It is argued that the concept of validity and reliability are multi-faceted and a touchstone of all types of educational research [Cohen et al., 2005, Hammersley 1993, Patton 1990]. And
Sapsford and Jupp, [1996] argue, ‗As with other research data, we must always be concerned about the validity and reliability of observations‘ [p.88].
Tomlinson [1989] contends that major problems to the validity of data gathered from research interviews are presented by:
The active and idiosyncrasy of human understanding and language and; The potential social influence of the interviewer.
Thus, in pursuit of the researcher‘s own interests they may miss the interviewee‘s construal and
reactions particularly through questionnaire use. Then again, over emphasis on the
interviewee‘s perspectives may fail to do justice to the researcher‘s own agenda. As such, it is
advantageous and an issue of validity to do both [Tomlinson 1989].
Reliability deals with the consistency of a measure across different testing externally or within itself internally. It is the extent to which the instrument yields similar results at all times when used under constant conditions [Bell, 1999]. Validity on the other hand deals with the question of whether the measuring instrument measures what it was originally intended to measure [Cohen et al., 2005].
Conversely, Hammersley [1993] argues, ‗consideration of validity takes us to the interface
between the findings of the study and the reality from which they were extracted, and,
essentially about truth conditions‘ [p. 178]. As such; validity becomes an important ingredient
and a key to effective research. For Patton, [1990], ‗the validity and reliability of qualitative data depend to a great extent on the methodological skill sensitivity and integrity of the
researcher‘ [p. 11]. However, Cohen et al [2005] contend that reliability is necessary but
insufficient condition for validity in research; were as, reliability is a necessary pre-condition of validity. Nonetheless, for [Nuttall, 1989; Patton; 1990, Dennis, 2001], there are several
different kinds of validity. ‗If a test achieves what the originators intended it to achieve, it is a
valid test, it ought to be noted that there is a connection between the validity and the reliability
of a test or examination paper‘ [Dennis, 2001, p. 366].
Nevertheless, Cohen et al., [2005] argue that threats to validity and reliability cannot be
eliminated completely, ‗rather the effects of these threats can be attenuated by attention to validity and reliability throughout a piece of research‘ [p.105], and ‗there will be several ways in which they can be addressed‘ [p. 104]. For Maxwell [1992], qualitative researchers need to
be cautious not to be working within the agenda of the positivists in arguing for the need for research to demonstrate concurrent, predictive and convergent, criterion related, internal and external validity. Rather, Maxwell [ibid] argues for the notion of ‗authenticity‘ while Mishler
[1990] also cited by Cohen et al., [2005] prefers using ‗understanding‘ as a more suitable term
than validity in qualitative research. Nonetheless, to me, it all boils down to a question of semantics since authenticity and understanding appear to be sides of the same coin. For Hammersley [1992], validity in qualitative research replaces certainty with confidence in our results and that, as reality is independent of the claims made for it by researchers; our accounts will only be representations of it [pp. 50-1].
To account for validity, Cohen et al., [2000] contend that in qualitative data, validity could be addressed through the honesty, depth, richness and scope of the data achieved, the participants approached, the extent of triangulation and the disinterestedness of objectivity of the researcher. Well, in this research, I attempted to improve validity through careful sampling, appropriate instrumentation and appropriate statistical treatments of the data. Cohen et al [2000] postulate that, in qualitative data the subjectivity of respondents, their opinions,
attitudes and perspectives together contribute to a degree of bias, ‗validity then should be seen
4.7.1. Internal Validity
I strove to account for internal validity by taking heed of some of the advice espoused by [Merriam 1988, LeCompte and Preissle 1993]. Internal validity was accounted for by providing explanations of particular events, issue or sets of data which attempted to actually sustain the data i.e. [accuracy of the description of the phenomena being researched]. LeCompte and Preissle [1993] argue that in qualitative research, internal validity can be addressed in several ways by using, low inference descriptions, multiple researchers, participant researchers, peer examination of data and using mechanical means to record store and retrieve data [p. 338]. Nonetheless, LeCompte [p. 323-4] argue, ‗in ethnographic qualitative research there are several overriding kinds of internal validity. These include: confidence in the data; the authenticity, cogency, soundness of the research design, credibility, auditability, dependability and confirmability of the data‘ [p. 108].
4.7.2. External Validity
External validity, as Cohen et al., [2005] argue; refers to the degree which the results can be generalized to the wider population, cases or situations, were as Bryman [p.30] writes,
‗external validity is concerned with the question of whether the results of a study can be
generalized beyond the specific research context. And that the issues of generalization can be
problematic as, Eisenhart and Howe [1992, p, 647] cited by Cohen et al., [2005] postulate, ‗ it
is possible to access the typicality of a situation, the participant and settings to identify possible comparison groups and to indicate how data might translate into different settings and cultures [p. 109]. The validity and reliability issues will be closely discussed in relation to the analysis of data under appropriate sections.