Criticisms of the lack of objectivity and generalizability are often associated with the qualitative method (Phillimore and Goodson, 2004). Extracting and analysing qualitative data is not a straightforward process due to the nature of the method. Furthermore, unlike the analysis of quantitative approach, there is not a clear method to develop the data collected (Bryman, 2012). Due to the nature of qualitative research, many doubts occur during and after the research process. Often researchers question the credibility of collected data and whether enough evidence is gathered to support the claims, or even determine if the accumulated results should be published. These doubts may be minimized through the aspect of trustworthiness. The above-asked questions capture concerns with validity, reliability, objectivity, and generalizability while broadening and deepening them (Marshall and Rossman, 2006).
“A thorough reporting of the process and the results of qualitative data collection and analysis is the key to justifying and assuring that trustworthiness exists in the study” (Henderson, 2006 cited Veal, 2011). According to Veal (2011), Bryman (2012) and Loh (2013) trustworthiness consists of four different components — credibility: the validity of the findings; transferability: the applicability of the findings in other contexts; dependability: reliability of the findings at another time; and confirmability: objectivity of the researcher while carrying out his/her research. The combination of these four terms constitute towards the trustworthiness criteria, thus forming conventional pillars for qualitative methodology (Phillimore and Goodson, 2004). Throughout this paper, the author will elaborate these four frameworks by contrasting them with the analytic induction method of data collection and analysis process of certain qualitative methods.
In terms of the analytic induction method to analyse qualitative data, the author assumes it to be a very tedious and unreliable process to configure data. The analytic induction process begins with an abstract definition of the research aim; proceeding to the hypothetical explanation of the assumed problem; which leads to the collection of data (Bryman, 2012). Depending on the consistency of the hypothesis, this could lead to two procedures: (a) the confirmation of hypothesis — end of data collection (b) to reformulate the hypothesis or exclude the deviant (ibid). When contrasting the various qualitative data collection methods with this data analysis method, the author can see a number of possibilities of trustworthiness being breeched. For instance, throughout the process of analysing the transcripts of in-depth interviews, the researcher could potentially select to discard the interviews if they do not support his/her hypothesis. Moreover, one of the steps of the analytic induction method is to reformulate the hypothesis. The author too questions this phase of the process. The author believes that this already implies alterations being made to the findings in order to better suit what the researcher wants to conclude, thus affecting the credibility factor of trustworthiness as well. Not only does this deem as a short cut, but also the author would suggest that these two steps also do not seem to comply with the ethical standards of research. One suggestion to counter this option would be for researchers to pre-compare other studies that have done similar research. Shenton (2004) and Porter (2007) both suggest that one way to combat credibility is to cross-reference similar strategies used by previous researchers, as that will help to eliminate the possibility of invalid findings. Moreover, by investigating the limitations of other authors in terms of trustworthiness, the researcher would be able to format a more suitable method to collect and analyse data. In addition, other procedures to counter trustworthiness would be to apply random sampling within the suited participant pool, familiarise oneself with the culture of the chosen sample, and feedback from peer researchers (Philimore and Goodson, 2004; Shenton, 2004). By choosing a random sample, the researcher will be able to remain objective, and not only selecting participants that will provide the researcher the results he/she pleases; thus decreasing the liability of confirmability. Also, by immersing oneself with the culture of the sample, the researcher will better understand the epistemology of the participants therefore aiding one to remain objective during data collection and analysis. Lastly, with fellow researchers crosschecking findings, it is possible that the researchers will gain insight to transferability. Often, researchers may stumble across blind spots when being overly focused on their own research; peer feedback might not only help to increase validity of your paper, but recommend your findings in other contexts. The author seems to find the dependability aspect in trustworthiness to be the toughest element to make reliable. Marshall and Rossman (2011) and Bryman (2012) mention this element to be rather problematic for researchers and may be hard to prove. As trends and lifestyles change throughout the years, it is inevitable the current findings may not be applicable in the future. However, if the exact same procedure were to be repeated with the same participants and same data collection method, will researcher be able to obtain similar results? The author would advise researchers to devote a section describing the step-by-step process of the implemented research design in detail, to ensure other researchers can execute the same strategy. Similarly to the suggestion of peer feedback (Phillimore and Goodson, 2004; Shenton, 2004), Porter (2007) also encourages reflective appraisal to strengthen dependability. “Reflective appraisal of the project evaluates the effectiveness of the process inquiry undertaken” (Porter, 2007).
Throughout the writing process of this research topic, the author cannot help but to think that limitations of trustworthiness are unavoidable. Although there are suggestive measures to combat these limitations throughout the data collection and analysis process, the author implies that researchers are only human and are bound to make these mistakes. However, these elements of trustworthiness can be minimised, and other data analysis methods could be better. Moreover, other methods of data analysis such grounded theory has been evolving over time to ensure more reliability in qualitative research. Trustworthiness is a crucial aspect within qualitative research, and should not be taken lightly. The author would advise researchers to also confider other forms of limitations when carrying out their research, in obtain the most reliable findings possible.
References:
Bryman, A. (2012) Social Research Methods. 4th ed. Oxford: Oxford University Press.
Loh, J. (2013) Inquiry into Issues of Trustworthiness and Quality in Narrative Studies: A Perspective. The qualitative report, 18 (65), 1-15. Available from: http://www.nova.edu/ssss/QR/QR18/loh65.pdf [Accessed 28 October 2015].
Marshall, C. and Rossman, G. B. (2006) Designing Qualitative Research. 3rd ed. London: SAGE Publications.
Marshall, C. and Rossman, G. B. (2011) Designing Qualitative Research. 5th ed. Los Angeles: SAGE Publications.
Phillimore, J. and Goodson, L. (2004) Qualitative Research in Tourism. New York: Routledge Taylor & Francis Group.
Porter, S. (2007) Research Methodolody. Belfast: Blackwell Publishing ltd. Available from: http://www.marjee.org/pdfs/validity.pdf [Accessed 30 October 2015].
Shenton, A. K. (2004) Strategies for ensuring trustworthiness in qualitative research projects. Education for Information, 22 (1), 63-75. Available from: http://www.crec.co.uk/docs/Trustworthypaper.pdf [Accessed 3 November 2015].
Veal, A. J. (2011) Research Methods for Leisure and Tourism. 4th ed. England: Pearson Education Limited.
