Discussion of Quantitative Empirical Data

Firstly, when discussing the analysis of the quantitative data, the reliability and validity of the survey must be discussed. The reliability of the survey can be questioned, as there has been no control over the distribution of the survey to the employees in the respective SMEs. It would therefore be possible for the respondent from the interview to intentionally select preferred employees whom the respondent knows will provide positive answers to the statements presented in the survey. However, when looking at the percentage of the employees who are participating in the survey, the average is almost a third of the total amount of employees. Therefore it can be discussed whether it is possible for the respondent to find a third of the employees whom the respondent knows will provide positive answers to the statements listed in the survey. In the discussion of reliability from this perspective, the answers from the surveys can therefore be argued to be relatively reliable. When discussing the reliability in terms of measuring and analysing the data, the tool for analysis, provided by SurveyMonkey (Surveymonkey.com), can be argued to be quite reliable due to the algorithms used for analysis on the website compared to personally analysing the data. When analysing the quantitative data though, the reliability can be discussed as the analysis is based on subjective interpretation of the data.

Here the notion of social constructionism must be taken into consideration.

As also mentioned in section 3.7 about social constructionism, the task of the researcher is to acknowledge their own intrinsic involvement in the research process and hereby reflect on how this might affect the findings in the research. With this in mind, the quantitative data collected have been interpreted based on the intrinsic involvement in the research. However, as the data is of quantitative character, it is difficult to question the data quantitative itself. Furthermore - regarding the reliability of the survey - the

67 very same survey has been distributed to all the respondents, and therefore the survey here can be argued to be reliable.

The validity of the survey can be questioned too, as innovation culture as a phenomenon is difficult to measure empirically. The valuation of whether an empirically grounded indicator truly represents the non-observative phenomenon, in this case innovation culture, implies the validation problem in this context (Møller & Hvid, 2018). However, the comparative case study is based on the research design called mixed methods, and both quantitative and qualitative research methods are used to understand the problem, as presented in the research question and thereby it can be discussed how big the validation problem is in the research.

When further discussing the empirical findings. It can be discussed whether the statements in the survey and the questions in the interviews really addresses that or the characteristics of a phenomenon that one would like to investigate (Kristensen & Hussain, 2019, p. 224). Here again, the use of mixed methods can be argued to address this issue. With the use of both methods, the research does not have to rely entirely on the data collected from one method or another, as it is supported by the other method for collecting empirical data.

The comparative case strategy must also be addressed in this discussion, as it can be discussed whether or not the cases fit the category most different cases, same outcome. When presented with the conclusion of the number of employees and the conclusions on Q1, Q2 and Q3 in the Comparative Case Study Matrix (Appendix 4), the cases can appear to be similar rather than different. However, when including the employee score, the culture and core competencies in the respective cases, these can be argued to be different, and the cases can therefore be argued to be most different cases, same outcome.

The amount of employees indicates that all of the companies can be defined as SMEs. The conclusion on Q1 indicates that the SMEs started their business activities relatively recently. The conclusion on Q2 indicates that all of the SMEs in the cases are self-owned companies, and the conclusion on Q3

68 indicates that the SMEs are selling their products or services on mainly a national market. All the questions and conclusions can be found in Appendix 4.

Based on the quantitative data, on whether a specific culture is tried to be created in the respective SMEs can be discussed, as the statement in the survey (“The company is innovative”) is not comparable with Q4 (“Are you trying to create a specific culture in the company?”) in the interview.

Therefore, the quantitative data is not relevant regarding whether or not a specific culture is created.

Regarding Q5 in the Comparative Case Study Matrix (Appendix 4), the interpretation of the analysed data can be discussed. The average employee score in Q5, on the statement “The company has an innovative culture”, is 4.17 and when analysing this, the culture in the SMEs must generally be identified as innovative. However, it can here be discussed whether the statement addresses the characteristics of innovation culture, meaning that the employees might have different perceptions of an innovation culture and therefore the validation of data presented from this statement implies a validation problem.

In Q8, the statement presented (“I know the company's overall strategy and my tasks are a part of that strategy”) concerns both the overall strategy of the company and the employee’s work tasks. The question in the interview (“Do the company has an overall strategy as you know of?”) however, is only concerned with the overall strategy of the company and not the employee’s work tasks. It can therefore be discussed how valid the quantitative data are regarding the analysis of the knowledge of the overall strategy in the company.

In Q9, the validity of the data can again be discussed when it is compared to the question in the interview concerning the same issue of changes in the market. The statement identifies whether or not the employees perceive the company as adaptable to changes in the market, whereas the question in the interview identifies how the company adapt to changes in the market and therefore the statement in the survey and the question in the interview investigates two different aspects regarding changes in the market.

69 In Q14 (“Does the employees choose their own work tasks?”) the case is the same as in Q8. However, in Q14, the question in the interview only focuses on the employee’s work tasks, whereas the statement in the survey focus both on the employee’s work tasks as well as the company’s overall strategy.

In both Q18 and Q19, the statement in the survey and the question in the interview are formulated, so a comparison is possible. Again the perception of internally and externally innovation and the perception of a systematic approach can be different depending on the respondent.

With the discussion of the quantitative data collected by the conducted surveys, in order to compare with the qualitative data collected from the conducted interviews and examine innovation culture in the studied SMEs, it becomes clear that the validity of the surveys can be discussed. However, the intention with the quantitative data has throughout the research been to use it as a support to the qualitative data, and the quantitative data therefore can be argued to provide useful support of the qualitative data.