3 RESEARCH METHODOLOGY
3.4 Validity and reliability of the research
Well-conducted, quality studies are an important part of research rigor. For instance, following the procedures of the research alone is not sufficient to produce reliable results (Murphy & Dingwall, 2007). Rigor is the extent to which a researcher keeps in mind the quality of the studies. Validity and reliability help a researcher to achieve research rigor (Heale & Twycross, 2015); therefore, these two elements demonstrates and communicates the trustworthiness of the research findings (Roberts, Priest, & Traynor, 2006). It is important to consider the significance of the validity and reliability of the data, tools, and instruments to uphold the research quality and trustworthiness. Accordingly, the present dissertation considers validity and reliability as important aspects of the research conducted in individual essays, where the validity and reliability procedures were carefully followed.
In quantitative studies, validity refers to the extent to which a concept is accurately measured. Validity has two further sub-dimensions, namely, external validity and internal validity. External validity refers to the ability to apply the findings of the study with confidence to other people or situations, and it ensures that the conditions under which the study has been carried out represent the time and situation to which the results relate (Calder, Phillips, & Tybout, 1982). The study sample must be representative of the population at the time of the research was conducted. Accordingly, the representative sample of people with reference to variables like their age, gender, and so on should be drawn from that population of interest. Although care has been taken to uphold the external validity in this research, on average, in most of the data collected in this study, gender and
education levels are overrepresented in both the Finnish and Pakistani populations. Specifically, the sample is skewed toward younger people and those with bachelor’s degrees. It should be noted that, in consumer research, samples of student consumers are widely used as a respondent source (Ashraf & Merunka, 2016). This study does not make claims about a population, but rather, it tests new theoretical ideas regarding sustainable consumption motives phenomenon embedded in consumers’ HI versus VC cultural values, using robust research methods.
Internal validity refers to the extent to which a research study has been correctly performed. The effects observed in the dependent variable are due to the manipulation of the independent variables and not other factors, which proves the internal validity of this thesis. Internal validity has three sub-dimensions (Punch, 1998). First, content validity relates to the relevance of the questionnaire to the intended settings. This validity can be achieved if a pilot study is conducted with people who are similar to the intended respondents of the research. Second, criterion-related validity involves comparing the questionnaire with the same or similar validated measures of the same concept. Third, construct validity that relates to demonstrating the relationship between the concepts measuring a phenomenon, for example, the relationship of the independent variables with the dependent variable. In quantitative studies, construct validity can be determined using the factor analysis statistical procedure. The content validity of measures in this study was achieved by pretesting the questionnaire among the Finnish and Pakistani respondents individually. For criterion validity, the questionnaire was compared with similar questionnaires. Since the questionnaires in all the essays were adopted from earlier studies, the criterion validity of the studies was not a problem. Moreover, the correlation tests using SPSS 20.0 and the discriminate validity test demonstrated the robustness of the criterion validity. Finally, to confirm the construct validity we followed convergence, homogeneity, and theory evidence criteria. For convergence, we used the convergent validity test of the instrument in AMOS 7.0 and SmartPLS 3.2.6 on the given constructs. The homogeneity and theory evidence was not a problem because the questionnaires successfully measured each construct, and they were adopted from earlier research with similar theoretical arguments and evidence.
Reliability refers to the accuracy of an instrument. It refers to the extent to which a research instrument gives same results repeatedly with different populations. In quantitative research, reliability can be tested in different ways (Heale & Twycross, 2015). For this purpose, the internal consistency of the measurement instruments/tools needs to be assessed. For example, this can be done by evaluating homogeneity or internal consistency, which refers to the extent to which all items in the scale measure the same construct. The stability of the instrument can be determined by assessing the constancy of the instrument with repeated testing. Moreover, the equivalence consistency of the instrument among the responses of multiple users of an instrument can be tested. In quantitative research, typically, the Cronbach’s alpha coefficient statistical procedure is used
for checking the internal consistency of the measures. A value of alpha closer to 1.0 is considered to have higher internal consistency (Bonett & Wright, 2015).
Researchers also use the composite reliability (CR) test of statistics to measure the instruments’ reliability (Fornell & Larcker, 1981). CR considers the varying factor loadings of the items of the scale measuring a construct. Accordingly, CR values were used to measure the internal consistency of the instruments employed in the essays of this dissertation. The values of CR of the instruments obtained were reasonable thus indicated acceptable reliability.