Business Venturing, 8(1), 59–73. doi:10.1016/0883-9026(93)90011-S.

Shane, S. (1995). Uncertainty Avoidance and the Preference for Innovation Championing Roles. Journal of International Business Studies, (1), 47.

Siew Kim Jean Lee & Kelvin Yu (2004). Corporate culture and organizational performance. Journal of Managerial Psychology, (4), 340. doi:10.1108/02683940410537927.

Sivakumar, K. & Nakata, C. (2001). The Stampede toward Hofstede’s Framework: Avoiding the Sample Design Pit in Cross-Cultural Research. Journal of International Business Studies, (3), 555.

Sonnentag, S., Volmer, J. & Spychala, A. (2008). Job Performance. In The SAGE Handbook of Organizational Behavior: Volume I - Micro Approaches. London: SAGE Publications Ltd, ss.427–448.

Spector, P. E. (2001). An International Study of the Psychometric Properties of the Hofstede Values Survey Module 1994: A Comparison of Individual and Country/Province Level Results. Applied Psychology: An International Review, 50(2), 269.

(47)

(48)

Culture Matters

Neshe Tuna

Culture Matters

Analysis of Culture in Sweden and

Finland and Its Influence on Innovation and

Job Performance

Business Administration

Bachelor’s Thesis

15 ECTS

Abstract

Table of Contents

1. Introduction

2. Literature Review

3. Research Methodology

4. Empirical Findings

5. Discussion

6. Conclusion

Bibliography