Results 1 to 6 of 6
  1. #1
    MicheleR
    Guest

    Identifying individual contributors in bulletin board thread



    Hi all --

    I'm working with data retrieved from a bulletin board. The data is broken
    into individual files, one for each discussion thread on the BB. So each
    file has anywhere from 3-10 different individuals participating, each making
    one or more posts to the discussion thread.

    My question is this: I am considering creating a number of codes such as
    R_1, R_2, R_3 etc and coding all of a person's posts with the same R_# so
    that I can do things like isolate all quotes that came from a particular
    poster, or see how many unique participants are involved in a discussion.
    However, an obvious problem arises when I look at a data set larger than one
    thread, since R_1 in this thread may or may not be a completely different
    person from R_1 in another thread.

    I could go through all the threads, make a list of ALL posters and assign
    each one a unique code but that means I may end up with several hundred
    codes just for the posters and that sounds a bit excessive.

    If anyone has dealt with a similar data set I would appreciate any
    collective wisdom on the topic.

    Regards --

    Michele


    ---
    Outgoing mail is certified Virus Free.
    Checked by AVG anti-virus system (http://www.grisoft.com).
    Version: 6.0.806 / Virus Database: 548 - Release Date: 12/6/2004



  2. #2
    Naybell Hernandez
    Guest

    Identifying individual contributors in bulletin board thread



    Hi Michele

    I haven't seen anyone responding to your question with any possible
    solution. In discussing this with my husband we also thought about creating
    a unique code but this code would have to be the same for a person in any
    post she does to avoid the problem you mentioned at a larger scale.

    The suggestion that I got from one professor was that as long as I can
    figure out a way to account for multiple mentions by the same person, I'll
    be OK. I will create this codes and send you the list of each person name
    with the code it was assigned to each. I'll also add the codes to my HU. If
    I sent you a HU where those codes are included can you merge it with yours
    so you don't have to type in these codes? Please, let me know.

    Adding this unique code we don't need to have two codes for the thread
    respondent, just one. I think you mentioned that we can merge codes as well.
    If we can do that then the 2 ThreadRespondent codes should be merge into
    just Role_ThreadRespondent.

    I hope this is not too much trouble. Please, Let me know.

    I will have all the codes by tomorrow (Tuesday).

    Thanks a lot,
    Naybell


    -----Original Message-----
    From: ATLAS.ti The Knowledge Workbench - User Forum
    [mailto:ATLAS-TI@LISTSERV.DFN.DE] On Behalf Of MicheleR
    Sent: Monday, December 13, 2004 1:38 PM
    To: ATLAS-TI@LISTSERV.DFN.DE
    Subject: Identifying individual contributors in bulletin board thread


    Hi all --

    I'm working with data retrieved from a bulletin board. The data is broken
    into individual files, one for each discussion thread on the BB. So each
    file has anywhere from 3-10 different individuals participating, each making
    one or more posts to the discussion thread.

    My question is this: I am considering creating a number of codes such as
    R_1, R_2, R_3 etc and coding all of a person's posts with the same R_# so
    that I can do things like isolate all quotes that came from a particular
    poster, or see how many unique participants are involved in a discussion.
    However, an obvious problem arises when I look at a data set larger than one
    thread, since R_1 in this thread may or may not be a completely different
    person from R_1 in another thread.

    I could go through all the threads, make a list of ALL posters and assign
    each one a unique code but that means I may end up with several hundred
    codes just for the posters and that sounds a bit excessive.

    If anyone has dealt with a similar data set I would appreciate any
    collective wisdom on the topic.

    Regards --

    Michele


    ---
    Outgoing mail is certified Virus Free.
    Checked by AVG anti-virus system (http://www.grisoft.com).
    Version: 6.0.806 / Virus Database: 548 - Release Date: 12/6/2004



  3. #3
    MicheleR
    Guest

    Identifying individual contributors in bulletin board thread



    Very interesting -- and an approach that had not occurred to me except in
    the context of "Gee I wish the PDs were separated by poster" LOL!!

    Did you make the assumption that the same user name was in fact the same
    poster, across all the threads?

    Thanks

    Michele

    ----- Original Message -----
    From: "Miller, Jane" <J.Miller@GCAL.AC.UK>
    To: <ATLAS-TI@LISTSERV.DFN.DE>
    Sent: Tuesday, December 14, 2004 7:53 AM
    Subject: Re: Identifying individual contributors in bulletin board thread


    > Hi Michele,
    > Haven't got much time I'm afraid to respond to your email
    > but I have analysed postings to discussion boards using Atlas
    > so I wanted to write something in case it's of any use to you.
    > It took me a while to figure out exactly what I needed to get
    > from the analysis and I ended up coding then recoding twice
    > before I got it the way I wanted... I ended up putting each
    > poster's messages in a primary document so the primary
    > documents were by participant and I then coded each message
    > according to which discussion it belonged to and if the message
    > was in response to another participant. Then I was able to
    > run the analysis using the query tool and supercodes to find
    > out differences between males and females and age groups
    > (organised in primary doc families) regarding the linguistic
    > nature of their postings.
    > Hope this is useful..
    > Jane
    >
    > Dr Jane Miller
    > Learning Support/eLearning Support
    > Department of Psychology (M422)
    > School of Life Sciences
    > Glasgow Caledonian University
    > 70 Cowcaddens Road
    > Glasgow
    > Scotland, UK
    > G4 0BA
    > Tel: 0141 331 3400
    > Fax: 0141 331 3636
    > Email: j.miller@gcal.ac.uk
    >
    > -----Original Message-----
    > From: ATLAS.ti The Knowledge Workbench - User Forum
    > [mailto:ATLAS-TI@LISTSERV.DFN.DE] On Behalf Of MicheleR
    > Sent: 13 December 2004 18:38
    > To: ATLAS-TI@LISTSERV.DFN.DE
    > Subject: Identifying individual contributors in bulletin board thread
    >
    > Hi all --
    >
    > I'm working with data retrieved from a bulletin board. The data is
    > broken
    > into individual files, one for each discussion thread on the BB. So
    > each
    > file has anywhere from 3-10 different individuals participating, each
    > making
    > one or more posts to the discussion thread.
    >
    > My question is this: I am considering creating a number of codes such
    > as
    > R_1, R_2, R_3 etc and coding all of a person's posts with the same R_#
    > so
    > that I can do things like isolate all quotes that came from a particular
    > poster, or see how many unique participants are involved in a
    > discussion.
    > However, an obvious problem arises when I look at a data set larger than
    > one
    > thread, since R_1 in this thread may or may not be a completely
    > different
    > person from R_1 in another thread.
    >
    > I could go through all the threads, make a list of ALL posters and
    > assign
    > each one a unique code but that means I may end up with several hundred
    > codes just for the posters and that sounds a bit excessive.
    >
    > If anyone has dealt with a similar data set I would appreciate any
    > collective wisdom on the topic.
    >
    > Regards --
    >
    > Michele
    >
    >
    > ---
    > Outgoing mail is certified Virus Free.
    > Checked by AVG anti-virus system (http://www.grisoft.com).
    > Version: 6.0.806 / Virus Database: 548 - Release Date: 12/6/2004


    ---
    Outgoing mail is certified Virus Free.
    Checked by AVG anti-virus system (http://www.grisoft.com).
    Version: 6.0.806 / Virus Database: 548 - Release Date: 12/7/2004



  4. #4
    Miller, Jane
    Guest

    Identifying individual contributors in bulletin board thread



    Hi Michele,
    Haven't got much time I'm afraid to respond to your email
    but I have analysed postings to discussion boards using Atlas
    so I wanted to write something in case it's of any use to you.
    It took me a while to figure out exactly what I needed to get
    from the analysis and I ended up coding then recoding twice
    before I got it the way I wanted... I ended up putting each
    poster's messages in a primary document so the primary
    documents were by participant and I then coded each message
    according to which discussion it belonged to and if the message
    was in response to another participant. Then I was able to
    run the analysis using the query tool and supercodes to find
    out differences between males and females and age groups
    (organised in primary doc families) regarding the linguistic
    nature of their postings.
    Hope this is useful..
    Jane

    Dr Jane Miller
    Learning Support/eLearning Support
    Department of Psychology (M422)
    School of Life Sciences
    Glasgow Caledonian University
    70 Cowcaddens Road
    Glasgow
    Scotland, UK
    G4 0BA
    Tel: 0141 331 3400
    Fax: 0141 331 3636
    Email: j.miller@gcal.ac.uk

    -----Original Message-----
    From: ATLAS.ti The Knowledge Workbench - User Forum
    [mailto:ATLAS-TI@LISTSERV.DFN.DE] On Behalf Of MicheleR
    Sent: 13 December 2004 18:38
    To: ATLAS-TI@LISTSERV.DFN.DE
    Subject: Identifying individual contributors in bulletin board thread

    Hi all --

    I'm working with data retrieved from a bulletin board. The data is
    broken
    into individual files, one for each discussion thread on the BB. So
    each
    file has anywhere from 3-10 different individuals participating, each
    making
    one or more posts to the discussion thread.

    My question is this: I am considering creating a number of codes such
    as
    R_1, R_2, R_3 etc and coding all of a person's posts with the same R_#
    so
    that I can do things like isolate all quotes that came from a particular
    poster, or see how many unique participants are involved in a
    discussion.
    However, an obvious problem arises when I look at a data set larger than
    one
    thread, since R_1 in this thread may or may not be a completely
    different
    person from R_1 in another thread.

    I could go through all the threads, make a list of ALL posters and
    assign
    each one a unique code but that means I may end up with several hundred
    codes just for the posters and that sounds a bit excessive.

    If anyone has dealt with a similar data set I would appreciate any
    collective wisdom on the topic.

    Regards --

    Michele


    ---
    Outgoing mail is certified Virus Free.
    Checked by AVG anti-virus system (http://www.grisoft.com).
    Version: 6.0.806 / Virus Database: 548 - Release Date: 12/6/2004



  5. #5
    Ann Lewins
    Guest

    Identifying individual contributors in bulletin board thread



    I'd keep R1 and R2 etc for each thread, but put a 'thread' identifier
    before each speaker id..... that would make it unique to that thread - but
    it would better retain the idea of total numbers of contributors to each
    thread. It would still be more ids...but because you are using fewer
    digits - and theres more information in the ID- it woud be of more use and
    less clumsy (also more useful for autocoding)
    so
    R1 AND R2 wd be instead in thread one
    T1-R1
    T1-R2

    in Thread 2
    T2-R1
    T2-R2

    (although I'd make the thread desciptor more interesting and explicitly
    informative
    say for a thread on 'history 'for instance)...

    T-HISTORY-R1
    T-HISTORY-R2

    or to take that logic further

    T-HISTORY-R1-F: (A female contributor to the history thread)

    you can do autocoding on any of those exlusive elements and save single or
    multiple hard return paras around them e.g. to code all the
    R1-HISTORY or
    -HISTORY or
    -R1 or
    -F: or

    cheers

    Ann Lewins

    Ann Lewins, Manager
    CAQDAS Networking Project
    Dept of Sociology, University of Surrey
    GUILDFORD GU2 7XH, UK

    email: caqdas@soc.surrey.ac.uk
    OR a.lewins@soc.surrey.ac.uk
    CAQDAS web site: http://www.soc.surrey.ac.uk/caqdas/
    Tel +44 (0)1 483 68 94 55
    Fax +44 (0)1 483 68 95 51

    DISCUSSION GROUP qual-software : join information etc
    see: http://www.jiscmail.ac.uk/lists/qual-software.html
    ++++++++++++++++++++++++++++++++++++++++++++++++++ +++++++++++
    CAQDAS NETWORKING PROJECT funded by the ESRC Research Methods Programme to
    support the use of software in qualitative data analysis;
    co-directed by Nigel Fielding, Professor of Sociology, University of Surrey
    and Ray Lee, Professor of Social Research Methods at Royal Holloway,
    University of
    London



    ----- Original Message -----
    From: "MicheleR" <mrothen2@twcny.rr.com>
    To: <ATLAS-TI@LISTSERV.DFN.DE>
    Sent: Tuesday, December 14, 2004 1:08 PM
    Subject: Re: Identifying individual contributors in bulletin board thread


    > Very interesting -- and an approach that had not occurred to me except in
    > the context of "Gee I wish the PDs were separated by poster" LOL!!
    >
    > Did you make the assumption that the same user name was in fact the same
    > poster, across all the threads?
    >
    > Thanks
    >
    > Michele
    >
    > ----- Original Message -----
    > From: "Miller, Jane" <J.Miller@GCAL.AC.UK>
    > To: <ATLAS-TI@LISTSERV.DFN.DE>
    > Sent: Tuesday, December 14, 2004 7:53 AM
    > Subject: Re: Identifying individual contributors in bulletin board thread
    >
    >
    >> Hi Michele,
    >> Haven't got much time I'm afraid to respond to your email
    >> but I have analysed postings to discussion boards using Atlas
    >> so I wanted to write something in case it's of any use to you.
    >> It took me a while to figure out exactly what I needed to get
    >> from the analysis and I ended up coding then recoding twice
    >> before I got it the way I wanted... I ended up putting each
    >> poster's messages in a primary document so the primary
    >> documents were by participant and I then coded each message
    >> according to which discussion it belonged to and if the message
    >> was in response to another participant. Then I was able to
    >> run the analysis using the query tool and supercodes to find
    >> out differences between males and females and age groups
    >> (organised in primary doc families) regarding the linguistic
    >> nature of their postings.
    >> Hope this is useful..
    >> Jane
    >>
    >> Dr Jane Miller
    >> Learning Support/eLearning Support
    >> Department of Psychology (M422)
    >> School of Life Sciences
    >> Glasgow Caledonian University
    >> 70 Cowcaddens Road
    >> Glasgow
    >> Scotland, UK
    >> G4 0BA
    >> Tel: 0141 331 3400
    >> Fax: 0141 331 3636
    >> Email: j.miller@gcal.ac.uk
    >>
    >> -----Original Message-----
    >> From: ATLAS.ti The Knowledge Workbench - User Forum
    >> [mailto:ATLAS-TI@LISTSERV.DFN.DE] On Behalf Of MicheleR
    >> Sent: 13 December 2004 18:38
    >> To: ATLAS-TI@LISTSERV.DFN.DE
    >> Subject: Identifying individual contributors in bulletin board thread
    >>
    >> Hi all --
    >>
    >> I'm working with data retrieved from a bulletin board. The data is
    >> broken
    >> into individual files, one for each discussion thread on the BB. So
    >> each
    >> file has anywhere from 3-10 different individuals participating, each
    >> making
    >> one or more posts to the discussion thread.
    >>
    >> My question is this: I am considering creating a number of codes such
    >> as
    >> R_1, R_2, R_3 etc and coding all of a person's posts with the same R_#
    >> so
    >> that I can do things like isolate all quotes that came from a particular
    >> poster, or see how many unique participants are involved in a
    >> discussion.
    >> However, an obvious problem arises when I look at a data set larger than
    >> one
    >> thread, since R_1 in this thread may or may not be a completely
    >> different
    >> person from R_1 in another thread.
    >>
    >> I could go through all the threads, make a list of ALL posters and
    >> assign
    >> each one a unique code but that means I may end up with several hundred
    >> codes just for the posters and that sounds a bit excessive.
    >>
    >> If anyone has dealt with a similar data set I would appreciate any
    >> collective wisdom on the topic.
    >>
    >> Regards --
    >>
    >> Michele
    >>
    >>
    >> ---
    >> Outgoing mail is certified Virus Free.
    >> Checked by AVG anti-virus system (http://www.grisoft.com).
    >> Version: 6.0.806 / Virus Database: 548 - Release Date: 12/6/2004
    >
    >
    > ---
    > Outgoing mail is certified Virus Free.
    > Checked by AVG anti-virus system (http://www.grisoft.com).
    > Version: 6.0.806 / Virus Database: 548 - Release Date: 12/7/2004
    >



  6. #6
    Miller, Jane
    Guest

    Identifying individual contributors in bulletin board thread



    Hi Michele,
    No, I didn't make any assumptions regarding the usernames because
    I was looking at gender and one of my criticisms of previous research,
    was that they had inferred gender from the username, which may
    not be reliable. So I conducted my research using a sample where
    I had access to their background details such as age, gender
    and course of study etc.. (university students) so I could
    match the username to the participant and reliably infer gender.
    Cheers
    Jane

    -----Original Message-----
    From: ATLAS.ti The Knowledge Workbench - User Forum
    [mailto:ATLAS-TI@LISTSERV.DFN.DE] On Behalf Of MicheleR
    Sent: 14 December 2004 13:08
    To: ATLAS-TI@LISTSERV.DFN.DE
    Subject: Re: Identifying individual contributors in bulletin board
    thread

    Very interesting -- and an approach that had not occurred to me except
    in
    the context of "Gee I wish the PDs were separated by poster" LOL!!

    Did you make the assumption that the same user name was in fact the same
    poster, across all the threads?

    Thanks

    Michele

    ----- Original Message -----
    From: "Miller, Jane" <J.Miller@GCAL.AC.UK>
    To: <ATLAS-TI@LISTSERV.DFN.DE>
    Sent: Tuesday, December 14, 2004 7:53 AM
    Subject: Re: Identifying individual contributors in bulletin board
    thread


    > Hi Michele,
    > Haven't got much time I'm afraid to respond to your email
    > but I have analysed postings to discussion boards using Atlas
    > so I wanted to write something in case it's of any use to you.
    > It took me a while to figure out exactly what I needed to get
    > from the analysis and I ended up coding then recoding twice
    > before I got it the way I wanted... I ended up putting each
    > poster's messages in a primary document so the primary
    > documents were by participant and I then coded each message
    > according to which discussion it belonged to and if the message
    > was in response to another participant. Then I was able to
    > run the analysis using the query tool and supercodes to find
    > out differences between males and females and age groups
    > (organised in primary doc families) regarding the linguistic
    > nature of their postings.
    > Hope this is useful..
    > Jane
    >
    > Dr Jane Miller
    > Learning Support/eLearning Support
    > Department of Psychology (M422)
    > School of Life Sciences
    > Glasgow Caledonian University
    > 70 Cowcaddens Road
    > Glasgow
    > Scotland, UK
    > G4 0BA
    > Tel: 0141 331 3400
    > Fax: 0141 331 3636
    > Email: j.miller@gcal.ac.uk
    >
    > -----Original Message-----
    > From: ATLAS.ti The Knowledge Workbench - User Forum
    > [mailto:ATLAS-TI@LISTSERV.DFN.DE] On Behalf Of MicheleR
    > Sent: 13 December 2004 18:38
    > To: ATLAS-TI@LISTSERV.DFN.DE
    > Subject: Identifying individual contributors in bulletin board thread
    >
    > Hi all --
    >
    > I'm working with data retrieved from a bulletin board. The data is
    > broken
    > into individual files, one for each discussion thread on the BB. So
    > each
    > file has anywhere from 3-10 different individuals participating, each
    > making
    > one or more posts to the discussion thread.
    >
    > My question is this: I am considering creating a number of codes such
    > as
    > R_1, R_2, R_3 etc and coding all of a person's posts with the same R_#
    > so
    > that I can do things like isolate all quotes that came from a
    particular
    > poster, or see how many unique participants are involved in a
    > discussion.
    > However, an obvious problem arises when I look at a data set larger
    than
    > one
    > thread, since R_1 in this thread may or may not be a completely
    > different
    > person from R_1 in another thread.
    >
    > I could go through all the threads, make a list of ALL posters and
    > assign
    > each one a unique code but that means I may end up with several
    hundred
    > codes just for the posters and that sounds a bit excessive.
    >
    > If anyone has dealt with a similar data set I would appreciate any
    > collective wisdom on the topic.
    >
    > Regards --
    >
    > Michele
    >
    >
    > ---
    > Outgoing mail is certified Virus Free.
    > Checked by AVG anti-virus system (http://www.grisoft.com).
    > Version: 6.0.806 / Virus Database: 548 - Release Date: 12/6/2004


    ---
    Outgoing mail is certified Virus Free.
    Checked by AVG anti-virus system (http://www.grisoft.com).
    Version: 6.0.806 / Virus Database: 548 - Release Date: 12/7/2004



Similar Threads

  1. Story Board View - For Interdependent Images
    By Michael in forum Suggestions
    Replies: 0
    Last Post: 2010-10-26, 02:36 AM
  2. Replies: 0
    Last Post: 2010-04-07, 11:16 PM
  3. when posting a new thread
    By Gerben Moerman in forum General
    Replies: 0
    Last Post: 2006-02-14, 02:15 PM

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •