4.8 Validity, Reliability and Ethics
Validity, reliability, and ethics are key concerns for qualitative researchers to conceptualise and guide data collection and analysis.In order to make valid research findings in certain professional fields, rigorous and ethical conduct of a research is needed to produce and transfer the research insights and implications to the readers and other researchers (Merriam, 2009).On the other hand, sufficiently authentic research findings secure the reliability of the approach (Lincoln & Cuba, 2000).Hence, strategies of internal validity, reliability, and external validity (Merriam, 2009) were adopted to enhance the validity and reliability of the present research.
Internal validity assesses the application of research findings in reality.People construct reality according to their own understandings of the world, where the validity of subjectivity-oriented qualitative research is relative because researchers are not able to completely interpret objective reality.In this study, firstly, in order to enhance the internal validity of the present research findings, triangulation of multiple methods and multiple sources of data was used to collect data through a combination of free talks, written documents, video and audio recording of seminars, classroom observations and interviews.Effectively, these different sources enabled me to cross-check data without the constraints of time and place, to address more information based on follow-up data collection, and to interpret and compare the data from multiple viewpoints in reality.Secondly, member checks were conducted with the Chinese student participants during the free talks and two rounds of interviews, which enabled me to avoid qualitative researcher’ bias and misinterpretation of the meanings of the participants’ saying and doing, including their clarifications and comments on certain data interpretation.Thirdly, two years’ engagement in data collection allowed me to be reflexive on the research design.Fourth, my position had impacts on the present research:that is, my role as an insider in the institution, my role for eight years as a lecturer in a Chinese university, and my educational and cultural background as a Chinese in Chinese and UK institutions.As Lincoln and Guba (2000) point out, emphasising the researcher’s roles in the research may provoke the question of ‘human as instrument’ (p.183), but this would also usefully grant a researcher sufficient personal and academic knowledge in carrying out the research.Finally, peer review with academic staff members on data analysis maximised the validity of the analytical process and findings of the research.
Reliability refers to the replicability and consistency of research findings.It is closely associated with the strategies of validity for ensuring credibility and consistency of qualitative research.Replicating research of personal experience doubtfully produces the same research findings because of dynamic human behaviour and various interpretive techniques of analysis, but repeated interpretations and measures of the same data most likely discover more variables and then comprehensively explain a phenomenon in reality.The primary criterion for qualitative research relies on the consistency between data collection and research findings (Lincoln & Guba, 1985; Merriam, 2009):‘if the findings of a study are consistent with the data presented, the study can be considered dependable.’ (Merriam, 2009, p.222).As such, the current study used strategies to rationalise and justify consistency between data collection and research findings.That is, triangulation, researcher position and peer review were used along with an audit trail as another strategy to describe ‘how data were collected, how categories were derived and how decisions were made’ (Merriam, 2009, p.223).To construct the audit trial in this study, a clear description of its path was demonstrated in previous sections, and recording memos were enacted to document and self-reflect on the processes of data collection and analysis.
External validity is related to the generalisability of research findings.Qualitative researchers are interested in understanding particular phenomena in depth, not to statistically generalise truth.By transferring research findings into other situations, they commonly use the strategies of thick description and maximum research variation.On the one hand, a detailed rich description of the research design can help the potential readers or other researchers extrapolate meanings from the information and accumulate them into their own reservoir of knowledge.On the other hand, purposeful selection of research variation maximises the range of application of the research findings by these readers and researchers.
Suggestively, validity and reliability should be aggregated in an ethical manner.Qualitative researchers are ethically encouraged to ‘protect their research participants; develop a trust with them; promote the integrity of research; guard against misconduct and impropriety that might reflect on their organizations or institutions; and cope with new, challenging problems’ (Creswell, 2009, p.87), and to avoid marginalising or disempowering their research participants.For this reason, ethical consideration therefore went into the data collection and analysis in the current study as well.It was done so with the general steps as follows.First, the completed ethical forms were submitted to the Ethics Committee for data collection permission.Second, permission for each data collection procedure was obtained from the individual participants before the implementation as to respect their willingness of participation.Third, the purpose and ethical issues of the research were explicitly conveyed to each participant before the data collection, by attempting to mediate their perspectives and not to disclose any information that may harm them.Fourth, anonymity and confidentiality of all participants have been maintained in order to protect them.