Math 151 , Day 33, Friday, April 23, 2004Hit reload... After class

HW Day33 Read (& reread) Sec 6.2-pp318-334  When that's under control, continue with pp334-337 (pp337-8 optional)  Read 6.3 p. 343-346 .  We'll skip 6.4.
I suggest this for each problem you find a P-value or sig. level for: Sketch the curve representing the sampling distribution of x-bar when H0 is true, or of the z you calculate from x-bar, and mark your observational result on it (like fig. 6.10, 6.11, 6.13)
from Moore
Sketching xbars for H0, p-value 
p. 323, 6.25 SSHA 
6.26 Spending on housing
- - - - - - - - - - - - - - - - - - -
Stating null and alternative hypotheses 
p. 325 6.27, 28, 29, 30 
- - - - - - - - - - - - - - - - - - -
Calculating p-value (one-sided), relating to Sig. level 
p. 328, 6.31 and 32 (extending 6.25 and 26) 
6.33 restating jargon
- - - - - - - - - - - - - - - - - - -
 Calculating p-value (one or two-sided), using z test statistic, relating to Sig. level 
p. 333, 6.34 price reduc. on coffee
  6.35 crankshafts true? Use your calculator to find the sample mean.
  6.36 cola? Use your calculator to find the sample mean.
- - - - - - - - - - - - - - - - - - - - - - 
More p-values 
p.341, 6.44 CEO pay.  Keep a copy of your z test statistic for use in 6.48 next time.
p. 343, 6.54  knife edge .05
p. 345, 6.55 and 56 effect of n
= = = = = = = = = = = = = = = =
*These will be part of Monday's assignment (& on the exam)
*p. 342, 6.52 1% vs 5%
*  6.53 define stat. signif.
p. 341, *6.46,  general z statistic, significance,(6.49 will be assigned too.)
p. 342 *6.50 patent protection; another z.
Read, 
to discuss
Optional 
(more practice) 
 
 

Stating null and alternative hypotheses 
p.340, 6.41,42 
- - - - - - - - - - - - - - - - - - -
Calculating p-value (one or two-sided), using z test statistic, relating to Sig. level 
p. 340 6.43 watered milk?

>>Makeup work for Exam 2 due Monday.  I'm in all afternoon today for consultation.
>>Exam 3 a week from today:  Ch. 4, 6.1, 6.2 most or all, 
part! of 6.3.  Sample exam handed out today.

>>Add your shoebox results to each of  the 2 sheets circulating:
    the 4 values || xbar|| z (assuming mean is 20)|| P-value=P(Z > z)|| Is P-value < .10?
Add a dot for each of your xbars to the dotplot transparency 

>> CI quiz If you missed the quiz Wednesday, you may take the quiz, today after class or Monday before (10 min before, in classroom) or after class---or by arrangement.
>>HW questions?
Cautions on Confidence intervals:(pp. 312-13)  Our formula
depends on SRS.
   Nonresponse or other selection bias
can destroy our conclusions.
   Outliers, skewness
can mess us up.  Nonnormality can mess us up, esp.  if sample size is < 15.
  
These cautions will hold for Significance Testing  also.
 
Significance testing 
Introduction Day32
Example:  H0: µ =1000 hrs.  (Average lightbulb life.)  Design a competing bulb:   Show it's better.
                 
Ha:   µ  > 1000 hrs.
       
Sample of size n = 25.  Population sigma = 150 hrs. Get xbar = 1060 hrs.  Are these bulbs better?
                 z = (1060-1000) ÷ (150/5) = 2.    
                  P(Z > 2) =  .0228  More than  2% and less than 3% chance of getting a result this high if we did it again.
                   "Significant at the alpha =.03 level.  Also at the alpha = .05 level"
                    "Not significant at the alpha = .02 level.  Also not significant at the alpha = .01 level"

Shoebox results: From Last year's dotplot
       White #s (green box) 2/17 = 11.8% of xbars found are significant. at 10%
       Yellow # (red top box) 13/16 are sig. at 10%   If µ is bigger than 20 by a goodly amount, the test successfully detects this.
(this year?)

2-sided (2-tailed) test:
H0: "Null hypothesis" A claim or statement about the population we would like to show is NOT true.
      H0: µ =1000 hrs.  (Average lightbulb life.)
Ha: "Alternative hypothesis" A claim or statement about the population we are trying to find evidence FOR.  A value either much bigger than or much smaller than the H0 value is evidence against H0 & for Ha.
      Ha:   µ  Not = 1000 hrs. (Quality control on assembly line--find if it is "off" either way.)
   Sample of size n = 25.  Population sigma = 150 hrs.  Suppose xbar = 940 hrs. z = (940-1000) ÷ (150/5) =  - 2

 P-value: We measure the probability of seeing something (again) as extreme as the observed value (or more so).
So you need to measure the P-value symmetrically both directions from the observed value--so the P value is double what it would be for a one-sided test.  P-value is approximately 5%; more precisely, 2·.0228 = .0456
Our test is just barely significant at the .05 level; it is significant at the .06 level, the .10 level.  It's not significant at the .02 level or "higher".

Meaning of "significance"  (note--"High" significance means small alpha or P-value.)
Question: How do we know that .05 is "significant?" (.05 is 1 in 20 chance of seeing the result by "dumb luck" if the null hypothesis is true.)  Read sec. 6.3, pp. 343-345
>>Significance levels vary by field of study; different fields have different "customarily acceptable" levels.
      In reality, no sharp border between "significance" and "not significant"
>>How small a P is "convincing evidence" against H0In practice...
        How plausible is H0?  Ha?  Strong evidence needed to reject "conventional wisdom."
        How expensive (mentally, economically) will abandoning H0 be?
>>"Statistically Significant" doesn't always mean "Important." (e.g. medicine: "Clinically significant.") Big enough sample sizes will allow you to distinguish even small differences.


Sievers home  Math151-Sp04/Days33.htm  2:30pm 4/22/04
This page belongs to Sally Sievers who is solely responsible for its content. Please see our statement of responsibility.