Susanne Friese <[log in to unmask]> wrote
>SdG Associates wrote:
>
>> Re: Sarah's point about working under a deadline
>>
>> Yes this is a real problem.  One issue is the need to educate funders
that
>> the analysis stage for qualitative data requires more time and money.
>
>You forget one other world - that is the commercial world where the turn
around
>always has to be quickly. You would get nowhere if you limit yourself to
one
>hour coding a day.
>
>Also for academic work, I think everyone has to figure out what their time
limit
>is. It could take a person just about an hour to get into coding and
stopping
>after one hour may  just lead to inconsistent coding, one may loose the
flow and
>by feeling that one is not making any progress (how far can one really get
in
>one hour?), one could also loose motivation.
>
>
>>  a) you identify the main themes you want to look at
>> b) divide your work up so that you say, this week I will look at Main
Theme
>> A, next week at Main Theme B, etc.
>> c) code just for that main theme - initially broadly - so if you are
looking
>> at conflict, you will code everything for conflict  This makes coding on
>> screen
>> easier because the initial coding decision is broad so you are not
weighing
>> up very fine distinctions between codes yet.
>
>If you however do this throughout your entire documents, then you create a
lot
>of work for yourself again. I like to follow the advise of Strauss und
Corbin
>here - to develop a coding schema based on the first few documents (at
least
>this is my modified version of it). Thus, it can be is ok to start out with
>broad categories because the sub- or related categries are not immediately
>visible/obvious. But maybe it is a good idea to begin more intensive work
after
>one has broadly coded 2 to 4 documents and try to get a bit deeper before
>applying the coding schema to the rest of the documents. This way one
avoids
>having to go through all documents again and again to define the smaller
>categories (of course one could also argue that this is necessary to get
deeper
>and deeper emerged into the text and analysis - however, if one has to work
>towards a tight deadline, this is not always feasible and one has to adopt
a
>more pragmatic approach).
>

I come rather late to this discussion - I rather agree with Susanne's
remarks above - also I feel that although Silvana's strategy to  say Code
main theme A this week and main theme B next week will sound appealing to
many people - it rather assumes that the identification of 'MAIN THEMES'
arrives at the beginning of analaysis, rather than in the middle or even
towards the end - and therefore it presumes a particular, and quite
deductive approach to analysis.

I REALLY feel strongly that this style of coding strategy makes sense to
particular users ONLY and its certainly one of the strategies that coudl be
put forward in a class to overcome dilemmas ( and I use Silvana's strategy
as well, at certain points in my classes, and I like the way she describes
it and I also I can see it works well in the workshop scenario). However,
one of the main criticisms of using particular packages is that overplayed
notion that each software is tying you down to a certain approach to
coding - a deductive or an inductive approach - of course this is nonsense.
But it starts to be NOT nonsense if we assume that the strategy Silvana
describes is the way everyone should start coding and that this is the way
they should be taught to analyse data.

This also plays on the issue that a few  users may  come to the software
expecting to learn from it a methodology and thats not the ideal way to
develop a method.   Silvana, we both saw a bit of this when we did that
research on workshop attendees didn't we?  So I feel teachers of the
software must be aware of this, and not assume that there is one dominant
useful approach to coding which is THE way to do it, and the way that
everyone shoudl do it, and the way that everyone shoudl be taught to do it.
Otherwise we rather let the software be the master, if we tend to do it
because the software really helps us do it that way .  Of course I know
Silvans's response was just in reply to the plea for help from the 'all
coded out' user, and I absolutely agree this strategy is ONE quite nice way
to work.
Maybe I am being deliberately contentious.
cheers
Ann Lewins
Resource Officer, CAQDAS Networking Project
Dept of Sociology
University of Surrey
GUILDFORD  GU2 5XH
email:   [log in to unmask]
CAQDAS web site:  http://www.soc.surrey.ac.uk/caqdas/
Tel +44 (0)1 483 259 455
Fax +44 (0)1 483 259 551
























%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%