This preview shows page 1. Sign up to view the full content.
Unformatted text preview: he alternative hypothesis—use the arrow keys to move the cursor left or
right to select the correct form of your alternative hypothesis. Press [ to select it.
8. Select Calculate and press [. The screen repeats the form of the alternative hypothesis selected. The test statistic, t, and the
pvalue, p, are reported, along with the sample’s summary statistics.
As before, you can also select the option Draw to see the test results. Inference for Two Sample (Independent) Means
Note: These procedures will ask if you want to use the pooled standard deviation procedures—these are
rarely used, so use them only if your instructor tells you to.
To make a confidence interval for the difference between two population means:
1. Press K and scroll to the TESTS menu.
2. Scroll down until you ﬁnd the option 2–SampTInt and select it.
3. If you have the data, select the Data option, and enter the list names with the data. See the top screen
to the right.
4. If you have only the summary statistics, choose the Stats option and enter the summary statistics
(sample mean and standard deviation, and sample size) for each of your samples. See the second
screen to the right.
5. Determine your conﬁdence level.
6. If you are using the pooled procedures, select Yes; otherwise, select No. You may have to scroll down to
see this line.
7. Scroll down and select Calculate. Press [. Displayed is the interval for the difference m1  m2. Other
details of the calculation are also displayed: the degrees of freedom and all the summary statistics. You need
to scroll down to see them all.
To conduct a hypothesis test:
1. Write down your null and alternative hypotheses.
2. Press K and scroll to the TESTS menu.
3. Scroll down until you ﬁnd the option 2–SampTTest and select it.
4. If you have the data, select the Data option and enter the list names with the data. This is shown in
the screen to the right. 5 7.
8. Inference for Two Sample (Paired) Means
The following instructions assume that one set of data is in L1 and the other in L2 and that the paired difference we will compute is
L2– L1. Alternatively, you can calculate the differences by hand and store the differences directly in a list, say L3.
To have the calculator compute the paired differences, L2 –L1:
1. Press K, select the Edit menu, and input your data into lists L1 and L 2.
2. Clear L3 by moving the cursor up to highlight L3 and pressing P.
3. Move the cursor back to highlight L3.
4. Type in the formula L2 – L1 by pressing this sequence of keys: F 2 c F 1. See the
screen to the right.
5. Press [.
You should now see the calculated differences stored in L3.
To ﬁnd a conﬁdence interval or conduct a hypothesis test on the paired mean difference, follow the instructions under
Confidence Intervals and Hypothesis Tests for a Mean using the differences stored in L3. Chi Squared GoodnessofFit Test
Note: Several versions of the TI83/84 calculator have this test builtin. If yours does not, skip the ﬁrst
set of instructions and move on to the second.
These instructions assume that you have the list of observed counts in L1 and the list of expected counts
(generated from the probabilities in your null hypothesis) in L2.
To perform a test with the builtin function:
1. Press K and scroll to the TESTS menu.
2. Scroll down until you ﬁnd the option X 2 GOF–Test and select it. If you don’t see this, move to the next set of procedures.
3. Enter the list names with your observed and expected counts and the degrees of freedom for the test (the number of categories –1).
4. Select Calculate and press [. The results display the test statistic, x2, and the pvalue, p, along with some other information.
To perform the test step by step:
1. After entering the observed and expected counts into L1 and L2, respectively, move the cursor to
highlight the list name L3. Clear this list and return the cursor to highlight the name L3.
2. With L3 highlighted, type the formula for the test statistic using the following keystrokes:
(F1cF2)VdF2
The result is shown in the screen to the right. Press [. L3 should ﬁll with the calculated values.
3. Press F G to QUIT the list edit screen.
4. Press F K to enter the LIST menu, and scroll to the MATH menu. Select option 5:sum(.
5. Type in L3 as shown in the screen to the right. Press [. This is your chisquared test statistic.
6. Press F O to get to the DISTR menu, and select X 2 cdf(. Type in the value of the test statistic computed in step 5 (or use the ANS feature), enter a large number such as 99999, and enter the
degrees of freedom for the test, all separated by commas. Press [. This is your pvalue. 6...
View
Full
Document
 Spring '14

Click to edit the document details