At 08:39 PM 11/28/2008, Scott Woods wrote: . . .
>  I don't think my students are as smart as their scores indicate, . 
> . . I expect that students whose reading percentile scores are 
> lower will present a greater difference between the control and 
> experimental conditions.

DD: Interesting thought. I would expect the higher to profit more. 
Two tail test, then. Null Hypothesis no difference between the two 
groups. However the more interesting item is if the two techniques 
prove different in results, pre experiment and post.

>  This seemed to be the case for the previous test, based on raw 
> scores.  If such a difference is more pronounced in a timed 
> situation, that would seem to validate the experimental condition 
> as improving how well students understand text.   (Anecdotally, the 
> overwhelming majority of my students report that it is extremely helpful).

DD: No, it merely indicates the different result from timed test and 
non timed. Either would indicate the probability of the truth of the 
hypothesis. If both types of test are statistically significant, BRAG.

>
>An additional experiment would test the influence of extensive 
>practice in reading graphic syntax on reading skill.  This 
>experiment would compare the results of a second test with changes 
>in reading skill as measured on a standardized test.

DD: Pre test, post test with both your test and the standardized one, 
and be sure to have that control group that doesn't get the special 
graphic syntax training. Great idea. Ignore the ethical problem of 
not providing the control group with what you feel would help them the most.

>  My prediction is that after a year of reading difficult material 
> in graphic syntax, as well as difficult material in normal text, 
> reading scores would go up (this would not be a surprising or 
> revealing result) and the difference between experimental and 
> control performance for lower performers whose performance improved 
> would become similar to that of higher performers in the first test 
> (this would be important). I expect that if good readers are good 
> readers partially because they understand syntactic connections 
> better, and consequently are less aided by graphic syntax, then 
> improving student understanding of syntactic connections by 
> extensive practice with graphic syntax should improve their reading 
> comprehension.

DD: Tricky, but some good testable hypotheses there. You are on a 
worthy track.

>By showing a strong correlation between standardized reading 
>comprehension scores and the difference between scores on graphic 
>and normal text, I should be able to isolate the factor of graphic 
>syntax from all the other excellent things I do in the classroom and 
>show that reading in graphic syntax format can improve reading comprehension.
>  Does this make sense? Is it sound?

DD: Looks firm to me. Now do it and put it to the empirical test. Does it work?

>Is there another way to show that reading skill can be improved by 
>this method?
>
DD: Probably, but I like what you propose. However don't use the 
phrase, "...show that reading skill can be improved by this method." 
Use, "...test the hypothesis that  reading skill can be improved by 
this method." The words, 'Show' and 'Prove' are trigger words to 
incite criticism from Scientific Methodologists. Not unlike confusion 
between 'reliability' and 'validity' do to statisticians. Keep up the 
good work! 

To join or leave this LISTSERV list, please visit the list's web interface at:
     http://listserv.muohio.edu/archives/ateg.html
and select "Join or leave the list"

Visit ATEG's web site at http://ateg.org/