Learn how easy it is to sync an existing GitHub or Google Code repo to a SourceForge project! See Demo
I really like KH Coder. Great job!
If I wish to analyze multiple documents (say abstracts) must I include them all in one text file with each abstract separated with a heading e.g. H1 or is there another method for consolidating a large corpus composed of separate text files?
You seem to have CSS turned off.
Please don't fill out this field.
Just found the Unify *.txt option! I will give that a try.
Thank you for the encouraging post!
Yes, you have to include all documents in one text file with each documents separated with a heading.
And as you wrote, KH Coder can automatically perform this data preparation. Go to [Tools] [Plugin] [Data Preparation] [Unify *.txt in the Folder] in the menubar. KH Coder will unify all *.txt files in the specified folder into one text file. And add headings to separate each documents.
I discovered KH Coder last week and I am also very positive about it! Thanks for the great work!
I am also analyzing multiple documents (propositions from cognitive maps). In total I have ten cognitive maps, so the amount of data is not large. I would like to establish a comparison among these documents. I tried to use the "Unify" option that you mention in this post, but it doesnt work. Is it possible to establish a comparison among documents with KH Coder? If so, can we compare more than two texts simultaneously?
Thanks in advance, and again, congratulations.
Thank you for your post and welcome to the KH Coder users community.
Did you try the quick start tutorial?
Let's think in the following way: the novel “Botchan” is not one document but a set of documents that includes 11 different documents (or 11 chapters). Then we compared 11 documents and picked up distinctive words at page 14 of the quick start tutorial. By using correspondence analysis, we could also visually interpret the characteristics of 11 documents (page 15). We could use codes to compare documents too (page 18).
If you tell me how the “unify” command did not work, I will look into it. You have to put one document into one *.txt file to use “unify” command. Or you don’t have to use “unify” command, you can just make a text file like “botchan_en.txt” yourself.
Hope it helps.
Thanks a lot for the quick reply! Your answer definitely helps. You are right, there are very detailed instructions in the tutorial to establish the comparison. It is very clear, thanks!
However I am having some problems with the correspondence analysis. I put together the propositions coming from my cognitive maps in a .txt, separated in chapters. When I run the "correspondence analysis" as suggested in the tutorial, I get several messages of: "failed plotting with R" together with something stating: "simpleError in c$cscore: object of type 'special' is not subsettable". After getting four messages like this one, I get the final error that I attach in this message.
Is there any problem in the way I included the document? Would you have any suggestions for me to solve this issue?
Thanks again! KH coder is helping a lot with my research!
Thank you for the post!
It looks like you are using one H1 tag, one H2 tag, one H3 and so on...
You should use H1 tags only if you want to just separate chapters/documents in one file. If the file contains 11 chapters, you should use H1 tags 11 times like in the tutorial file: “botchan_en.txt.”
If you separated chapters by H1 tags and you also want to separate sections in each chapter, then you should use H2 tags to separate sections. In this case, you use H1 tags to make chapter headings and use H2 tags to make section headings. If you only have “chapters” and you don’t have “sections” in your data, then you should use H1 tags only.
Please note that after you edited your data file, you have to run pre-processing again before you perform any analyses.
Sorry for the lack of the detailed manual. Please feel free to ask more questions here.
Thanks! It works! For some reason I misunderstood the concept of tags. But it is clear now. The tutorial is short, but very clear! It helps to go step by step.
This is just what I needed. If I have further questions I will definitely ask again.
We are using khcoder for analysing architectural theory texts. The problem is - some of them are essays and some of the are whole books. Some are short manifestoes. Meanwhile one cannot set "term frequency" as percentage value (only "real" value, which affects the anylysis in unwanted way). Can You help us?
Can you describe the problem in more detail?
What are you trying to do? What is the goal of your analysis? And what prevent you from achieving the goal?
Also, when you have a new question, please make a new topic.
Ok - I will make a new topic.