Yoshikoder 0.6.3 – a preview
A preview release of version 0.6.3 of the Yoshikoder is available from the home page. I described the changes in a previous post a while back.
Remember that this is only a preview: online help is not up to date, and I’m sure there will be some other things to iron out. Nevertheless statistical document comparisons are ready. And you saw them here first…
Hello Will,
Great tool, and I look forward to trying out the new version. The multiple document addition and statistical document comparison will be useful!
Not sure if this is the right place to ask this question, but I am curious if there’s any way to set the concordance window default to a number other than “2″. If not, this would be a useful feature, as folks running concordances or concordance reports on different documents have to go back to the YK prefs every time a new doc is entered to reset the window token width (if it’s a number other than 2, which is the default). I am using the Mac Yoshikoder version 0.6.2.
Ian Lamont
September 2, 2006 at 3:14 am
Dancing preferences? I just can’t replicate this on the 0.6.2 or 0.6.3 preview Yoshikoder. All my test machines keep their concordance preferences across sessions. Does anyone else have this problem?
Will Lowe
September 8, 2006 at 4:52 pm
Hello Will
I’m currrently writing my MA-thesis, in which I do a content analysis of Norwegian party programmes, and was thrilled to stumble upon the Yoshikoder on the net a couple of weeks a go! I do believe this tool will greatly facilitate my work on the thesis.
One question: Is it possible to make patterns consisting of two or more words? Let’s say I would want to search for the term “civil society”, how do I go about using the YK?
Also, I would very much like to see document comparisons with confidence intervals!
Thanks a lot–the software is excellent!
Best regards
Pål
MA-student, Political Science, University of Oslo
Paal Skogrand
October 20, 2006 at 3:03 pm
Pål,
I’m glad you find it useful.
Yoshikoder doesn’t, as it stands, do multiple word patterns at all. However, multiple word patterns are now the most requested new feature, so will be in the next version.
I deliberately avoided doing word frequency comparisons with confidence intervals, because I’m not sure they make sense. First, the counts for individual words will be very low and so rather unreliable. Second, the point of having a content dictionary is to group into categories sets of words that are exchangeable with respect to content. This avoids having to mentally aggregate multiple noisy indicators of content, measured at the word frequency level.
Having said all that, you can always construct your own intervals from the information in a unified frequency report for two documents, exported as a spreadsheet. Email me if you want the mathematical detail. Or you can make categories containing a single pattern, and use the dictionary comparison routines. This latter approach is more consistent with testing claims about vocabulary, than exploratory data analysis, so it may not be so useful to you.
Hope that helps.
Will Lowe
October 20, 2006 at 3:27 pm