From what you describe, the use of mixed methods seems feasible and sensible to me. But I think that there are two issues to keep in mind. First, the field of mixed methods has developed quite rapidly into a subfield of methodological debate in its own right over the past few years. I think that it is necessary to take that into account and "link up" with current debates. For example, it has become standard to characterize research designs using one of the many mm-typologies out there (e.g. Creswell and Plano-Clark). Second, regarding later publication, one needs to keep in mind that there are still many journals that have very particular ideas about what makes relevant and credible research. So I would double-check whether there are any journals in your field of research that publish mixed methods.

Kenneth provided excellent advice on mixed methods. It seems that the publication bias is toward quantitative studies although this may be changing. Apart from methodological issues, try to focus your study for timely completion. My outside advisor recommended that my study be reduced by 80%. She said that getting the study done was the most important thing. It turned out to be good advice. My study won an award and yielded several top-tier publications. Your topic is timely and has considerable practical value. Good luck with your research...Nick

My first question would be the purpose of conducting separate quantitative study on the users, as opposed to doing qualitative interviews with them as well as with the managers. In other words, what would you gain by shifting into a qualitative mode for that set of participants?

My second question would be about the feasibility of such a survey. In particular, is the sample size large enough to produce sufficient data for statistical analysis?

In general, it seem like your design could be expressed as some kind of concurrent data collection, such as Quant + Qual, where each serves a separate purpose within the overall project. From a mixed methods point of view, you would have a relatively weak version of this design because there would be little opportunity for "integration," which is a key feature of mixed methods. Instead, this seems to be mostly a matter of convenience, where the qualitative methods are used for the smaller of two sets of participants.

A very different design would begin with qualitative interviews of the managers that generated hypotheses to be tested with a survey of the users, which could be represented as qual --> QUANT.

What are your research questions?

It depends on your research questions and what you are aiming to learn. That being said, I would not use one method for one group and another for the other group. I would expect that users would have opinions they might like to discuss in more detail, which will offer you unique insights into the way that e-tickets are used and thought of. If your aim is to understand what people think of e-tickets and how effective they are, i.e. how successful the project was, then you need detail from both sides, not only the managers.

While I think mixed methods is a good idea, I think what you propose misses out on a key element as well as an opportunity to learn more (i.e. about user opinions). I would encourage you to consider interviewing BOTH groups, not only one.

Dear Jone

As other colleagues have already indicated, it is very important to clarify the objectives of your study. The hallmark of mixed methods is to produce a whole that synergistically integrates the component parts (qualitative and quantitative components). At the moment I see that your study is multimethod, with two separate components, one qualitative oriented to the managers and another quantitative oriented to the users.

Regards.

Dear

For me it depends upon what sort of dimensions you want to explore in that particular industry. But i will refer you that if your plan is to build a new theory for the e-ticketing while analyzing the old theories, that would be nice approach to apply the mixed method. But for validating the old theories, through qualitative approach would have a lot of issues in the methodological ground. (e.g one way you will be applying the positivism, post positivism for qualitative and on the other hand you will apply interpretevism these are totally anti). So applying the mixed method you would have to be very much conscious.

Regards

What are the objectives of your study or rather what do you want to achieve? your objectives would somehow guide us to understand the best method that would be used to collect data that will be meaningful to your analysis and subsequent research findings.

We also need to keep in mind your sample size as it can also determine the best method to use.

I will need more information about your study to determine whether a mixed method approach would be appropriate. In this regard, a clear definition of your research question/s supported by your philosophical commitments, i.e. ontology and epistemology, your tentative methodology as well as methods and techniques.

Apologies for I will piggyback on Jone's question here because his methodology is similar to the one I am proposing to use for my own research.

There seems to be little support for the mixed methods approach. Marks allocation would be an issue of concern too.

The difference with my study is that I propose to use focus groups, interviews and questionnaires equally to both the service (microfinance institutions) users and the service givers. I will be looking at vast amount of data for analysis when I have it all gathered.

Would a mixed methods approach weaken a research study? Thanks for your insights.

Leah,

I don't think you would have any problem justifying mixed methods in Ireland -- although there are some parts of the world that are still behind in accepting that approach.

I would, however, worry about collecting so much data without a coherent design. In particular, the usual mixed methods design proposes some kind of "division of labor" where the specific tasks assigned to the qualitative and quantitative methods match the strengths of those different methods.

Thank you David for your answer. I have justified my methodology and gotten no concerns about it but reading the comments above concerned me. Yes, you are right the mixed methods approach is acceptable here and there is literature available to support it. Thanks