My Doctorate in Business Administration Resource Blog
I using an online Likert scale questionnaire to collect the data. The population from whom the sample is being selected consists of 180 staff. The sample will be randomly selected and will consist of a 10% sample.
The each member of the population will be randomly allocated by a computer macro a number between 1 and 180. The select adidas flux ion of the sample will be made from another macro that will generate 18 random numbers between 1 and 180. The output numbers will be matched to the main population list.
People selected will be sent an email by email mail merge. The email will contain a link to the online survey. The aim is to gather a minimum of 9 10 responses for the analysis part of the assignment.
The main challenges at the present time (AKA the beginning) are:
Making sure the computer macro work and generate the random numbers.
That the selection of staff from the randomly generated list are available to complete the survey.
That enough people will respond to the request to complete the survey.
That I am clear in my own head what the outcome of the survey is going to be.
I am looking at Factor analysis as a possible analysis technique (Field, 2013).
I am planning on getting the survey completed and out to recipients the week beginning 17th March.
Field. A, (2013) Discovering Statistics Using IBM SPSS Statistics, Sage, Los Angeles, USA. What I like most about it are the issues of higher validity and reliability, the focus on taking small amounts of notes rather than making full transcripts and the fact that it can be implemented as a project in its own right.
The projec adidas flux t aspect is quite interesting in terms of getting the Reference Group together which for me is similar to having a Project Board as part of a PRINCE2 or similar project process. Organizing the Reference Group will I think be an interesting in terms of challenging process.
The other interesting or challenging issue would be deciding on whether to use (at least) one other interviewer. On large projects CI tends to make use of more than one interviewer and in so adidas flux me of the papers I looked at up to six. Although there is always the option of carrying out all the interviews myself the big advantage is using one other interviewer would enhance the reliability of the research by allowing cross referencing and reducing bias.
Having another interviewer would add an interesting dynamic to the process of the research project. How to recruit someone to work with is an interesting problem. One thought would be to get the help of a Masters student who might be working on a similar project and to use them as a research assistant. There are possible opportunities for this in the social sciences department. This would add an interesting additional challenge to the ethical approval process.
With CI the use of two interviewers lends reliability and reduces bias through running interviews in parallel and then immediately after the interviews have finished the interviewers immediately compare notes and decide what the key issues or themes are. From these the next round of interviews are developed.
Sharing ideas and experiences I believe is a good way of building a creative dynamic throughout the interview process. The biggest issue that I found with doing just two interviews was typing up the transcripts from the voice recording. The first interview I did was 37 minutes long and took about 2 3 hours to transcribe. The second was only 15 minutes long but took about an hour to transcribe. The interesting thing about transcribing that I found was making sure that I did not add or remove any words. I think dealing with interviewing 20 or 30 people would require a serious amount of organization of the material and careful planning prior to starting the work. As already stated CI technically does not make use of long transcriptions of the interviews. In fact according to Dick voice recorders are not recommended. Dick recommends only brief note taking or using mnemonics to identify themes and issues during the interview and having only a page of notes at most.
To me this implies that there is a need for quite a disciplined approach to the process. The key issue for CI is ensuring that the initial interview question is further extended and more focused in order to gather information on more specific issues. For the assignment I only focused on one of the three theories that I am aiming to use in my full research project. These are, , Institutional Theory and Organizational Justice. For the assignment I only referred to Organizational Justice but after the interviews I carried out I believe that all three theories are relevant to the thesis at least as drivers into a direction for the project. CI is normally or often used in areas where there is little existing theory or where an area is under researched.
For the research project I believe that CI has the potential to lead to new theory development or an extension of an existing theory.
I asked the interviewee that I used last week if I could re interview her as part of the Convergent Interviewing process. The reply I got was interesting, basically along the lines of interviewed me last week why do it again? got me thinking about the difficulties of interviewing and why people volunteer to be interviewed. The question is, why do people agree to take part in interviews or, what in it for them?
A number of points and questions arose from this thought:
Getting interviewee participants is not easy. There has to be a benefit to the interviewee otherwise why would they agree to participate?
What is the incentive for the interviewee participant? I think there is a very narrow and short lived engagement process that occurs in the mind of a potential participant. Possibly the potential participant initial thought is based on a feeling that someone is interested in what they have got to say about something although I not sure what this says about randomly selected participants. Why the hell would a randomly selected person want to participate in an data collection exercise? Surely there even less motivation for them to participate?
I can see how purposeful selection could lead to greater engagement but again there has to be some kind of incentive for the participant but I assuming if someone is told that they have been selected on the basis of their expertise then they might be more forthcoming.
The potential difficulties of getting interview participants and getting people to agree to more than one or two interviews is potentially a difficulty if the researcher is undertaking Convergent Interviewing or just wanting to go back to speak to people again. There are also issues around timing, availability, getting a room and also if you are going back mult adidas flux iple times to people issues around confidentiality might come up because of something as simple as people going off for meetings.
When planning interviews and selecting participants I think depending on the number of people you need to interview it would be good to have more people to interview than absolutely necessary in order to keep numbers up in case some participants decide not to turn up or walk out during an interview.
In some ways small numbers of participants might be easier to manage than large numbers because you can build a closer relationship with potential interviewees and they might be more likely to make a commitment and a contribution.