1. VIP Member
Join Date
Nov 2007
Posts
5,425

## Re: discourse analysis

hinomura: I've specifically avoided breathing 'null hypothesis' etc. I think post-student is feeling awash enough in sorting through what he's now going to do to make sense of the data. Similarly:
your research must have accuracy and reliability that is, does it measure what it purports to measure and can it be repeated by another researcher and achieve the same results?
The tests are given, the data is in! NOW we tell him that?
This is a 'make the best of what we've got' situation.

2. ## Re: discourse analysis

Originally Posted by David L.
hinomura: I've specifically avoided breathing 'null hypothesis' etc. I think post-student is feeling awash enough in sorting through what he's now going to do to make sense of the data. Similarly:
your research must have accuracy and reliability that is, does it measure what it purports to measure and can it be repeated by another researcher and achieve the same results?
The tests are given, the data is in! NOW we tell him that?
This is a 'make the best of what we've got' situation.
Then a correlation analysis would be perfect with the data he has. Does A go up as B goes up? Give a simple correlation coefficient and BAM, he's done.

3. VIP Member
Join Date
Nov 2007
Posts
5,425

## Re: discourse analysis

Then a correlation analysis would be perfect with the data he has. Does A go up as B goes up? Give a simple correlation coefficient and BAM, he's done.
Seems so simple?
Firstly,
All you have said is, are they (forms of language skills) positively correlated?
Of course they are! As my fluency in speech improves, there isn't going to be a corresponding negative decrease in my writing skills.
Secondly, we already know that language skills are positively correlated.
Thirdly, how is he going to compare the two experimental groups? Apart from determining whether a correlation coefficient is greater than zero (that is, a 'real' relationship exists between the variables FOR EACH GROUP individually (foregone conclusion), one cannot determine by a statistical test that the correlation coefficient for group A is statistically different from that of group B.

Page 2 of 2 First 12

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•

Search Engine Optimization by vBSEO 3.6.1