Hi Caroline,
I do agree with Chris on this. I guess it depends on the method. I did
research in three languages and looked into the issue of translating and
back translating. Most valuable in multi- language research is to translate
as little as possible, if possible only for writing the article or thesis.
In grounded theory ( strauss & corbin 1998), this is advised. I read another
article from Norway (can^t remember anymore, but I thought it was in Social
Science and Medicine), telling about a study in which every interview was
translated and they found out later, that they had rather not done this.
Exactly as Chris said, because of interpretational and conceptual reasons;
this is not possible without integration of the context.
A colleague of mine, who did her PhD using two languages, confirmed problems
with back translation with questionnaires.
This issue therefore sets limits to positivist multi-language research, in
which this would be a great source of bias, and favours interpretivist
research. Of course, the issue has to be understood by institutions as well
as supervisors.
Best wishes,
Ans
-----Oorspronkelijk bericht-----
Van: A forum for discussion on midwifery and reproductive health
research. [mailto:[log in to unmask]]Namens Chris McCourt
Verzonden: Freitag, 30. November 2007 11:04
Aan: [log in to unmask]
Onderwerp: Re: Back translation: any guidelines?
hi Caroline
I have used back translation in the past to check the validity of
translations and found it really valuable. In our case, though, the
situation was different as the translator employed lacked in-depth
knowledge of the subject, so we found although the translation was
technically correct it was conceptually incorrect - very badly! What I
learnt was that back-translation is a good strategy to highlight
problems - so i would say to take a few samples of text makes a lot of
sense. The second lesson is that is better to have a bi-lingual
researcher or worker - which applies in the case of your student, though
I still feel it will really help to check and affirm the work she has
done.
Chris
-----Original Message-----
From: A forum for discussion on midwifery and reproductive health
research. on behalf of Marshall, J. L.
Sent: Fri 30-Nov-07 8:54 AM
To: [log in to unmask]
Cc:
Subject: Re: Back translation: any guidelines?
Hi Carolyn
To help with this question I feel it is important to know quite a lot
more about the way the research has been planned prior to the conduct of
the interviews. For example: what was the sampling strategy that led to
the selection of the people who were interviewed; how was the topic
guide or interview schedule developed, what framework has been used to
develop the study etc? I suggest that all these things should be
informing the way that she chooses to analyse the data.
I think it is likely that it would be more helpful for her to look at
all of the interview data and identify the major themes but without more
information about the study it is difficult to say.
I think Jennifer Mason's book provides a really good overview and might
be a useful resource: Mason J. (2002) Qualitative researching. Sage
Publications.
I hope this is helpful?
Joyce
--
Joyce Marshall
Lecturer (Midwifery)
Department of Health Sciences
Area 4, Seebohm Rowntree Building
Heslington
York YO10 5DD
Telephone (01904) 321831
Fax (01904) 321820
Email [log in to unmask]
Roth, Carolyn wrote:
> Dear all, one of my MSc students is conducting a study using 5
interviews (ranging from 15 minutes to 1 hr) which were conducted in
Greek and have been translated by her into English. She wants to
validate the translations and has asked another bilingual friend to
back-translate into Greek.
>
> She wanted some advice about how much she needs to look at. I've
suggested several small sections from each interview, 'randomly'
selected.
>
> I wondered if anyone had experience or knows any commentary on this.
>
> Thanks in advance,
>
> Carolyn Roth
> Senior Lecturer
> Midwifery Department City
> Community & Health Sciences
> 24 Chiswell Street London EC1Y 4TY
>
> Tel 0207 040 5869 Fax 0207 040 5866
>
>
|