The quality of any research is generally assessed by its reliability, validity, generalisability and credibility. Reliability is the tendency of a body of research to produce consistent results if measured repeatedly (Maylor and Blackmon, 2005; Bryman, 2016). Validity is concerned with the integrity of conclusions generated from a study (Bryman, 2016); a research is valid if it captures the truth of the situation and is not influenced by outside influences or personal preferences (Maylor and Blackmon, 2005).
According to Brinberg and McGrath (1985), validity carries different meanings during various stages of a research process. During the three stages of research process – the
‘prestudy’ stage, the ‘central’ stage and the ‘follow-up’ stage – validity means value or
worth, correspondence or fit, and robustness or generalisability, respectively.
Generalisability refers to the extent to which the findings of a study are relevant in another setting or situation (Stokes, 2011). Credibility questions whether or not the results are well founded; research is considered credible if the findings are supported by the evidence (Maylor and Blackmon, 2005).
Yin (2009) describes four criteria for judging the quality of a research design: 1) construct validity: identifying correct operational measures for the concepts being studied; 2) internal validity: seeking to establish a causal relationship whereby certain conditions are believed to lead to other conditions, as distinguished from spurious relationships; 3) external validity: defining a domain to which a study’s findings can be generalised; and 4) reliability: demonstrating that the operations of a study – such as the data collection procedures – can be repeated, with the same results. Tashakkori and Teddlie (1998, p. 89-93) have presented some strategies for checking the quality of qualitative data, which were partly applied in this study.
These criteria for quality control guided the research. Reliability was ensured by following a case study protocol (Yin, 2009) consisting of the following sections:
overview of the study, data collection procedures, data collection questions and a guide for the report. The protocol was important, as it helped the researcher maintain the focus of research and anticipate problems. For example, the necessary sections in the case study protocol obligated the researcher to identify all important aspects long before the case study was conducted.
To tackle the issue of validity, multiple sources of evidence or triangulation were used.
Triangulation can be classified into four types: data triangulation, investigator triangulation, theory triangulation and methodological triangulation (Patton, 2002). In this study, construct validity was verified through these types of triangulation. Theory triangulation was achieved through combining several theories to interpret and understand each case. In each case study, findings were supported by one or more methods of data collection. For example, in Case I, data was collected through interviews and observations alongside specific documentation. In Case II, data was collected through interviews and observations, and in Case III, questionnaires and data from workshops were used to produce findings. In Case III, investigator triangulation was used to validate the interpretations of the data and cross-check the conclusions.
In terms of generalisability, some cases are specific to a particular target market and generalising them may be inappropriate. Maylor and Blackmon (2005) suggest that if the research is not applicable to other settings, generalising should be avoided, while Yin (2009) claims the lessons learned from case studies can be subject to analytical generalisation. In this instance, the generalisable findings or lessons learned in one case study were applied in reinterpreting the results of the other case studies. In addition, the sub-studies that resulted from the case studies were presented at scientific conferences and published in peer-reviewed journals, highlighting the validity and generalisability of the research.
This dissertation is a combination of several in-depth case studies chosen from various perspectives to enhance the researcher’s understanding of the central phenomenon under investigation: ‘How socially driven businesses and innovations contribute to social sustainability’. The individual cases were chosen from various contexts, and in each case, critical attention was paid to the quality criteria throughout the research process. Various methods were employed to determine the trustworthiness of the qualitative research (Tashakkori and Teddlie, 1998): 1) prolonged engagement: the researcher spent nearly four years studying the phenomenon under investigation; 2) persistent observation: the researcher identified the characteristics and most relevant aspects of the phenomenon under investigation through persistent observation of the phenomenon; 3) use of triangulation techniques: triangulation techniques, including triangulation of sources, methods and investigators, were used in all the individual cases; 4) peer debriefing: the researcher discussed various aspects of the study with peers to clarify interpretations and probe biases, which contributed to the credibility of the study; and 5) reflexive journal: a reflexive diary was used, in which the researcher recorded a variety of information about the phenomenon under investigation.
5.3.1 Reflections of the researcher
This section describes the researcher’s role while conducting the empirical study. It provides a discussion about the researcher’s experiences and reflections during the empirical research process. As this study was qualitative in nature, the distance between the researcher and the participants was rather diminished due to the fact that the participants were considered the best experts to know about the subject matter. Therefore, their perspectives have been taken into consideration, which demands the understanding and increased familiarity between the researcher and the participants. It can be very well assumed that the empirical material collected during the research process was
‘representative of the participants’ social worlds’ (Eriksson and Kovalainen, 2008).
The researcher could be considered a primary instrument of the study, which means that the research process has been mediated through the researcher’s understanding rather than through any technical equipment. Therefore, the researcher’s values, age, gender, race, ethnicity and experiences have shaped the research process and the understanding and interpretation of the research material is impacted by these factors. The researcher has therefore been an active agent during the research process. The hermeneutic approach is visible in this research. The researcher’s prior knowledge of the target market, which can be observed in the case studies has formed a basis of this research. For example, in Case I, the rationale behind selecting Srinagar as a research area was the result of the severity of winters that the researcher had personally experienced. The prior knowledge about the uncomfortable housing and unbearable climatic conditions developed certain pre-understanding. In Case II, selection of SHP as a case study was the result of researcher’s familiarity with the topic of green energy businesses in emerging markets that the researcher had studied prior to conducting this research. However, this prior knowledge is only regarded as preliminary version of understanding, which has modified during the research process. The researcher has interpreted and understood things based on her
pre-understanding. Her gender, experiences, culture and expectations have played a big role in understanding the overall reality. As the prior knowledge base grew, the position towards this knowledge also modified. However, the researcher has been an outsider while gathering the research material, which gave her a better opportunity to stay increasingly analytical and even critical, towards the issues that she was examining.
As mentioned earlier, the evaluation of the research was done during the whole research process, not only at the end. In addition to the above mentioned quality criteria, the criteria that assessed the goodness or trustworthiness of the research were developed such as dependability, credibility, conformability, and transferability (Eriksson and Kovalainen, 2008). First of all, the researcher has produced a dependable research as it is logical, traceable and documented. Second, it is credible as the researcher has sufficient data to support the findings. Third, it is conformable as it links findings and interpretations to the data in ways that are easily understood by others. Lastly, it is transferable as some similarities can be found in other research contexts.