Writing a statistic problem in education

Home > Statistics Every Writer Should Know > The Stats Board > Discusssion

Writing a statistic problem in education
Message posted by Varn Bruce (via 170.211.200.125) on January 18, 2002 at 11:34 AM (ET)

Help me think of a problem in education that could be solved with statistics

READERS RESPOND:
(In chronological order. Most recent at the bottom.)

Re: Writing a statistic problem in education
Message posted by JG (via 64.10.126.132) on January 19, 2002 at 5:37 AM (ET)

The slogan 'leave no child behind' makes no sense from a statistical point of view.

Every physical process has error and variation and we have to worry about alpha, beta, and gamma error.

In any case, policy should be based on data and not ideology.

Not all children are 'teachable', not all teachers are minimally competent, not all administrators are minimally honest, not all parents have minimally reasonable expectations, not all employers want people who are more than minimally educated, etc.

Re: Writing a statistic problem in education
Message posted by JG (via 64.10.124.110) on January 19, 2002 at 4:50 PM (ET)

All data gathering, especially data gathered for decision purposes has to condider sampling issues and all testing and evaluation has to consider the validity and reliability of the process.

Re: Writing a statistic problem in education
Message posted by JG (via 64.10.124.92) on January 19, 2002 at 10:45 PM (ET)

It is known that 5% of police are compulsively honest, 5% are compulsive criminals, and the rest go with whoever is the most powerful.

Similar statistics my apply to education. The student population may divided between those who will never learn, those who will learn in spite of the system, and those who will learn if it is the popular or correct thing to do.

Re: Writing a statistic problem in education
Message posted by JG (via 64.10.124.92) on January 19, 2002 at 10:52 PM (ET)

We know that in some neighborhoods learning, studying, and even literacy are looked down on and may even be dangerous. We desperartly need data on this issue - how bad is it, which identifiable sub-population is involved, what can be done about it, etc.

Your $5 contribution helps cover part the $500 annual cost of keeping this site online.