CI and Testing to compare Means

NEW TYPE OF QUESTION

Test whether men at AASU have a mean travel distance that is greater than the mean travel distance for women. Use a = .01. Collect data on 3 men and 3 women. Ho: mmen = mwomen Ha: mmen > mwomen

Note: these are really too small samples.

THEORY:

______is a t-score

when you compare it to a t-value in the t-table A-53

your df =

______

EEEKK!! Let XL do it.

This number is always between

((Smallest n) –1) and (the sum of the n’s – 2)

Let’s use the steps I proposed last time…..

1.Read the Problem

A. Ha: is < or > then it is a one-tail test

Ha: is a not equal to sign then it is a two-tail test

B.  Note what a value is shown in the problem

C.  Is the data continuous? You will use the t-tables

D.  Work with your data.

Find the sample means and look around for s the standard deviation of the data

E. Calculate df the degrees of freedom by computing

n – 1, i.e. the sample size = number of pieces of data and subtract 1 from it OR

crazy df if comparing means

(approximately <min of n’s-1> to <sum of n’s-2>

2.  compute a t-score

( - m)

------where m is the ‘guessed’ mean

mentioned in the hypotheses and s was

s / given to you in the problem

or for this new situation

/

3. go to the t-tables in A-53

A.Look for 2-tail or one-tail

B. Look for a...... so now you know which column

C. Find the correct row by using your df

D. Pluck off your t-value

E. Put a negative in front of your t-value if there is a < in your Ha hypothesis

4.COMPARE

A.If Ha is < and your t-score < - t-value then “go with”

Ha

B. If Ha is > and your t-score > - t-value then “go with”

Ha

C. If Ha is not equal to sign then if either A. or B. is true then “go with” Ha.

The t-tables are found in text

Upper tail probability

df .10 .05 .025 .01…

1 3.078 6.314 12.706 31.821….

2 1.886 2.920 4.303 9.925 ….

3 1.638 2.353 3.182 4.541 …

. . . . ….

45

50 1.299 1.676 2.009 2.403

. . . . .

1000 1.282 1.645 1.960 2.326

------

80% 90% 95% 98%

Confidence level

Now let’s instead build a 90% confidence interval for the difference in the mean number of miles traveled to campus by men and women.

+- t*

6