Comparing target and reference data

290 views
Skip to first unread message

Deborah Davidson

unread,
Mar 6, 2024, 7:20:41 AM3/6/24
to AntConc-Discussion
Hi Professor Anthony, 

I have been searching for frequencies, n-grams and collocates (+accumulated collocates) of the key word lead* in national health policy documents.
 
Now, I would like to compare the national docs with local hospital reports.

I see that I could use one as a reference database (national) and one as the target  database (local).  So have loaded them up as such. 

But now what?  how do I see the comparison? is it by undertaking all the analysis in one and then swapping, or will I see compared norms in the results? 

Many thanks 
Deborah

Deborah Davidson

unread,
Mar 14, 2024, 4:58:12 AM3/14/24
to ant...@googlegroups.com
Does anyone know the answer to this question?

Deborah Davidson


--
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/5ef6fbf4-14fb-49a9-8728-703b19365fbbn%40googlegroups.com.

Laurence Anthony

unread,
Mar 14, 2024, 5:02:21 AM3/14/24
to ant...@googlegroups.com
Hi Deborah,

When you say "key word", do you just mean "search word"? Also, when you say you want to compare the two corpora, in what ways do you want to compare them? If you can provide some more details, I think we can give you some suggestions for what to do.

I hope that helps!

Laurence.


Deborah Davidson

unread,
Mar 14, 2024, 6:38:27 AM3/14/24
to ant...@googlegroups.com
Hi Professor Anthony,
Yes I mean search word/term.

My interest is in looking at what the n-gram and collocate results are in the different corpora I am exploring.

- national policy documents
- local NHS trust reports
- interviews with executives, healthcare professionals (medical and nursing leads) and lead managers

Being able to compare the results hopefully will enable me to understand whether and how leadership is articulated in each of these systemic layers? Are they different? Similar? 

I’m using corpus analysis as a first step towards a post-structuralist discourse analysis. 

In your videos you talk about how corpora can be compared, so I was  interested to know how a comparison is generated in practice (I.e. is it just a focus on norms? Or does it show the different results for each corpora side by side, in which case how do you generate that practically - there’s no video on this aspect.

Hope that is clear.

Many thanks 
Deborah Davidson


Laurence Anthony

unread,
Mar 14, 2024, 8:04:55 AM3/14/24
to ant...@googlegroups.com
Hi,

Here are a few things you can do:
1) Simply create results for the two corpora (i.e., NPD and LNTR) with the different tools in AntConc, treating each as a target corpus. You can then export the results and compare them with dedicated stats tools or just viewing them in something like Excel

2) You can load one corpus as a target corpus (e.g. NPD) and one as a reference corpus (e.g. LNTR) and then use the keyword tool to find which words in NPD occur unusually frequently compared with the same words in LNTR. The results here are called 'keywords'. This is perhaps the most common comparison tool used in corpus linguistics.

3) You can take any results using one corpus (e.g. n-grams in NPD), save the lists, and then load that list as a query on the other corpus (e.g. LNTR). 

Note that rather than fishing for results, I would recommend that you start out with a linguistic question. Once you formulate that, the tool(s) you need to use will probably be more obvious.


I hope that provides a useful start to your analysis!

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/
###############################################################


Deborah Davidson

unread,
Mar 14, 2024, 8:07:39 AM3/14/24
to ant...@googlegroups.com
Thank you - those are really helpful suggestions I will  follow. 
Many thanks once again.

Deborah Davidson


Kira Ogburn

unread,
Mar 15, 2024, 4:52:24 AM3/15/24
to AntConc-Discussion
Hi Deborah,
     I would recommend Paul Baker's book Using Corpora in Discourse Analysis to you. It's a great jumping off point for using corpora and corpus analysis tools in general and I think you will find a lot of useful information for exactly the type of study you are looking to do. Best of luck!

Deborah Davidson

unread,
Mar 15, 2024, 6:31:49 AM3/15/24
to ant...@googlegroups.com
Many thanks for that suggestion Kira. I shall take a look - it may well be accessible on-line.
BW
Deborah Davidson


Reply all
Reply to author
Forward
0 new messages