Atlas.ti Reflections

Atlas.ti and I had to part ways.  I'm a bit bummed about that because I really did put a lot of time, effort, and money into trying to make it work.  I am very much planning to try it out again in the future...but ultimately I had to make the decision that it wasn't going to enhance my work on this project during this semester.  Here are some thoughts on why.

1. Me.  My vision struggles make it difficult for me to get accustomed to new programs.  I don't get much out of classes or tutorials that are meant to introduce various functions of the program.  I have to learn on my own through reading and trial and error.  When I sit in workshops or classes where others are demonstrating programs, I tend to get frustrated because I fall behind and miss so much of what is happening.  It's not that I can't learn how to use program or that I'm resistant to using technology.  It's just that it takes a great deal of time for me to feel comfortable using a new program in a way that works for me and my vision abilities.   
2. The program.  There were several challenges that I ran into when attempting to use Atlas.ti.  My first issue was the fact that the program can't be used on a Mac, and I don't currently own a PC.  I thought I solved that problem when I bought Parallels and Windows.  But then Atlas wasn't capable of playing my video.  While it can play some mp4 files, it cannot import the specific kind I needed to use for my project.  I didn't mind converting my video.  Unfortunately, I discovered that all other versions of the video diminished the quality.  Because the visual quality was extremely important to my data, I felt that using the converted videos would negatively impact the transcription.  I decided to transcribe using VLC to play the mp4 file, but still planned to use Atlas for my analysis.  But I eventually discovered that I needed to watch my video again and again to do my analysis...and that meant using VLC.  
3. The decision.  I thought about using codes in Atlas to mark up my transcript while projecting my video in VLC, but at this point a computer disaster led to more complications.  I borrowed a laptop from OIT, but it was unable to project my video.  I borrowed a laptop from a friend, but it was a Mac and couldn't run Atlas.ti.  While it might have been an option to use both computers in order to use VLC and Atlas simultaneously...at this point, I decided that it just wasn't the best option for me to use to complete this project.  I don't think that I would have been able to take experience the true benefits of the program and would have made my analysis harder not easier.
4. The future.  I would like to try out Atlas in the future on my own.  I've considered using it for a literature review or to code data for my own research.  I do know that I want to use it for something that I don't feel rushed to complete.  That way I can take my time to become familiar with the program and figure out what can and can't be enlarged and what features are the most beneficial for me.  

A Few Thoughts on Goodman

This was a good article for me to read.  I had been annoyed at one of my articles for my lit review (you know that thing I still need to write--ahhh!) because the researchers kept making generalizations based on their data/findings.  What I realized through reading this article (along with the other two articles on interviews) is that the problem was not the generalization in and of itself.  The problem was that they didn't do a good job of backing up their claims.  Likely they are correct that deaf teachers do the things they noted.  But they only had one deaf teacher in their study.  And their mention of other research was very much in passing and not at all detailed.  How did they determine that her actions are indicative of the the actions of all deaf teachers?  Had they given more of a description on earlier findings or made their own analysis more visible I would have been more apt to buy into the "validity" of their claims.    I suppose Goodman did a nice job of convincing me that DA findings can be generalizable with his clear example and criterion.

Wrapping Up Hutchby & Wooffitt

First of all, I am officially typing this from my repaired Macbook.  I cannot tell you how excited this makes me!  :-)

These 3 chapters were all over the place, so I'm going to blog about a few things that I think are the most applicable to me and my research.

Interviews--This touched on a lot of the same things as the articles.  But also--Yep...this is why I loathe structured interviews.  Thanks H&W for articulating my complaints and for pointing out the research that indicates these interviews are not neutral.  "Whatever the ostensible topic, context or purpose of the interview, the interviewer and respondent are engaged in social action"  (p.181).

Children's Talk--Language (development of children) is my area of interest so it's no surprise that this section stood out to me.  H&W discuss two types of research in education and sociolinguistics (and all those other related fields).  1--Development of linguistic skills is the priority.  2--Current linguistic competence is the priority.  I think in my field, too often we are guilty of focusing on the first type of research instead of valuing what we can learn from the ways students are already demonstrating linguistic competence in their own sub-cultures.  That is very much an inclusive "we."  Because I come at language development from an institutional perspective...it's natural for me to observe student language in the classroom.  I'm not sure that it occurred to me to study language in the second way until very recently.  I've been doing a study looking at a student's language across various (adult led) contexts.  My own research findings for a study I'm presenting at LRA has led me to wonder how my student participant's language is different with his peers when teachers are not around to guide the interactions.  (Because what I've found is that his "side conversations" with his classmates demonstrate both quantitative and qualitative differences in the language he uses.)  I'm interested to look into Goodwin's research because it seems to be exactly the future directions I am suggesting in response to my own research.  I was also interested in the study by Danby and Baker and several of the others.  Note to self:  See pps. 194-199 for references when you're ready to check these out.  I think this is particularly of interest in deaf education where the language used in social vs. academic settings is extremely visible especially at a school that uses  mandates simultaneous communication from teachers in the classroom.  What could we learn about those differences that might help us in our approach to students with "delayed language".  And are they as delayed as we thing they are?

Order of Disorderly Talk--Again.  This section stood out to me for obvious reasons.  I think that's why my analysis is so hard for me...it is definitely 33 minutes of 'disorderly' talk,  Reading the examples of the patient and therapist in this section reminded me of the teacher-student struggle to co-construct meaning in deaf education classrooms.  Quite frankly, it's exhausting because both participants are working so hard!  My grandfather had a brain tumor and stroke 13 years ago that left him with pretty significant aphasia.  My family often remarks when I'm home that he is much more involved in conversations and seems to be able to participate in ways that he can't when I'm not there.  For years they thought because I rarely visit he was more motivated to communicate.  But Heeschen and Schegloff's research indicates that my role as an unimpaired co-participant is important because it allows us to co-construct meaning.  I think this can also be true of teachers in the classroom...but when we study classroom talk, the language isn't really naturally occurring because it's still institutional talk.  On p. 199, H&W talk about giving stimulus materials like cartoons to assess speech.  Yep, that's exactly the kind of research that I'm used to...in fact that is was one of the SIWI studies we did last year.  Only it didn't work at all...the students gave such little language that we weren't able to get much from the analysis.  Makes me think--what if we had just watched them interacting in a social context?  How would our data have been different?  Were there minimal responses due to a resistance (note to self--check out Danby & Baker, 1998) or because the formal setting led to more severe production difficulties (note to self--check out Heeschen & Schegloff, 2003)?  With deaf students, often the teachers are the ones with the responsibility of speech and language therapy.  Most deaf educators have some training on the topic...but they don't have nearly the amount of specialized coursework and experience as SLPs.  (Sidenote--UT does a much better job of emphasizing these skills than most deaf ed programs...go UT!)  Still speech and language development in many positions is one of the main responsibilities.  I have been known to say (during my classroom days)--"My number one goal for my students is for them to learn how to use language to interact with others."  I was a Language Arts teacher--of course, I wanted my kids to read and write.  But above all, I wanted them to be able to communicate using any "sign system."  This section made me think, that there is a lot that CA (and in my opinion also DA) approaches could do to shape interventions and inform practices of deaf educators.

"Grammar"--Okay that's it.  I need to read Goodwin.  Because everything that interested me cited him.  Love that he looks at the use of gaze, gesture, and body movement as part of the 'grammar' of social interaction.  To me these are the things that I feel can never be included in a transcript (no matter how Jeffersonian it is!)  And that's why I have continued to watch my video again and again to analyze it and why I feel like there is no way for me to adequately portray the interactions in a paper.  I really liked how Antaki et. all talked about making the sound available...but this is why I can't imagine doing DA without the video.  And why ultimately I couldn't use Atlas.ti   I think to use anything less than the original quality video is distorting the data that I am analyzing.  Even then there are things (certain details of signs, gaze, gesture, etc.) that are not fully available even in the highest quality video.  The video has already changed the interaction enough.  I think this is why I'm glad that I chose to use sign language.  It forced me to attend to the details of the interaction that would have been easy to miss if I had been using spoken language.  Moreover, it forced me to make decisions about which components I considered to be part of the grammar of the interaction.   And it made me much less willing to make compromises when it came to my transcription and analysis.  I think that these experiences better prepared me because I will be more likely to understand why all of these things are important regardless of the main method of communication.

Additional Note--In other respects, I very much wish I had stuck to a "simpler" data source....because this has been even more difficult than I had imagined.  I now understand why Kimberly looked at me with very wide eyes when I asked if she had heard of the Berkley Transcription System!  Sometimes I think that using a second language (especially a signed language) has been overwhelming and eclipsed some of the details of the analysis process.  It also made it more difficult to get feedback from others.  I'm a little envious of the data sessions my other group members got.  Even Emily.  Although she used deaf students, her transcript was an English translation and the interaction was among a group of students who are not delayed in language development at all.  Hollie and Journey were able to read her transcript and give input.  But my context and data were so unfamiliar to my group that they weren't able to give much assistance, especially since we ended up with very little time.  So I'm glad that I'm trying to tackle this now since it's the only way to really use DA in my own research...but it is a little overwhelming that I don't have very many examples or colleagues in my field to go to for support.  Can there be a DART for people who know sign language and language delays?  No?  Not so much?  :-)  (I was WAY too spoiled at FSDB.  It's one of the largest deaf schools in the nation...so I was surrounded with amazing colleagues and resources.  It's easy to forget that deaf ed is not like that in "the real world."  :-)    

Interview Articles

Where were this articles when I was in Intro to Qual trying to analyze interviews and write up my findings?  I interviewed 6 beginning deaf educators and thought that there was some really good data in the interviews, but I struggled with the study because we didn't talk about transcription in class and the time we spent on analysis included mostly photographs and documents with very little discourse.  I found a few "similar studies" and followed their lead...but they weren't the best examples.  I felt that the way I transcribed lost a lot of the data that I knew was there.  And when it came to analysis I wasn't sure exactly how to go about interpreting my findings.  I'm pretty sure I did half the things the articles warn against.  :-)  Both of these articles would have been very helpful at the time.  I'm glad I read them now because I'm planning to do a similar interview study for Program Eval II in the spring.  I'm hopeful that these articles will help me both with my current analysis for my DA project and my Program Eval project.

So...some notes for me to look back at next semester and some quotes and thoughts on things that stuck out to me.

Antaki et al, 2007:  What is not analysis?

1.  (Under-analysis through) Summary "Transcription prepares the data for analysis.  However, it is not analysis in itself" (p. 13).  Summarizing losing information from the interview and doesn't add any information.  Summarizing can distort the the original message.

2.  (Under-analysis through) Taking Sides

Position-taking and critical dis-alignment limit the complexity of the discourse.  "A particular danger is that the desire to sympathise or censure, when not allied to careful analysis, can lead to the sort of simplification that is the antithesis of analysis" (p.18).  

3.  (Under-analysis through) Over-Quotation  or Isolated Quotation

Compiling of quotes.  Referring to quotes rather than analyzing them.  Using quotes as "self-evident".  

4.  Circular Identification of Discourses and Mental Constructs

Different speakers draw on the same repertoires/ideologies/discourses.   1-Quotes must be shared, 2-A claim must be made to the existence of these repertoires, etc, 3-Details of the interaction must be examined, 4-Commonalities between quotes must be explained in detail.  The same is true for mental constructs (attitudes, schemata).  If all of these steps are not taken there is danger of circular identification.  How does the speaker backtrack, justify or qualify his utterances?  What in previous research is relevant?  

5.  False Survey

Be careful not to attribute findings to the "group" represented by the participant. 

6.  Simply Spotting Features

How are discursive devices used in this particular example?  "Good analysis always moves convincingly back and forth between the general and the specific."  

Potter & Hepburn, 2011:  8 Challenges for Interview Researchers

Reporting of the Interview.

1. Make the interview set-up explicit.  What category have the participants been recruited under?  What task are the participants given?
2. Display the active role of the interviewer.  What was asked?  How was the talk delivered? (prosody, delay & overlap, emphasis, volume, tempo, etc.)  
3. Represent talk in a way that captures action.  What is of consequence to the interaction?  (overlaps, closing intonation, latching of turns to one another, rising and falling intonation, raised volume, stretched vowel sounds, different kinds of breaths, laughter, etc.)
4. Tie analytic observations to specific features of interviews.  How are links between claims and quotations made evident to the reader?

Analysis of the Interview.

1. Flooding.  What social science agenda(s) are (implicitly) present in the interviewers questions that could influence participants?
2. Footing.  What position do interviewers question/speak from?  What position do participants speak from?  As an individual?  As a member of a category?  What is relevant?  How do we know?  
3. Stake and Interest.  How and when do the interviewer and interviewees stake and interest appear during the interview?   What do agreements and disagreements do?  
4. Cognitivism and Individualism.  What role does the cognitive and psychological language used in interview play?  

"The irony is that qualitative interviews are massively overused, but their potential has been massively restricted."  (p.32)

 

In my "HU"

Update:  I updated this on Tuesday night really early on Wednesday morning.  Last Thursday night after class when I originally wrote this...I was still totally lost and not at all liking my analysis.   So instead of writing my draft for the last two days, I ended up finding a better method and direction for my analysis.  So the good news is I actually feel like I have findings to write about.  The bad news is I have a lot of writing to do tomorrow night!  =)

So...I gave up on Atlas. There are multiple reasons--all of which you can read about in my other post. So my wanna-be-HU is in a folder in Dropbox. The most important things are not in subfolders.

• Video-- A VLC file, so if you attempt to play it on any other player it won't play properly. There are converted versions in a folder...but they're pretty terrible if you ask me. 
• Transcript--I added a "key" to it. It has the codes used most often, but there are a lot not in there. Partially because the BTS manual is a 30 page document so it can't really be condensed. But also because most of it wouldn't be understandable to people who don't know ASL. So I only included what I thought my group would need today. 
• Transcript with notes--Since I'm not using Atlas, I did all my comments and coding in track changes in word.  (Well really I did it in comments on Google Docs b/c that's much easier to read and work with for me...but it was easiest to share it this way.)  
• Document In Use--A document the class is using. 
• Coconstructed document--A photo of the document they are creating taken the day after the lesson. 
• Teaching Record--The teacher's reflection on the lesson. 

I've found that analyzing the transcript is tough because it's not really ASL.  I really like the transcript because it does a great job of capturing things.  But I can't analyze, because it's already changing the data way too much.  I realized that I just had to continue watching the video and pausing to make notes on the transcript.  At first I attempted to "code" in GoodReader.  I was using different colors of highlighting and comments to look at the different types of questions the teacher was asking and to comment on what those questions were "doing."  But it wasn't working because I found that I really needed to continue to revise the transcript as I was coding.  Because BTS has codes for discourse moves and actions, there were things that I really wanted to include in the transcript instead of coding in notes. But because GoodReader uses pdf's, I couldn't make changes.  Also, I really didn't like the focus of my analysis.  So after all that...I I abandoned that approach and started over this weekend using Google docs where I could edit the transcript and code with comments.  

It took me a long time to really be able to do analysis that I was happy with.  I felt like I was too close to the data to really do what I wanted with my analysis.  I spend a lot of time with these students looking for very different things and a ridiculous amount of time transcribing this video. After doing that it was really hard for me to switch gears...because my brain was in translation mode.  I had really hoped my group could help...but it was hard with limited time to get them to a point where they could understand what was even really happening in my video.  My students are extremely language delayed so the lesson is really hard to follow even if you are familiar with sign language.  But...after watching it again and again I finally started to be able to let go of the individual sign choices and focus more on my original questions.

At this point, the things I'm focusing on:

• I don't really know what to call this, b/c it's not exactly a communication repair.  But I'm looking at strategies used to assist in communication.  Basically what does the teacher (and also the students) does when she doesn't get a response (or doesn't get the response she wants) to a question.  (repeating the question, rephrasing the question, using a visual cue, roleplaying).  This will likely end up being what I write about for this project.  
• When students comments are not taken up.  This is something I'm looking at for another study I'm doing, because I'm doing a case study on the student whose responses are most often not taken up.  And I'd like to use this to look at why.
• When the participants repeat or rephrase each other's language.  Again something that I'll likely look at for my other study...but it will also be relevant to my first question.  

I don't really know how to give you specific questions about my analysis, other than do you have better ways to say whatever it is I'm trying to say above?  (Especially the first one.)   I don't really think that my video or transcript will make much sense without me to help you navigate them...but if it does make any sense and you have any thoughts I'd love to hear them.

My biggest question for you is about Jeffersonian--Any thoughts on how I should tackle that?  I'm a bit stuck.  There are some features in BTS that I didn't use that are much like Jeffersonian.  Should I add those in to the parts I'm using in the paper and also look to see if there are Jeffersonian things that are not included in BTS and use those from Jeffersonian?  So sort of combine the two and use what's applicable from each?  I guess I just don't have a vision for how to make it "accessible" to my audience.  Any suggestions?  Or thoughts on what you're expecting?

A Blog Post about Blog Posts

As I sit to write this there are a few things I find to be ironic.

1-I'm writing a blog post about research about blog posts.

2-I started my last post with an "I don't know much preface."

3-I just read an article on perfectionism which pretty much summed up my thoughts on blogging.  

So...I hate blogging...but I also kind of love blogging.  I have a blog...and once upon a time I wrote in it.  That's the thing blogging is writing.  And I hate writing...but I also kind of love writing.   The thing is that I only like writing when I'm done writing.  But it takes me way to long to actually write things.  Even a simple blog post.  I said I'd blog about my marathon...but I never did.  I said I'd blog about my trip to Haiti...but I've only gotten through the morning of day 2.  Because I am the slowest writer in the world.  (Also I like hyperboles and I'm kind of busy.)  I think the problem is that I'm the kind of person who likes to edit...and doesn't know when to stop editing or how to hold off on editing until I've actually written enough to edit.  I hit the backspace key a lot.  I just did.  This is the third time I've started this sentence.  True story. 

I found this research to be interesting, because this is something I wonder about with teaching.  I've been doing this whole higher ed thing for long enough to have done A LOT of blogs, wikis, and discussion boards... And I'm not the biggest fan--for the same reasons mentioned in the articles.  I'm not quite the quiet ballerina I was when I started out on this academic adventure, but I'm still not a huge fan of this whole "public display" thing.  The articles sort of helped me think through why and also think about how this is important to know in my own teaching.  

When I ask students to do blogs or discussion boards, what do I need to be aware of?  How might that interaction differ (or not differ as the case may be) from in class encounters or written responses that are not public?  

Note--One way to write a blog post quickly without over-editing is to forget your computer charger and work and be mid-post when the battery warning tells you that you're running out of time.  Whoops. Ha! 

SFG and Me

Basically I have no idea what I'm doing... =)  Kidding.  Kind of.

Kimberly thought Systemic Functional Grammar might be valuable to our work so she started reading up on it and introduced us to it over the summer. Then she asked me to start coming up with a way to use it to analyze student writing to see what we could learn. So I'm attempting to do that!

I don't know what you know about SFG--but the biggest piece is that there are 3 functional building blocks (of word groups) instead of abstract parts of speech. Participants (noun groups). Processes (verb groups) and Circumstances (adjective/adverb groups). Here's a simple example.

(This picture is from this website. Which helped me learn the basic groups.)

I've placed our students in low, medium, and high groups and added a same-age hearing peer group. I'm taking each groups' writing samples and analyzing them to see what kinds of participants, processes, and circumstances the groups are using. It's pretty fascinating. Difficult because I'm really jumping into it without knowing enough about it or having a real plan . (Not always the best approach! ha!) But neat because I'm seeing already how some of the objectives we've set in the past for some students based on what we saw as a need area may not have been the most appropriate objective simply because we didn't have a good way to determine what might be the next logical step. I think in deaf ed. often we don't know how to scaffold their learning appropriately--how do you know what's in their ZPD if you you don't really know what should come next?

It's been a really concrete way of seeing how the writing differs from across the 4 groups. For example I can see that the low group is only using first person pronouns, whereas the high group is using first, second, and third person pronouns. It's helpful to me to see what kinds of language students are using to make meaning instead of being distracted by the micro-level errors that can be easier to see in their writing. I really think it has a lot of potential as a way to look at the expressive language development (written, signed, or spoken) of deaf students, which is what I'm most interested in. I'm a teacher through and through...so I want my research to be helpful in understanding the things I struggled with most in the classroom. When Kimberly first brought up SFG I doubted it's ability to be helpful in a practical way (mostly because I saw the teachers' responses at the summer PD--ha!)  But as soon as I started creating tables of my findings, I realized that I was seeing things that I hadn't been able to see with all the other ways I've looked at and analyzed student writing. SFG is crazy and tedious so I'm still not sure that I see it as a tool for teachers, but I can definitely see the potential is has for research in language development, learning, and education.

Right now I'm just finishing up an initial analysis of our experimental groups' baseline independent writing (recount) samples. But in the next few weeks we'll start using it inform objective setting and progress monitoring. In fact I have a lot to do...because we're doing a PD on it with the teachers on Friday!  And then we'll continue to use it with other genres and see what we learn.  

So that's it...I'm just getting my feet wet. But I definitely see it as having a lot of application to things I'm interested in. I wish there was someone at UT who did some work in SFG/L. Hopefully after this semester, I'll find a little time to explore it more on my own. I need to start learning more about it so that I can determine if and how I might want to approach a dissertation using it. Kimberly and Hannah are presented on it at AAAL in March, but I'll be in Haiti.  But...Kimberly and I are putting a proposal in for the International Systemic Functional Congress in Argentina in April.  I'm hoping I'll be able to go because it'll be a good opportunity to get some feedback on our work and learn more about current research in the field.  And there is a pre-congress institute on SFL and language education this year...which would be perfect.  We shall see...