R ELIABILITY , V ALIDITY & C REDIBILITY

4.11.1 Reliability

Reliability refers to the ability to replicate the design of previous research and achieve the same results. In other words, it relates to the consistency and replicability of the research (Saunders

et al., 2016). Several measures have been taken to facilitate the reliability of both the qualitative and quantitative data of this thesis.

To mitigate the threat of participation error in the qualitative data collection, the interviews were conducted during office hours at a time considered most convenient by the participants.

Furthermore, all interviews were conducted in a closed space to ensure that the interviewees were able to speak freely without being overheard (Saunders et al., 2016). For the internet-mediated interviews, it is assumed that the interviewees were able to find a closed space themselves. Moreover, potential social influence affects the interviews in situations where the participants are reluctant to divulge information on certain topics. There is a risk that the interviewee is inclined to cast their respective organization in a positive fashion, which in turn reduces reliability.

For the quantitative data, measures were taken to facilitated reliability through the utilization of methods, parameters, and measurements established in previous research. In order to examine the phenomena thoroughly and avoid misinterpretations, questions were designed with rating-scales, as well as matrix-scales (Bryman & Bell, 2015). The clarity and universal meanings were enabled through pilot-testing. The pilot-testing was performed in an attempt to minimize any potential bias. Furthermore, all the information retrieved from the questionnaire was anonymous in order to mitigate potential social influences (Söderlund & Öhman, 2005).

This was considered particularly important due to the topic of this questionnaire, as respondents might otherwise be inclined to portray themselves as more environmentally responsible than they actually are.

4.11.2 Validity

Validity is concerned with the accuracy and generalisability of research and can be interpreted as the integrity of the conclusions derived from the findings (Bryman & Bell, 2011). The validity of this research has been facilitated through internal and external validity (Saunders et al., 2016).

Internal Validity

The internal validity of the quantitative data refers to the ability of the questionnaire to measure the aspects that it intends to measure, referred to as the measurement validity (Saunders et al., 2016). The intent of the questionnaire was to analyse whether the area of improvement identified in the preceding qualitative aviation sector interviews corresponded with consumer demand. Furthermore, the purpose was to gain insights into the attitudes and perceptions of the consumers in regard to carbon offsetting, as well as whether or not a proposed digital solution could improve their perception. It was found that the questionnaire produced valuable insight into the intended aspects, as such, measured what it intended to measure. The responses collected from the questionnaire provided the researchers with the data required to fulfil the research objective, and as such establishes measurement validity.

External Validity

External validity refers to whether the findings of the research can be generalized to other relevant groups or cases (Saunders et al., 2016). In relation to the qualitative data, the interview participants have been carefully selected with a purposive non-probability sampling method in an attempt to achieve generalisability. The main criteria for the interview subjects within the Scandinavian aviation industry were substantial knowledge within the current carbon offsetting and environmental practices of their respective companies. Furthermore, it was deemed appropriate to conduct interviews with both airlines and their offsetting partners in order to ensure that the data collected was from different perspectives. The participants of the blockchain expert interviews were selected based on their knowledge and experience within blockchain technology, with a particular focus on individuals with an insight into the usage of blockchain for environmental purposes or supply chains.

In an effort to facilitate external validity in the questionnaire, the data was gathered from a large sample of respondents. However, the application of a convenience sampling method is not able to ensure the representativity of participants, and as such the data are not necessarily generalizable (Malhotra et al., 2017). Evidence of this can be seen in the distribution of responses in terms of both gender and age, which essentially reduces the external validity of this research. However, as a sequential mixed-methods approach has been utilized in an attempt to validate and substantiate the previous findings of the qualitative data collection in the

aviation industry, the findings of the questionnaire are considered to generate some degree of external validity.

Nonetheless, this thesis is a single case study on the specific circumstances of voluntary carbon offsetting in the Scandinavian aviation industry, which entails that the findings are rather specific to the present conditions. As such, it is difficult to establish the generalisability of this research (Saunders et al., 2016). It could, however, be argued that the findings may be generalizable to voluntary carbon offsetting practices in the aviation industries elsewhere, assuming that the professional actors and consumers share similar perceptions and attitudes like those in Scandinavia.

4.11.3 Credibility

Credibility is considered to be the most crucial criterion in establishing the trustworthiness of the research. The main purpose of credibility is to guarantee the connection between the findings of the research and the socially constructed realities of the participants of the research (Saunders et al., 2016). The triangulation method has been applied in order to ensure the credibility and trustworthiness of this thesis. The research has been carried out by two researchers, who explored and analysed the findings from individual perspectives.

Furthermore, the primary data was collected from two sources in order to attain consistency in the findings and increase representability: qualitative semi-structured interviews and a quantitative questionnaire. The collected data was combined with a range of theoretical perspectives and secondary data in order to analyse the findings. Utilising more than one source of data through the triangulation method, the multiple researchers, data sources, methods, and theories help in ensuring the credibility of the collected data, analysis, and subsequent interpretation(ibid).