Note: forwarded message attached.
How much free photo storage do you get? Store your holiday snaps for FREE with Yahoo! Photos. Get Yahoo! Photos
--- Begin Message ---
--- the forwarded message follows ------ Begin Message ---Hi,
I am finally down to my last final paper -on India's District Priamry Education Projects effectiveness-why and why not---30 pages to be submitted tomorrow by midnight-by email. I'll start writing today.
Today I submitted a very hectic paper -on statistical data analysis-our S030. I think it could have come less hectic if the course had been organized according to same principles of instruction as we are taught to practice in future.
The teacher was new (stop gap for someone who is on leave and teaches the course usually) and did have lots of practical data analysis experience. And I think she did a good job of doing what she was supposed to do - to do to the letter -what her predecessor had done.
The handouts, the assignments and even the final project seem to never have been revised for past many years. There was one memo to help us with our final project- by some TF who had done something similar in 1999 - maybe she had done the same assignment.
The assignments were the same ones which our seniors (TFs) had done -exactly. They were very helpful and even helped me understand my deficiencies and alowed me an opportunity to rewrite the last HW. But the question remains - Isn't quality a moving target?
Shouldn't there be constant revisions in curriculum? I think this year our seniors (TFs) have suggested an overhauling of the system.
One suggestion I would make is - that upto regression analysis we had already learnt in S012 -- and that part was again covered till Spring break. I would have organized the curriculum a bit differently if I had to teach it myself. I would spend just 2 weeks and one assignment on topics upto regression analysis (single predictor) -the students are supposed to know that stuff before they joined the course.
And then divide the partial correlation and multiple regression into atleast two separate assignments --with lots of sections for discussions -and divide the students into groups -which a TF for each group. Might include practice assignments for discussion during sections (like in S012)
Have a time table posted on the course website about what topics will be covered on which days. Instead of the TFs giving us tips about assignments -- I would suggest that we have more assignments --one for each topic--unlike the last two we had. HW4 was on two topics simultaneously - partial correlation and multiple regression -- both related topics ofcourse - but mind blowing - for me atleast to do them together.
Buidling up a final multiple regression model is no joke -- it needs a separate assignment. Similalry, understanding the overlappings (in partial correlations) and their foreshadowing for multiple regression - is another battle altogether.
However, I feel I have learnt a great deal about data analysis and research - which hopefully would be useful to me in future. Without use of computer knowledge (SAS program) taught to us, it would have taken me years to do the final project in India-which we did in 15 days. I think that is the reason why Indian research is so slow and backward, since only now MNCs like Gallup poll have started training their trainees with SPSS etc. I have an email friend who is doing a summer internship with Gallup in Delhi and she will start learning SPSS now.
God help those in Delhi Univ.
Any ideas?
Umesh _______________________________________________ Iepcohort2005-list mailing list [EMAIL PROTECTED] http://gse.harvard.edu/mailman/listinfo/iepcohort2005-list
--- End Message ---
--- End Message ---
_______________________________________________ Assam mailing list Assam@pikespeak.uccs.edu http://pikespeak.uccs.edu/mailman/listinfo/assam
Mailing list FAQ: http://pikespeak.uccs.edu/assam/assam-faq.html To unsubscribe or change options: http://pikespeak.uccs.edu/mailman/options/assam
