Inquiry about log likelihood And effect size

[N AG]

Hello respected and fellows,

I was watching the antconc video tutorial by Sir Anthony posted one year ago for keyword feature. My question is how will I interpret that this log likelihood result is significant or insignificant and this effect size is strong? Is there any fix reading point or figure that help to interpret them? I haven't any statistical basic knowledge to decode the. Kindly guide me about these two features. I have attached the screenshot as well as link of your tutorial video where i have seen these features but you did not mention either results wer significant or Not. Thanks in advance. 

https://youtu.be/SludW4FHatI?si=h0pz3UIEzH4yu0Go

[N AG]

Hello Respected sir and fellows,

Can anybody help me how to interpret the findings of log likelihood and effect columns while using Keywords analysis. I have attached the screenshot. Thanks in advance.

Nadia. 

[N AG]

Hello Respected sir and fellows

In keyword Analysis, for log likelihood results antconc by default used 0.05 value and Dice option for effect settings. Now, I do not know how to interpret these results that what these readings are showing. 

Many thanks.

Nadia.

[Laurence Anthony]

Hi Nadia,

To understand the statistics used, I suggest you read my article below:

https://www.laurenceanthony.net/resources/statistics/common_statistics_used_in_corpus_linguistics.pdf

As for how to interpret the values, there was a recent discussion in the group about the meaning and interpretation of likelihood values and effect sizes. I suggest you read that.

https://groups.google.com/g/antconc/c/YfaUQ2CJWTE/m/YUryRY6NDwAJ

Laurence.

[N AG]

Sure sir, I will read now. I have already read one of your paper about these topics. May be this link will be the same. Thanks again for your reply. 

Sir I have seen in antconc tutorial that we can hide and show tags by using global settings. I am using the latest version of antconc but I did not find the option to hide tags. There is only one option "type" that is not working. How can I see this option active? Or is it not availble in this version? Thanks in advance. 

Nadia. 

--
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/105a91fc-7a59-4aa5-9aa4-edd415045092n%40googlegroups.com.

[Laurence Anthony]

Dear Nadia,

You can now choose how to display your data in the tool settings. If you want to see both the word and the tag, choose the "Type+POS" option.

Laurence.

To view this discussion on the web visit https://groups.google.com/d/msgid/antconc/CAEjwtaa53xR1zs-GaDRfhx%2B7AwoRZPZXU_f1BX7nwVF%2B1cN%2BBw%40mail.gmail.com.

[N AG]

Hello Respected Sir,
This thread is for all whoever can guide. Sir Anthony Might be busy in his upcoming project and unable to respond it. So, group fellows can guide me. I'll appreciate it.

https://groups.google.com/g/antconc/c/YfaUQ2CJWTE/m/YUryRY6NDwAJ

Sir has mentioned me this thread. I got that when we do keyword Analysis we should not use effect size but default Log likelihood. My question is how I will interpret my data in keywords. What columns i should Copy for my data analysis table form.keyword frequency, keyness, norm frequency and should i exclude the effect size column in my table?
I do not know which columns are necessary in keyword analysis. Kindly guide me step by step. I will use AntConc and tell me which is the default number that help us to determine that it is strong, or weak differnce. I am not  Student of statistics. Like In The link, one example was mentioned 0,1 And above 1 digit or O. O5 value stands for What?

Regards
Nadia

To view this discussion on the web visit https://groups.google.com/d/msgid/antconc/CAL6Fgv2-VNad4U50QSrwxWGjx1bY2qLnWkBNDEmJvL8UPs5hRA%40mail.gmail.com.