Challenges in performance measurement

Get Complete Project Material File(s) Now! »

Method

This chapter presents the chosen research methodology and method. The data collection and the method of analysis are extensively discussed and motivation for chosen research choices is provided in the context of fulfilling the purpose of this thesis.

Research philosophies

Research philosophies are philosophical frameworks that guide the research process and specify methods that should be adopted in order to execute the research with a clear purpose and strategy. Collis and Hussey (2013) and Braa and Vidgen (1999) explore and compare three main philosophies of methodology: positivism, interpretivism and pragmatism. Positivism is a concept aiming for explanation and prediction by providing theories to understand a social phenomenon, which is assumed to be measurable (Braa & Vidgen, 1999). Thus positivism is often associated with quantitative research methods where the results of research can lead to law-like generalisations (Collis & Hussey, 2013; Remenyi et. al, 1998). Interpretivism aims for interpretation and understanding of complex social phenomena by developing theories or patterns (Collis & Hussey, 2013). It strives to explore the meaning attributed to experiences of individuals, which ties interpretivism to qualitative research (Rae, 2013). Pragmatism is an approach that aims for intervention and change and therefore provides knowledge or insights that are useful in action (Braa & Vidgen, 1999). In order to adequately answer the research question of this study, an overall research philosophy of pragmatism is used as the aim is to contribute constructive knowledge that is useful in practice (Braa & Vidgen, 1999; Johnson & Onwuegbuzie, 2004).
As the authors also aim to provide an understanding of the importance of performance measurement as well as new insights to the phenomenon of performance measurements for social enterprise start-ups, an additional interpretivist approach is appropriate. The philosophy of interpretivism allows the authors to consider the attitudes that social entrepreneurs might have towards performance measurement (Collis & Hussey, 2013).
Therefore the research might not be law-like generalisable and is not undertaken in a value-free way, which does not hold a philosophy of positivism (Remenyi et al., 1998).

Research approaches

For this research a purely deductive research approach is not feasible, since the development of propositions or the hypotheses testing of current theories is not possible as the existing literature within the research field is not yet saturated (Dubois & Gadde, 2002; Robson, 2011). Moreover it is not appropriate to solely work inductively, because the frame of reference on existing theories and frameworks provided the authors with a direction for this study. The knowledge contribution is thus not only generated from empirical data (Dubois & Gadde, 2002). Therefore the most appropriate research approach for this study is an abductive approach, which resembles a conceptual bridge between the deductive and inductive research approaches. It allows researchers to generate new conceptual models by associating data with concepts, based on existing literature and theories as well as empirical findings (Dubois & Gadde, 2002; Lipscomb, 2012; Murphy et al., 2017).

Research method

There are two commonly known research methods, quantitative and qualitative research. In a quantitative research hypothesis are tested by examining the relationship among variables which are measureable and thus quantifiable (Creswell, 2013). A qualitative research on the other hand focuses on the complexity of a phenomenon and seeks to make sense of the meaning of experiences of individuals, which is why it is mostly consisting of non-numeric data (Creswell, 2013). The concept of performance measurement within social enterprise start-ups remains mainly unexplored by existing literature and there might still be important variables to examine that are unknown. Thus, a qualitative research through interviewing was found to be the most suitable method for this study.

Data collection

Primary data

The method of interviewing was chosen for the primary data collection, which is the predominant mode of data collection in qualitative research (May, 1991). Interviews can, depending on the research objective, vary in their structure and question formulation. As the authors are interested in understanding the opinions and attitudes of the interviewees towards performance measurement in social enterprise start-ups, a semi-structured interview with open-ended questions was found the most appropriate. A semi-structured interview means that while some questions are prepared to ensure the discussion of main topics of interest, other questions are developed during the course of the interview (Collis Hussey, 2013; Wengraf, 2001). Additionally open-ended questions enable the interviewee to give longer answers, which subsequently enables the researchers to gain a deeper understanding of the phenomenon. Therefore a semi-structured approach, with pre-written open-ended interview questions and the ability to deviate from the questions depending on given answers, was found the most appropriate for this study.

Interview guide construction

In a semi-structured interview the interview guide construction is crucial and should be based on relevant literature as well as the researchers own knowledge (King, 2004). Thus the pre-written open interview questions were formulated on the basis of the frame of reference as well as the initial conceptualisation of the authors. From the frame of reference, themes like start-up phases, measurement systems, stakeholders and challenges were used to understand the topic. In order to put these in a logical order the interview questions were divided into the main themes of background, stakeholders, measurement systems, benefits, contextual factors and priorities.
The background themed questions had the purpose of establishing the background information of the business, including the purpose and the mission of the organisation. This is especially important in the start-up phase as formalising the business model is crucial in the initial stages of a social enterprise start-up (Perrini et al., 2010). With the questions on stakeholders, stakeholder relations and influences on the organisation were discussed. This can be linked to the multiple constituency approach to performance measurement where all stakeholders should be considered in order to measure effectively (Costa & Pesci, 2016). The authors were also interested in what kind of funding the start-up has received as it was found in the research that funders as external stakeholders can be especially important (Luke, 2016).
The topic of desired outcomes towards stakeholders was included in the interview questions in order to establish a base for performance measurement related questions where reasons for measuring or not measuring were explored. At this stage the questions were focused on what kind of measurement was in place in each of the social enterprise start-ups and how measurement was approached. This was linked to the benefits that the start-ups could seize, as questions on consequences of measurement were formulated. Contextual factors of founder backgrounds and funding types were identified as influential for the consequences, creating the next main theme of questions. The challenges identified in the frame of reference made this especially interesting as exploring reasons why measurement was not done was important for the authors. Finally the priorities of the start-ups were questioned, as the authors wanted to discover at what stage the social enterprise start-up was in order to further investigate how the stage would link to performance measurement. As the interviewed start-ups differed in many ways, the questions were adjusted accordingly. The specific interview questions can be found in Appendix 1.

Selection process for primary data sources

For the selection of the interviewees the concept of theoretical sampling was followed and thus the informants were selected based on their apparent knowledge and experience (Murphy et al., 2017). Due to the limited time frame of this research, the authors were however not able to continually seek out for additional informants that could contribute with their answers to shape the emerging conceptual model based on priorly collected data, which is why the concept of convenience sampling was additionally followed. Convenience sampling enabled the researchers to use networks (Collins & Hussey, 2013) to initiate the first contact with the interviewees, which ensured their availability and successful collaboration. For the sake of cost constraints, the interviews were conducted over Skype as most of the interviewees were not located in Sweden. The interviews were recorded to ensure accuracy and the complete documentation of the given answers, which has been communicated to the interviewees prior to the interviews. Six social enterprise start-ups were interviewed which enabled the authors to see the research topic from the perspectives of the interviewees as well as to understand how and why they have their particular perspective (King, 2004). To ensure the research was conducted in an ethical way, it was made sure that all participants were informed and aware of the purpose of this study as well as the use of the empirical data. Additionally the authors emphasised that the names of companies and individuals were used anonymously, which established a basis of confidentiality and trust.
To assure the relevance of the primary data sources, the authors decided upon a set of characteristics that the interviewees had to possess in order to be found appropriate for this study. These characteristics have been identified as essential through the literature review, to ensure that the interviewees are competent to contribute with knowledge and insights to the research topic of performance measurement within social enterprise start-ups. In regard to the choice of the interviewed social enterprise start-ups, the questioned organisation had to be categorised as a social enterprise, which is widely based on the formulation of a social goal (Barraket & Yousefpour, 2013; Luke, 2016). In order to increase the spectrum of the research, various social goals were incorporated which subsequently enhanced the applicability of the proposed model as it covers a broader context. Additionally the organisation had to be in the start-up process, which is why only organisations that were not older than three years were considered (Robehmed, 2013). All these characteristics have been reviewed based on background research on the organisations in question, whilst they ultimately were ensured by clearly stating the purpose of the research in the initial contact with the chosen interviewees.

Secondary Data

It is often beneficial to combine primary data with secondary data, which is defined as information gathered from existing sources (Bryman, 2012; Saunders, Lewis & Thornhill, 2009). By re-analysing existing data the researchers get access to more information, which can be used to form a valid foundation for the research of this thesis. The secondary data was collected through reviewing existing literature regarding social enterprise start-ups and performance measurement. The sources used were primarily online databases such as Primo, Web of Science and Google Scholar and in order to find relevant documents the following search parameters were used: Search Words: social enterprise start-ups, social business, performance measurement, impact measurement, start-up, social enterprise, start-up process, social ventures, assessment and evaluation; Literature Type: Books, Peer-reviewed articles, Internet; Publication Period: 1980-2017; Publication Languages: English, Swedish, Finnish, German. To ensure reliability and high-quality information the number of citations of the articles was taken into account.

Method of analysis

Combining the pragmatic and interpretivist approach of this study, the grounded theory approach was found as suitable for analysing and conceptualising the empirical data. The method of grounded theory aims to generate theory by collecting and analysing primary data and is thus an essential research method for the development of new insights into relatively unaddressed or new social phenomena (Murphy et al., 2017). It additionally considers the context in which individual behaviour takes place, which enables the researchers to gain a conceptual understanding of the issues that make up the environment of the phenomenon (Fendt & Sachs, 2008; Murphy et al., 2017). As this research aims to gain insights into how a social enterprise start-up could approach performance measurement and the field of social enterprise start-ups is relatively unaddressed, the grounded theory method enabled the authors to develop new insights and understandings based on primary data.
According to Murphy et. al (2017) the grounded theory is based on four core principles: emergence, constant comparison, theoretical sampling and theoretical saturation. Emergence refers to the openness of a grounded theorist towards new developments and data that emerge during the data collection and analysis. Constant comparison means that emerging data, existent data and existent literature are continually compared in order to construct a theory of social reality that is grounded in past and present data. Through the theoretical sampling, data sources are chosen on the basis of usefulness. Theoretical saturation means that the research process is only completed if the theoretical categories are sufficiently saturated and comprehensive in scope and depth.
There are two different approaches within grounded theory, the Gioia approach and the twin-slate approach (Kreiner, 2016; Murphy et al., 2017). While both approaches follow the same process and core principles, the twin-slate approach allows the early integration of existing literature, which enables the researchers to conduct the research in an abductive manner. It also increases the efficiency of the research as it ensures that no time is spent on ideas that already have been covered in existing literature, whilst it is connected to prior work which increases the chance of an impactful theoretical model. Therefore the twin-slate approach was found suitable for this research as it allows a theory-data interplay, which leads to a logic of discovery based on connecting conceptual ideas and empirical data (Van Maanen, Sørensen & Mitchell, 2007).

Research process

As mentioned by Gioia, Corley and Hamilton (2012), there are four features of methodology that enhance the development of grounded theory, which have been considered thoroughly to ensure a significant outcome of this research. As for the first feature, the research design, a well-defined, yet relatively unexplored phenomenon of interest was identified, which is performance measurement within social enterprise start-ups. Initial research questions were formulated, derived from existing literature, whilst keeping the concept of emergence in mind which allows the researchers to adapt their research questions in course of the research (Gioia et al., 2012). The second feature of data collection includes the principle of theoretical sampling. Thus the data sources should be knowledgeable agents that possess knowledge and experience within the research topic, which is why social enterprise start-ups were interviewed. Interviews are widely used as a qualitative data collection method in connection to the grounded theory model, where a flexible approach tailored towards the research questions is of preference (Gioia et al., 2012; Murphy et al., 2017; Strauss & Corbin, 2008).
Subsequently the feature of data analysis is brought up, which is essentially based on data coding. After the interviews have been transcribed the researchers “code” the data, which means the data are fractured, conceptualised and integrated to form theory (Strauss & Corbin, 2008). According to Corbin and Strauss (1990), there are three main coding phases: open coding, axial coding and selective coding. The first phase of open coding entails the basic process of breaking down the data by comparing, conceptualising and categorising. The transcripts were thoroughly read through and then separately broken down into themes, which then were compared and combined. During the process of axial coding, the identified themes or codes through the open coding were linked together based on contexts and patterns. The last step of selective coding selected the core theme and systematically related it to the other identified themes, whilst relationships were clarified and conceptualised. Thus the process of the data analysis began interactively as the data is collected and labeled with themes and codes, which are refined and clarified over time through axial and selective coding.
Besides the analysis through coding, several scholars have suggested the strategy of “memo-writing” as well as comparative techniques to further analyse the collected data and uncover patterns (Corbin & Strauss, 1990; Murphy et al., 2017; Strübing, 2014). Memo-writing is the process of writing notes that document one’s thought process throughout the research process. These memos can be potential clues, theoretical connections or insights that enable the researchers to identify components for the conceptual model that leads to the theory formulation (Murphy et al., 2017). Therefore the authors have constantly written individual memos to capture their understanding and draw connections between common themes during the whole research process, but especially whilst conducting the interviews. After each interview the individual memos were compared to identify similarities and differences which further helped to develop a conceptual model. This process can be seen as a comparative technique. Apart from comparing the written memos, the emerging data was constantly compared with prior interviews as well as secondary data sources to identify patterns and interrelations.

Table of contents
1 Introduction
1.1 Background
1.2 Problem
1.3 Purpose
1.4 Research questions
1.5 Delimitations
1.6 Definitions
2 Frame of reference
2.1 Social enterprise start-­‐ups
2.2 Performancemeasurement systems
2.3 Challenges in performance measurement
2.4 Summary of frame of reference
3 Method
3.1 Research philosophies
3.2 Research approaches
3.3 Research method
3.4 Data collection
3.5 Method of analysis
3.6 Research process
3.7 Method critique
3.8 Research quality
3.9 Method summary
4 Empirical Findings
4.1 Introduction of the interviewed social enterprise start-­‐ups
4.2 Interview results
5 Analysis
5.1 Open coding
5.2 Axial Coding
5.3 Selective coding
5.4 Proposed model
6 Conclusion
7 Discussion
7.1 Implications
7.2 Limitations
7.3 Suggestions for future research
8 Referencelist

GET THE COMPLETE PROJECT
Performance measurement in social enterprise start-ups