MATH 251, P&S I, Fall 2007, Mon. Nov. 19, Day 37 hit reload.after class.

Reading:  Start 7.2, thru p. 497.   Read the rest, lightly, in self-defense (no HW required on it, exactly)
Hand in Yes: 

Sec. 7.2, two-sample, by hand.  A table is a good way to organize the work, see example
 For problems involving calculating a CI and/or a test, give the Difference and  SEDiff as well as the answers asked for.   You can check your hand calculations using the Excel Two-sample calculator.
7.83 iron deficiency  (FYI, most US pediatricians recommend iron supplements for both...)
7.82 cocaine & birthweight
7.68&69, 70, 71. Bread Read them and do only this: Do 7.68a, b.  (You need to calculate 2 variances "by hand". Since n = 2, it's not too onerous with a simple calculator; but you may use any aid you like.)     Tell what analysis you would do for 7.69.   Tell whether the test in 70 will be significant, without computing anything.  Do 7.71.
7.66 flat screens  Read a, remember how to do it. Do b, c. 
    +7.67 new screens now
7.59 what's wrong?
7.60, 61 short answer questions
7.57 soft drink size--missing information  (A frustrating thing about published info--like this, they often leave out what you need to check up on them, investigate further.)

 Read, 
discuss
 
 

 

 Optional
(more practice)
Exam 3 Due today by 3pm under my door or in my hand.
Quiz back. Nine 9/10 or above, two 8/10, 3 lower. 
Homework questions:  Day 36
~~~~~~~~~~~~~~~

Sec. 7.2, Comparing two means"Two-sample tests". Notes Day 36
Summary:  Two SRS's, independent, from distinct  populations. (Populations are normally distributed) Robust? Yes...but...p. 493

We use the Difference of the two x-bars,  diff =  xbar1 - xbar2 = .
 The Standard Error of the difference  xbar1 - xbar2 :
 substitute sample standard deviations for population standard deviations, and our z's are replaced by t's.
<>For testing, if Ho is "population means are equal"
"Two-sample t-statistic" <>
For doing by hand:  df = smaller of (n1- 1) and (n2- 1).  Will give a "conservative" result.
From a computer:  df = complicated formula on p. 498.  Produces non-integer degrees of freedom.  Very good if both sample sizes are at least 5. Unsuitable for doing by hand.
CI for µ1 - µ2, difference of means,  is 

--SPSS will do our computations when we are given raw data. Next. Handout

Sievers home  Math251-Fall07/Day2s37.htm  3:30pm   11/19/07
This page belongs to Sally Sievers who is solely responsible for its content. Please see our statement of responsibility.