Hi

Might depend somewhat on your preferred ordering of topics, how you go about 
teaching hypothesis testing, and when you introduce distributions amenable to 
talking about confidence intervals.  

I've always liked the sign / binomial test to introduce HT because students are 
able to compute (binomial theorem) probabilities of different outcomes, 
something they subsequently have to depend on tables / programs for.  Also 
possible to compute Type II probabilities for different Ho false values.  Not 
readily amenable to confidence intervals, except perhaps for normal 
approximation (but less than ideal for CI?) or if you introduce equivalent z 
tests for proportion.

After sign and its normal approximation, HT can be presented in the context of 
sampling distribution of mean.  Here is where it would be easy to do CI first 
or in conjunction with HT, given their equivalence (in the crude sense of 
whether CI contains Ho value).  I like to use simulations for HT here to show 
that M falls beyond z-alpha sigmas the theoretically expected number of times.  
If I wanted to stick with simulations, I'd have to think about whether it would 
be as easy to first introduce the idea that MU-null falls outside M+/- 
z-alpha*sigma expected proportion of times.  On the face of it, appears more 
challenging to grasp but might just be a matter of familiarity with the former 
approach.

So off the top of my head, not clear that introducing CI first has any 
advantages and may have some costs (at least the way I introduce HT ... e.g., 
regressing to sign test after discussing sampling distribution of mean strikes 
me as odd).

Nor would I separate equivalent CI and HT approaches too much.  Students might 
mistakenly get the idea that the approaches are more markedly different than 
they are in fact, notwithstanding all the promotion in some quarters of CI over 
HT.

Take care
Jim

James M. Clark
Professor of Psychology
204-786-9757
204-774-4134 Fax
[email protected]

>>> Marte Fallshore <[email protected]> 17-Apr-12 3:54 PM >>>
I was at the Rocky Mountain Psychological Association meeting last weekend, and 
there was a talk on confidence intervals. It got me thinking about teaching 
about confidence intervals before I get to hypothesis testing and then 
integrating it with each hypothesis test we do.

Has anyone out there done that? How did it go? Have you found a book that may 
be does something like that? Thanks,

Marte
 
 
************************************************
Marte Fallshore
Department of Psychology
Central Washington Univ.
400 E University Way
Ellensburg, WA 98926-7575

509/963-3670
509/963-2307 (fax)
Room 462, Psychology Building

Everyone is entitled to their own opinions, but they are not entitled to their 
own facts. ~Daniel Patrick Moynihan
 
When I give food to the poor, they call me a saint. 
When I ask why the poor have no food, they call me a communist. 
        ~Dom Heider Camara

I teach for free; they pay me to grade. (anon)
************************************************

---
You are currently subscribed to tips as: [email protected].
To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13251.645f86b5cec4da0a56ffea7a891720c9&n=T&l=tips&o=17352
 
or send a blank email to 
leave-17352-13251.645f86b5cec4da0a56ffea7a89172...@fsulist.frostburg.edu


---
You are currently subscribed to tips as: [email protected].
To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=17355
or send a blank email to 
leave-17355-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu

Reply via email to