Rebecca's Research Blog

I’ve finally taken the plunge and got all my epistolary interviews with students and tutors underway. This was supposed to happen much earlier, but I got held up by the issue of which conferences could be archived which has really only just been sorted out.

I didn’t want to end up interviewing a random selection of people, so had to wait until I found out which conferences would be archived.

I’m now interviewing everyone on the four archived conferences who agreed to an interview. I’ve focused on another conference which hasn’t been archived, as the tutor wrote on their consent form that it had been an interesting group. I’ve also focused on another unarchived group, because lots of its members agreed to be interviewed.

Together with the interviews I did when I was expecting to interview other groups, that’s 22 in all. I’ve started 16 today, which is too many to have running at the same time – but I didn’t want to delay any longer.

Some of the students and tutors have given their OU email addresses. I know they use these when a course is running, but I’m worried they won’t check those mail boxes otherwise. It will be interesting to see whether everyone takes part in the interviews.

Two interviews with tutors from the last tranche petered out. I’ve decided to leave these, as the tutors were obviously busy and I don’t want to pester them.

I was just sending the first part of my email interviews out to students, when I thought I should check the link to my web page.

I’d forgotten that I’d tried out my Flickr badge there. It works very effectively but lots of my Flickr pix were taken at the Guinness factory when I went to the CAL conference at Dublin.

Thought I’d better not give the impression that I’m obsessed with alcohol on my home page. I’d use the badge in this blog, but I can’t work out how to do that. Something to do with the template, I guess. I’ve pasted it below for when I have the time to sort it out.

1. Open the relevant FirstClass conference. Right-clicking on the column names in grey allows you to choose which columns will be displayed. Choose Attachments, Name, Last Modified, Kind, Subject. (Kind is just in there to aid the sorting process). Sort by Name by left-clicking on the grey Name column heading. Drag the columns into order. 

2. In the main Snag-It window, set up a profile with the settings Input=Auto scroll window, Output=File, Effects = Space Formatted. (To do this, change the profile settings at the bottom of the screen, then press the big plus sign near the top right of the screen.) 

3. Click Capture, and Snag-It will scroll down and pickup the entire window. This takes a couple of minutes. Then save it as a text file. 

4. Copy all and paste into a Word document. Find and replace the double line spaces and the thousands of double spaces. 

5. Now set it up to be converted to a table using commas to separate the columns. First, remove all the commas which are already there. Find and replace them with an obvious string such as %%%. 

6. Take each author’s name. Find and replace the name with a comma after it. Use their full name rather than just their first name, otherwise you’ll run into problems when a name has been included in a subject line. 

7. Convert the year of posting to the year followed by a comma. Replace the word in the Kind column with a comma, thus putting a division after all the times. 

8. Go to the Table tab, and convert text to table, choosing the comma to sort the columns. 

9. Replace the commas which were in the text originally, by replacing all the %%% with commas. 

10. Paste into Excel, sort by the title column. Replace the numbered Re’s eg Re(2) becomes 2. Replace the remaining Re’s with 1 and put 0 by all other entries.  Delete bracketed numbers (1) and then Re: 
11. Add a column and manually note which postings have attachments.
12. Now for the laborious bit. Put it all back into Word ,on a landschape page in a small font. Add two columns and manually paste in all entries and the posting history.

Gill has just pointed out to me a brilliant feature of Snag-It. It will pick up text which could not normally be cut and pasted. So the menus on a FirstClass conference, which appear as text, but which I was having to copy across manually, can be picked up in a text capture grab of a scrolling window.

The conference I’m archiving at the moment has around 650 postings. I was having to copy all the authors across, which wasn’t too tricky, but then having to move dates, times and titles manually, which was time consuming, laborious and boring. Now it’s all done with a screen grab and a series of find and replaces. Far quicker.

Hoorah for Snag-It! Everyone should have a copy.

I’ve now worked out a method I’m happy with of archiving a FirstClass conference. It gives me access to the attachments and history of each posting, and provides for easy data sorting.

Theoretically, I could archive a conference in a day, if I worked late. However, the limitations are:

• I get bored. It’s very, very boring and, hey, the Internet is just a click away
• I get RSI. Endless cutting and pasting. My fingers begin to tingle. I’ve swapped my mouse to the left to double my time, but there’s still a physical limit to how much can/should be done in a day.

I decided to put all the data into an Excel file, so that it would be easily sortable. Although FirstClass conferences can be copied and pasted, the contents pages cannot be, so I had to type in all these details. I have columns for: 

• post author 

• date of post 

• time of post 

• title of post 

• whether the post has an attachment 

• which place in a thread the post appears (eg ‘Test’ is in position 0, ‘Re: Test’ is in position 1, ‘Re (2) Test’ is in position 2 etc). This avoids a sort by title which ends up with all the Re’s clumped together 

So far, so good. It’s time consuming, but it’s doable. However, I run into a problem when I come to paste in the text from each post. Excel interprets each return as a signal to move to a new cell. My organised table quickly becomes a terrible mess. There may be a way to get round this, but I’m not good enough at Excel to find it.  I therefore paste it all into a Word table. It remains sortable and, if I work out a way to do it in Excel, it can all be pasted back in there. I shrink the other columns as far as possible, change all the text to 8pt, and view at 75%. I end up with 76 pages of very small type. By doing this I’ve lost the formatting – I don’t know which size the text was originally presented, but that doesn’t matter for my analysis and nobody appears to have used distinctive font sizes anyway. 

There are more attachments than I expected. I download the Excel, Powerpoint and Word attachments into a folder and note at the bottom of the text field of my Word table which attachment goes with which post. My licence for SPSS has expired so I can’t currently download the three SPSS attachments.

Now, how do I pick up the History information which tells me who has read or downloaded a posting and when they did that? Again, this can’t be cut and pasted. I try the idea of entering it manually in abbreviated form, but that’s a lot of work for very little return. I settle on opening the History of nine postings at a time, arranging them on my screen and taking a screen dump. It’s slow, but manageable.

I make a number of false starts. I try picking up all the FirstClass postings and dragging them into a folder on my hard disk, but I only succeed in picking up aliases. I try cutting and pasting text consecutively into a Word document, but the header information won’t cut and paste. I try and fail to use the FirstClass archive function (don’t have the right permissions, and I’m not sure that I want to end up with a FirstClass archive).

Some sort of image would be good, so I buy and download SnagIt software which allows me to save rolling screens. I arrange the conferences according to date, author, title, size and take pictures of each. This works for most of the conferences, but one seems to have author search set as default and refuses to treat any other arrangement seriously. I’ve had this problem with FirstClass before and have failed to solve it. And its Help function is atrocious.

I start to take screen dumps of each posting in turn, but this is laborious, and I have to type in the title of each one, and I can’t quite see how I’d use these in future. I stop doing this. Time for a new approach.

A how-I-did-it list to remind me of how I did it when I come to write my thesis.

I’ve been archiving a FirstClass conference for the last two days. I’m drawing on my experience of the last two conferences I worked with – which were both archived by someone else. I had several problems with that data which I wanted to avoid here.

1. The files had gone through so many transformations that they were full of little rectangles and returns which no find-and-replace operation would pick up on. It took a lot of formatting and reformatting and shifting from Mac to PC and back again to get them into readable shape.
2. One of the conferences didn’t show the dates when posts were made, only the date on which they were archived. This was no use whatsoever for a temporal analysis. Also, the participants in that forum reported a lot of technical problems with postings going missing or popping up out of order. It was impossible to work out in what order messages had been posted, or how threads hung together.
3. The postings couldn’t be sorted according to eg poster, thread or date without an enormous amount of manual work.
4. None of the attachments had been saved, which meant that the main body of the work was not available for study, and I couldn’t see how the study had been developed, amended or finalised.

Just got an email to say I have access to the four asynchronous conferences I intend to archive for my PhD.

I’m dropping everything to archive them at once.

🙂  🙂  🙂  🙂  🙂  🙂  🙂  🙂  🙂  🙂  🙂  🙂  🙂

Content analysis of computer conferencing transcripts – which one should I use? Roisin Donnelly, B Holmes, J Gardner
http://ltc.dit.ie/roisindonnelly.html

to come!