Using Antconc to identify important words

[Tawnya Smith]

Thank you for permitting me to post to this Discussion Board.

I am working on my dissertation and plan to use Antconc as part of the methods. Although I am familiar with Antconc, I have little experience using it.

What I would like help with is in making sure I am using it in a way that makes sense for my project and in understanding how to accurately interpret the results.

Here is the description of my project:

Looking at one specific academic journal, I am examining language from a specific article genre from 1996-2013. I am looking at these articles from a historiographic perspective in an attempt to answer two questions:

• How do these articles grapple with the concepts of "students' rights to their own language" (https://cccc.ncte.org/cccc/resources/positions/srtolsummary) and "Black language" (https://cccc.ncte.org/cccc/demand-for-black-linguistic-justice)?
• How do the language and rhetorical moves in these articles line up with habits of white language supremacy (https://cccc.ncte.org/cccc/white-language-supremacy)?

Using the three statements linked above, I created a reference corpus (Linguistic Justice Key). I did a keyword search using my corpus as a target corpus and the AmE06 as the reference corpus. This list produced many of the terms I was anticipating.

My plan is to examine the articles from 1996-1999, 2000-2009, 2010-2019, 2020-2023 for evidence that they engage with the ideas of linguistic justice represented in the reference corpus. 

I expect that the results from Antconc will point to articles that require closer reading and analysis. Articles that do not use any of the keyterms from the reference corpus suggests no direct engagement.

Does this plan sound viable? Are there immediate issues or concerns that you see?

I welcome your advice.

[Laurence Anthony]

Hi Tawnya,

On reading your description, the plan certainly seems feasible. Through the new AntConc 4 metadata search functions, there's probably a way to set up a single corpus in which you can then search for specific years. Unfortunately, the setup is not that simple. There was a discussion about metadata searches here in the discussion group that you can search for, and you might want to look at the help page, too. Of course, you can also simply create separate sub-corpora for the different years and search within those separately. If you combine your years into single files, the Plot tool might be really useful as it will show exactly in which years the hits appear. 

I hope that helps.

Laurence.

###############################################################
Laurence ANTHONY, Ph.D.
Professor of Applied Linguistics
Faculty of Science and Engineering
Waseda University
3-4-1 Okubo, Shinjuku-ku, Tokyo 169-8555, Japan
E-mail: antho...@gmail.com
WWW: http://www.laurenceanthony.net/
###############################################################

--
You received this message because you are subscribed to the Google Groups "AntConc-Discussion" group.
To unsubscribe from this group and stop receiving emails from it, send an email to antconc+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/antconc/b71ceb06-ca05-44c0-a8e1-55e19ad1c734n%40googlegroups.com.

[Tawnya Smith]

Thank you for the suggestions.

Because I am using a historiographic approach, I did create sub-corpora for the time periods I indicated earlier.

Before that, I created a keyword list from the comparison of the Linguistic Justice corpus to the AmE06 using both Likelihood and Textdisperion (4-term). After deleting proper nouns, academic notations, and common-use words, this resulted in a list of 132 words closely associated with the ideas of linguistic justice.

Other than searching word-by-word, is there a way to use this list to examine the sub-corpora?

[Laurence Anthony]

Hi,

>Other than searching word-by-word, is there a way to use this list to examine the sub-corpora?

Using the advanced search function in AntConc, you can search for all the words at once, which should reduce your time. In the KWIC tool, by default, results are ordered by 'KWIC pattern' frequency, so you'll immediately see which of the list words are the most salient and the patterns that they appear in.

I hope that helps.

Laurence.

To view this discussion on the web visit https://groups.google.com/d/msgid/antconc/3c08e736-4db9-4285-997b-a15741d88a57n%40googlegroups.com.