Math 151 , Day 30, Friday, Nov. 8, 2002Hit reload to get current versionAfter Class

HW Day30  Read (& reread) Sec. 6.1, read ahead, Sec. 6.2 at least to p. 334.  I will ask you next time to tell me  some new words we'll need to understand.    (I have a MISPRINT p. 329, tan box, first formula::  z = x -BAR minus mu-sub-zero (etc.), not z = x minus mu-sub-zero (etc.) as written.  If your book is newer, this may be fixed.)
Memorize the definition of a C.I. p. 302, esp. the "repeated samples" bit, or below,
                 and the formula p.306 for a CI for the mean, including the picture explaining C<-->z*.
    Closed Book Quiz Monday, see below
Note on reading:  p. 306, table at top:  "Tail area" is area in One tail, which is what you look up in the Normal table.
Separate sheet: For each of your sets of 4 shoebox #s, find the mean, and tell whether you believe the mean for that box is 20, or something bigger.  Bring to class to pool.  The shoeboxes are outside my door if you missed doing them in class.
- - - - - - - - - - - - - - - - - - - - - - - 
Cautions; general review and extension 
 p. 314 6.14 internet, response rate
 p.317, 6.19 newts
p. 318, 6.22  men/women CI's
 p.316, 6.18  consumers/pharmacies
- - - - - - - - - - - - - - - - - 
Sample size for C.I., review 
p. 3.11, 6.10, 6.11, 6.12 
p. 315, 6.16  enlighten the unstatistical
  6.17hotel mgrs. 
Read, to discuss

6.13, 6.15 (cautions) 
6.23 (Carter election)
- - - - - - - - - - - - - - 

Optional
Notes re Activstats: bottom of this page.
Additions to Birkenstock box results:  2 of 2 "60% CI's" captured  µ :  total (66+2)/(107+2) = 62.4%
New Shoeboxes:  Take 4 from each, write them down (White from green box, yellow from red-top box) For HW, find means (for each box separately.)  Know  which box!  Does that box have pop. mean 20, or some number >20?
HW questions?
Closed book quiz Monday:  1)  Give a definition of a level C Confidence Interval for a parameter.
2) a) Write down the formula for a level C confidence interval for the unknown mean of a normal population.
       (Assume the standard deviation of the population is known.)
    b) Tell or show with a picture how "C" connects with your formula.
Confidence interval estimate of a(n unknown) population parameter: Confidence Interval of the form  estimate + margin-of-error  for the mean µ with Confidence level C: (p.306)
Formula: (Table A, or Table C, t dist. bottom row) Why does the formula work?
  1. If a particular xbar is within m of the population mean, then the interval xbar + m contains the population mean.

  2. & If a particular xbar is farther than m from the population mean, then the interval xbar + m doesn't contain the population mean.
  3. We choose z* (and from it m) so that the probability that Xbar is within m of the population mean    is C.

  4. How? Probability C is between -z* and +z* in the standard normal table,
        between -z* ·(s.d. of Xbar) and +z* ·(s.d. of Xbar) around µ,  in the normal distribution of Xbar.
  5. Table C, bottom row, is a restating of  table A, normal table, but with probabilities (areas) on the edge, and z values in the body.  To get the z* for C = 60% from the normal table, note that this is the middle 60%, which leaves 40% to be split between the 2 tails.  So 20% above z*,  and 80% below.  Go into the body of table A, find .8000 is between values .7995 and .8023, closer to .7995.  The z value with .7995 below it is .84.  Table C gives it more precisely as  .841.
Assumptions: pp. 312-13
SRS--other random samples get other formulas.  Nonrandom or biased  samples can't use C.I.
    Sometimes we can plausibly think of data as SRS from large population (rolling dice, repeated weighings on scale)
Xbar is normal!  OK IF 1) population is normal, or 2) n big enough for Central Limit theorem.
    Outliers?  Trouble (xbar is sensitive).   Slight outliers ok (see next)
    Skewness?  n> 15 allows CLTh to overcome all but strong skewness.
Sigma for population is known.  Rarely true in practice.  Large n?  substitute s calculated from sample. Small n--Ch. 7.

Relation of m(margin of error, half width), C (confidence level), and n (sample size),  (and sigma)
    C and z* get bigger and smaller together (bigger C means bigger z*, and vice versa) (standard normal sketch)
m = z* (sigma)/ sqrt(n)
    Want bigger C?  Must accept bigger m.  Trade off confidence vs. accuracy.
    But bigger n will make smaller m. This makes sense: bigger sample size, more info-->more accurate estimate.
            (square root makes it Expensive: have to quadruple n to make m half as  big)
    So smaller m can be achieved only by
        » accepting lower confidence level (smaller C),
        » or by increasing sample size (bigger n).

Visual example: Author website, whfreeman.com/scc, Applet: Confidence Intervals.
      You can change the C, with the same xbars, see the m change.

    Sigma:  We can't change it, it comes with the population.  But bigger sigma (more population variability) will give bigger m (wider CI), i.e. less accuracy in prediction (for the same C and n).
Science  projects directed by Prof. Wahl:  Experiments on chickens bred to be "identical"--very low variability from one to the other.  Therefore very small samples suffice.

Planning ahead:  Choose sample size big enough to satisfy desired: margin of error, confidence level.
    Given C and m (and sigma), find n.
        Method:  Use C to find z*.  Plug in to formula for m, and solve for n.  Or memorize formula for n and plug in to it.
          n = (z* sigma / m)2   Note:  z* sigma still on top.  m and n change places, and whole thing is squared!
            Round up!  If you get n = 5.06, you need a sample of size 6 to get your margin of error at least as short as you want.
~~~~~~~~~~~~~~~~
Start here Monday. Sec 6.2: "Significance tests use an elaborate vocabulary, but the basic idea is simple: an outcome that would "rarely" happen if a claim were true--is good evidence that the claim is NOT true." (p.314)


Activstats for Ch. 6: DO Ch 16 if you didn't.  P. 16-3 is the same as 16-2 with an added activity, + some new comments.
 Ch. 17 (optional)  introduces Confidence intervals. (Alternative, read Moore 6.1)  It can give you some
 good reinforcement for what we are doing.  But many instructors (me included) think his  demos with ybar in the center of the
 distribution may give a misleading sense of the definition of confidence interval (tho his words are right.).  So if you work with
 this chapter, be sure to do p. 17-4 (Appendix) after p. 17-1.
Activstats Ch. 19--Hypothesis tests.  It's good, but not in exactly the order we're doing it.Optional
Notes for p. 19-1: Act uses "hypothesis test" for what Moore calls "significance test".  This is common.
activity 1:  Try it again.  Each time it will reset the bar proportion to a different value.
activity 2:  Reasoning of hypothesis test. I think it's good, except it uses "unlikely" in the sense of "not very believable" when it's talking about P(Red) = .50. , and we are trying to use "likely" only about random phenomena.  It's not a random phenomenon whether P(Red) = .50.  In any particular experiment, it either is, or isn't, and if I peek inside the computer program I can tell you which--no probability about it.
activity 3:  Here the use of "unlikely" is correct--we're taking a sample from the population, so our proportion is a random phenomenon.
activity 4: A succinct exposition of the reasoning and terminology.
p. 19-2 ACT does only 2-sided tests.  But it's good.
activity 1 is video--therapeutic touch
activity 2 is data on above.  p=.4something
  activity 3 gives a chance to organize the info and do it
activity 4 introduces alpha level..
activity 5 relates C.I. and hypoth test but I think it's confusing:  "C.I.contains those values for which we would not reject Ho."  Skip it if you like.
p. 19-3 Goes through actually doing tests, p-value and alpha.

Sievers home  Math151-Fall02/Day-30.htm  2:30pm 11/8/02
This page belongs to Sally Sievers who is solely responsible for its content. Please see our statement of responsibility.