This is an observation that comes from my experience of working on projects and interviewing project managers. Often people say they want feedback from stakeholders. Preferably in a formative sense so that feedback can be reviewed and enacted upon within the project if necessary. Often people choose to use a project wiki and publicise this in order to invite comments. I wonder how effective this is. Will/do people really comment and give feedback in such a public space?
Others often quote final reports and journal articles as effective means of dissemination but also as channels for feedback. Such summative methods are unlikely to provide useful feedback that can be incorporated unless the project has a second phase perhaps either through a new grant or other means.
Friday, April 25, 2014
Thursday, April 24, 2014
Coding
Just finished reading a great book by Jonny Saldana, The coding manual for qualitative research (2013, SAGE). I've learned a lot of things, here's a summary.
I should have started coding as I was collecting my data rather than waiting till the end of the data collection period which was over 18mths.
The value in analytic memos should not be underestimated. Another way of saying this is reflective comments which I guess is what I did but next time I would formalise these and use them as sources of evidence for coding.
There are so many ways to code! I've tried to identify methods that may be suitable for my study and these include structural, descriptive, In Vivo, values, provisional, hypothesis, landscaping, focused and elaborative coding.
I have begun first cycle coding using Initial Coding (Charmaz, 2006) as a starting point. I'll then look at the codes generated and decide which other method of coding may be more appropriate for the next pass at the data.
During the first cycle coding process I have added many observer comments to the transcripts which I will use as a further data source.
I've so far resisted using CAQDAS or computer aided qualitative data analysis software ie NVivo. Mainly because it does not run on Mac and I don't have a windows computer. I have tried Leximancer to do a first pass on the data without much success. May come back to this and use as a triangulation method. This book has outlined other methods that are manual but make use of excel and word to manipulate and organise data. I'm going to manually write codes on the transcripts and then use these programs to arrange codes and data.
I need to document some of this process that i end up using and include in dissertation as a chapter. Only one or two paras would make it into a journal paper.
When I look at the artifacts such as reports and applications, critically reflect on them as they "reflect the interests and perspectives of their authors" (Hammersley & Atkinson, 2007 as cited in Saldana, 2013, p.54). They can also carry "values" (Hitchcock and Hughes, 1995 as cited in Saldana, 2013, p.54)
other notes:
Simultaneous Coding: "socail interaction does not occur in neat, isolated units" (Glesne, 2011, p.192).
Structural Coding: useful for hyposthesis testing (amongst other applications) - could be phase 2 aim..
Descriptive Coding: summarise in a short sentence or few words the topic of a paragraph of data. Basic method
In Vivo Coding: applicable to action and practitioner research (Coghlan & Branick, 2010; Fox, Martin & Green, 2007; Stringer, 1999)
Process Coding: useful for studies searching for "ongoing/interaction/emotion taken in response to situations or problems often with the purpose of reaching a goal or handling a problem" (Corbin & Strauss, 2008, pp96-97)
Values Coding: reflects participant's values, attitudes and beliefs representing her or his perspectives or world-view. One application is to explore experiences and actions in case studies. Must remember though that "values coding is values laden" (p.114)
Provisional Coding: builds on or corroborates previous research and investigations.
Causation Coding: Attribution refers to reasons or causal explanations. An attribution answers the 'why?' question. We should carefully consider the nuanced differences between a cause and a reason and a motive and to keep our focus primarily on people's intentions, choices, objectives, vakues, perspectives, expectations, needs, desires and agency within their particular contexts and circumstances (Morrison, 2009).
Finally, "coding is not a precise science it is primarily an interpretive act " (p.193)
From codes to themes -
I should have started coding as I was collecting my data rather than waiting till the end of the data collection period which was over 18mths.
The value in analytic memos should not be underestimated. Another way of saying this is reflective comments which I guess is what I did but next time I would formalise these and use them as sources of evidence for coding.
There are so many ways to code! I've tried to identify methods that may be suitable for my study and these include structural, descriptive, In Vivo, values, provisional, hypothesis, landscaping, focused and elaborative coding.
I have begun first cycle coding using Initial Coding (Charmaz, 2006) as a starting point. I'll then look at the codes generated and decide which other method of coding may be more appropriate for the next pass at the data.
During the first cycle coding process I have added many observer comments to the transcripts which I will use as a further data source.
I've so far resisted using CAQDAS or computer aided qualitative data analysis software ie NVivo. Mainly because it does not run on Mac and I don't have a windows computer. I have tried Leximancer to do a first pass on the data without much success. May come back to this and use as a triangulation method. This book has outlined other methods that are manual but make use of excel and word to manipulate and organise data. I'm going to manually write codes on the transcripts and then use these programs to arrange codes and data.
I need to document some of this process that i end up using and include in dissertation as a chapter. Only one or two paras would make it into a journal paper.
When I look at the artifacts such as reports and applications, critically reflect on them as they "reflect the interests and perspectives of their authors" (Hammersley & Atkinson, 2007 as cited in Saldana, 2013, p.54). They can also carry "values" (Hitchcock and Hughes, 1995 as cited in Saldana, 2013, p.54)
other notes:
Simultaneous Coding: "socail interaction does not occur in neat, isolated units" (Glesne, 2011, p.192).
Structural Coding: useful for hyposthesis testing (amongst other applications) - could be phase 2 aim..
Descriptive Coding: summarise in a short sentence or few words the topic of a paragraph of data. Basic method
In Vivo Coding: applicable to action and practitioner research (Coghlan & Branick, 2010; Fox, Martin & Green, 2007; Stringer, 1999)
Process Coding: useful for studies searching for "ongoing/interaction/emotion taken in response to situations or problems often with the purpose of reaching a goal or handling a problem" (Corbin & Strauss, 2008, pp96-97)
Values Coding: reflects participant's values, attitudes and beliefs representing her or his perspectives or world-view. One application is to explore experiences and actions in case studies. Must remember though that "values coding is values laden" (p.114)
Provisional Coding: builds on or corroborates previous research and investigations.
Causation Coding: Attribution refers to reasons or causal explanations. An attribution answers the 'why?' question. We should carefully consider the nuanced differences between a cause and a reason and a motive and to keep our focus primarily on people's intentions, choices, objectives, vakues, perspectives, expectations, needs, desires and agency within their particular contexts and circumstances (Morrison, 2009).
Finally, "coding is not a precise science it is primarily an interpretive act " (p.193)
From codes to themes -
Subscribe to:
Posts (Atom)