Validity and Reliability

“Ensuring validity and reliability in qualitative research involves conducting the investigation in an ethical manner” (Merriam 1998 :198).

asks, ‘do the findings capture what is really there?’ Internal validity is defined as “the degree of confidence that the results are true given the study situation” (Davis 2005 :149). External validity on the other hand, is concerned with the extent to which the findings of one study can be applied to other situations and is defined as “the degree to which the study’s results can be generalised across populations, settings and other similar conditions” (Davis 2005 :150).

Demonstrating that research satisfies both internal and external validity requirements is a challenge. Triangulation, by using multiple investigations, multiple sources of data, or multiple methods to confirm emerging findings help to enhance internal validity. External validity is enhanced through rich, in-depth descriptions so that readers can determine whether the findings apply to their situations. Also, the use of several sites, cases and situations, especially those that maximise diversity in the phenomenon of interest could enhance external validity (Merriam 1998).

The multiple method research design of using a preliminary Delphi study, face- to-face in-depth interviews in different sites and the multiple method-participant case studies support the internal and external validity of this research. The overall research validity was supported by the use of combined procedures and the validity of the individual Delphi study and Case study phases is explained below.

According to Bonnemaizon and colleagues (2007), the validity of a Delphi study is based on a rigorous selection of experts whose combined expertise reflects the full scope of the problem and the number of experts. The minimum recommended number

is 5-7 experts, with 8-10 offering the best precision and cost ratio. Going beyond 12 experts makes further information marginal. The method’s validity relies on a strict implementation process, of usually three iterations, to obtain a satisfactory consensus. Using statistical measures of central tendency in subsequent rounds affirms the responses and supports validity and reliability (Dalkey, Rouke, Lewis and Synder 1972; Jackson 2003). These criteria were all satisfied during this phase of this research.

Similarly, using multiple cases helps strengthen the precision, validity and stability for the case study phase. This is a common strategy for enhancing external validity or generalisability of findings (Merriam 1998) because the case studies gain credibility through triangulating the descriptions and interpretations (Stake 2005). Furthermore, valid responses are more likely from the participants as they are interested in and informed about the topic. Protecting anonymity for questionnaire respondents implies this encourages greater truthfulness (Ary et al. 2006).

Reliability refers to the extent to which research findings can be replicated, i.e. if the study is repeated, will it yield the same results? Reliability in Social Sciences is problematic because human behaviour is never static (Merriam 1998); therefore dependability or consistency can describe reliability. Rather than demanding the same results the research is deemed reliable if the results make sense and are consistent and dependable. The question is therefore not whether the findings will be repeated, but whether the results are consistent with the data collected. Collectively, using triangulation through the use of multiple methods of data collection and analysis strengthens reliability (Merriam 1998) and suggests the research is reliable.

3.5 Conclusion

The chapter has described the methodologies employed in the research and provided justification that the data gathering and analysis has been thorough. The choice of the constructivist paradigm was made so the collective views of the respondents are used to construct a picture of their reality. The research design called for a multi-method and multi-participant strategy, which entailed a preliminary Delphi study and Case study approaches, using multiple data gathering procedures, including face-to-face in-depth interviews, electronic in-depth interviews and an electronic survey.

The choice of multiple categories of participants, including external school stakeholders (directors / officials of school governing agencies and academics) and internal stakeholders (principals, school human resources practitioners and both serving and retired school teachers), was explained. Multiple methods were used to aid triangulation, which helps secure an in-depth understanding of the subject and enhances the validity and reliability of the study.

The approach to the analysis was explained. To deal with the large amount of qualitative material gathered, the data was recorded, analysed and interpreted using methods in accordance with the Murdoch University Code of Conduct for Research and other ethical considerations. The next chapter contains detailed analysis of the data and the resulting findings from the research.

CHAPTER FOUR: FINDINGS

4.1 Introduction

The previous chapter discussed the methodological approaches, data gathering processes and analytical procedures employed to answer the two main research questions, which respectively address the issues of strategic human resource management (SHRM) implementation in the school sector and the influence of SHRM in improving teacher attraction and retention. This chapter details the analysis of the three-round Delphi study and the Cases Studies. The two central research questions are:

1. How effectively is strategic human resource management (SHRM)

implemented in the WA school sector currently?

2. Are the human resource management functions (appraisal and performance

management, professional development, recruitment and selection, and reward) aligned in a SHRM approach and what SHRM strategies could be implemented to improve attraction and retention of teachers in WA?

The Delphi study was used to provide preliminary answers to both research questions. The first question was further explored via the multi-site case study using face-to-face in-depth interviews of principals and human resource managers. The second question was addressed in the multi-method case study, which included the ten multi-site interviews, an electronic in-depth interview of retired teachers and an electronic survey of serving schoolteachers. The analysis of the data generated from each phase of the study are presented in sequence, with the analysis of the Delphi coming first, followed by the Case studies A and then B.