Distance Education Statistics Lab E Eric Parslow and Russell T. Hurlburt University of Nevada, Las Vegas

This lab will consist of 4 parts:
1. Using the "confide" program in ESTAT.
2. Using the "ttest1" program in ESTAT.
3. Using "datagen" in the computation of confidence intervals.
4. The quiz.




PART 1 - "confide"

1. In ESTAT, open the program labeled "confide".

2. Click on the "Tutorial" button and follow the directions in order to learn how to eyeball-estimate confidence intervals.

3. Practice using the "New Data Set" button.

4. Your eyeball error should be consistently under 10%.



PART 2 - "ttest1"

1. In ESTAT, open the program labeled "ttest1".

2. Windows ESTAT users, take Path 1. Macintosh and DOS ESTAT users, follow the 'shortcut.' That is, inspect the histogram and proceed directly to answer the 'Reject H₀, yes or no' question, skipping the intermediate estimations.

3. Your task is to estimate, by eyeball, whether or not the null hypothesis shoud be rejected. You do not need to perform the complete eyeball-estimation. (You may do so if you wish, but it is not required).

4. The null hypothesis is given at the bottom of the screen. The question is simply Yes or No, should the null hypothesis be rejected?

5. You should be able to estimate this fairly accurately without doing all the computations at the top of the screen.

6. Simply answer "yes" or "no" and 'continue' to the end of the problem see if your quick eyeball-estimation was correct.

7. Repeat this process with new data sets. After a few trials, you should begin to get a 'feel' for when a sample did or did not come from the population with the stated mean.

8. Within reason, you should get all of the problems correct (no one is absolutely perfect).

9. If you are having difficulty performing this task quickly and accurately, use the "Tutorial" in order to gain a stronger understanding of the procedures used in doing the quick eyeball-estimation.



PART 3 - 'The confidence interval and datagen'.

1. Open datagen in ESTAT.

2. Create the data set X = 12,15,21,13.

3. Windows ESTAT users: Ignore for the moment the fact that datagen computes the confidence interval limits automatically.

4. Recall that the equation for the limits of the confidence interval is .

5. Note that datagen provides X bar = 15.25 and the standard error 2.0155.

6. Look up t_cv in Table A.2 in the book. The confidence interval has n - 1 degrees of freedom.

7. Use your calculator to compute the confidence interval.

8. Your answer to #7 should be that the confidence interval is 8.84 < < 21.66. Windows ESTAT users: verify that those are indeed the values that datagen provides for the confidence interval minimum and maximum.

9. This was computed as follows:
(a) look up the critical value of t in table A.2 under the column for .05, a two-tailed test, with 3 degrees of freedom. t_cv = 3.182
(b) obtain the standard error, shown automatically by datagen (2.0155), or you could divide s (4.03) by the square root of n (n = 4 so the square root of n is 2)
(c) multiply the standard error (2.0155) by the critical value of t (3.182) to obtain 6.41.
(d) obtain the sample mean X bar from datagen (15.25)
(e) add and subtract 6.41 from the sample mean X bar (15.25) to obtain 8.84 and 21.66.

10. If this procedure is at all confusing, review the exercises for Chapter 8 in the book. Windows ESTAT users: You can also practice by making up your own data sets and computing the confidence interval, and comparing the values you obtain with those provided by datagen.


PART 4: THE QUIZ

Once you are comfortable with eyeball-estimating the confidence interval limits and the one-sample t-test, and are confident that you can use datagen to compute the confidence interval, you are prepared to take the quiz.


You will need the quiz password, which is
7654321