Professional Documents
Culture Documents
CALCULATOR
AND MINITAB
IN MATH 1080
STATISTICS
2
This manual will lead you through the use of the TI-83 and TI-84 graphing calculators
as you go through your Introductory Statistics course. Exact key strokes and
resulting screen shots are shown for selected problems.
Unless otherwise noted, the procedures are the same for both calculators. The few
differences between the TI-83 and the TI-84 calculators are noted and procedures
given for each one.
The software package Minitab® is used for functions that are not available on the
calculators (primarily graphs).
MINITAB
Graphs on Minitab…………………………………………………………. p 5
Histograms…………………………………………………………………… p 7
Labeling the x-axis……………………………………………….. p 8
Add your name………………………………….………………... p 8
Add data values…………………………………………………… p 9
Type (frequency, relative, etc)………………………………….. p 10
Printing……………………………………………………………………….. p 11
Polygons………………………………………………………………………. p 12
Stem-and-Leaf Diagrams…………………………………………………. p 16
Correlation & Regression on Minitab…………………………………. p 17
Correlation…………………………………………………………. p 17
Regression………………………………………………………….. p 18
Regression graphs………………………………………………… p 19
Prediction Intervals………………………………………………………… p 21
4
TI-83 & TI-84 Calculator Functions
Confidence Intervals
1 Proportion……………………………………………………….. p 48
1 Mean σ Known………………………………………………….. p 49
1 Mean σ Unknown………………………………………………. p 51
2 Proportions………………………………………………………. p 53
2 Means Independent Sample Large Samples…………….. p 55
2 Means Dependent Samples (Matched Pairs)…………….. p 57
Hypothesis Tests
1 Proportion………………………………………………………… p 59
1 Mean σ Known………………………………………………….. p 60
1 Mean σ Unknown………………………………………………. p 61
2 Proportions………………………………………………………. p 62
2 Means Independent Sample Large Samples…………….. p 63
2 Means Dependent Samples (Matched Pairs)…………….. p 64
Linear Regression……………………………………………………. p 67
Scatterplot………………………………………………………….. p 67
Regression equation……………………………………………… p 68
χ2 TEST…………………………………………………………………….. p 70
ANOVA ………………………………………………………….…………. p 72
5
GRAPHS ON MINITAB
Minitab is located under Math/CSci Programs
Open Minitab
In the Data window notice that there is a blank row above the numbered rows. This is called the Title Row. You may enter
titles or labels here and refer to the columns either by the column label (C1) or by the label you entered.
We will use the rent data from the class notes for chapter 2 to construct our histograms:
Data Set 2) Consider the following data giving monthly rents paid by a sample of 29 households in one city:
429 585 732 675 850 1020 620 750 660 540 578 956 1030
1070 930 871 765 880 975 650 1020 950 840 780 870 900
800 750 820
SORT
Sorting data is not needed for constructing the graphs.
It will, however, make it easier to count the class frequencies and to find the mode.
Data
Sort
In the window that comes up:
OK
You will now see the data sorted in increasing order in column C2
7
HISTOGRAMS
Graph
Histogram
Simple
Enter C1 in the window for Graph Variables
OK
Note the resulting histogram:
Histogram of RENTS
5
Frequency
0
400 500 600 700 800 900 1000 1100
RENTS
Note also that its default classes are not the ones we want.
To construct the classes we wish, double click anywhere on the x-axis.
Select the tab that says Binning
For Midpoints
MidPoint/CutPoint Positions
In the window enter the midpoints separated by a space:
449.5 549.5 649.5 749.5 849.5 949.5 12049.5
OK
For Lower Limits
CutPoints
In the window enter the lower limits separated by a space:
400 500 600 700 800 900 1000 1100
OK
For Upper Limts:
CutPoints
In the window enter the upper limits separated by a space
300 400 500 600 700 800 900 1000
OK
Note: In the window you may also enter the first two numbers and Minitab will extend that pattern throughout
the entire data set.
8
The resulting histogram, using midpoints:
Histogram of RENTS
5
Frequency
0
400 500 600 700 800 900 1000 1100
RENTS
Histogram of RENTS
5
Frequency
0
449.5 549.5 649.5 749.5 849.5 949.5 1049.5
MIDPOINTS
YOUR NAME HERE
Graph
Histogram
Simple
Enter C1 in the window for Graph Variables
Select:
Labels
Data Labels
Use y-value labels
OK
Histogram of RENTS
7
7
6
5 5
5
Frequency
4 4
4
3
3
2
1
1
0
449.5 549.5 649.5 749.5 849.5 949.5 1049.5
RENTS
YOUR NAME HERE
To produce the different types of histograms you must first construct the frequency histogram above.
Then:
PRINTING
File
Print Graph
Select the printer you want (for the room you are in) – in C3058 the printer in that room will already be the default selection.
Landscape mode should be the default – check it first.
Note: ALL graphs are to be printed in Landscape mode with your name in the lower left
corner of the graph
12
POLYGONS
To construct a polygon in Minitab you must first construct the corresponding histogram, then draw the lines segments, then
erase the histogram.
Using the RENTS data first construct the frequency histogram. Remember that polygons use midpoints. Also remember that
you need an extra midpoint on either side of the existing data values. You will have to manually enter all of the midpoints:
Histogram of RENTS
7
7
6
5 5
5
Frequency
4 4
4
3
3
2
1
1
0 0
0
349.5 549.5 749.5 949.5 1149.5
RENTS
YOUR NAME HERE
Before drawing the polygon we must first have the Drawing toolbar showing:
Tools
Toolbars
Select: GRAPH ANNOTATION TOOLS
This will add another toolbar immediately below the menu bar.
Begin on the left, click on each point of the polygon, moving from left to right.
Minitab will automatically connect each point to the previous point.
Double-click on the last point.
13
Histogram of RENTS
7
7
6
5 5
5
Frequency
4 4
4
3
3
2
1
1
0 0
0
349.5 549.5 749.5 949.5 1149.5
RENTS
YOUR NAME HERE
We will now remove the histogram bars. Double click in any of these histograms bars.
14
OK
(You may have the repeat this procedure if there are any bars still showing)
15
The resulting polygon (change the labels as in a histogram):
Polygon of RENTS
7
7
6
5 5
5
Frequency
4 4
4
3
3
2
1
1
0 0
0
349.5 549.5 749.5 949.5 1149.5
RENTS
YOUR NAME HERE
16
Graph
Stem-and-Leaf
Increment: enter 10
OK
Printing
File
Print Session Window
Check out the DotPlot and Bar Chart and Pie Chart options under the Graph menu.
17
1) Enter the Bill data in column C1. Include the label “Bill” above the column.
2) Enter the Tip data in column C2. Include the label “Tip” above the column.
CORRELATION
Minitab procedure for finding the correlation coefficient:
Select
Stat
Basic Statistics
Correlation
In the dialog box, enter the column labels of C1 C2 (in that order)
(you can do this simply by double-clicking on each one).
Click OK
REGRESSION
Select
Stat
Regression
Regression (again)
Make the following entries (careful which one is asked for first!):
C2 (or Tip) in box for “Response” variable.
C1 (or Bill) in box for “Predictor variable.
In the session window, there will appear a lot of information that we have not studied
yet. Find the information below, erase everything else (the session window is a word
processor), print your name in the session window, then use the print session
command to print the results.
Analysis of Variance
Source DF SS MS F P
Regression 1 93.14 93.14 8.73 0.042
Residual Error 4 42.66 10.67
Total 5 135.80
GRAPHS
Minitab procedure for the scatter diagram, graph of the regression line,
and prediction intervals.
Select
Stat
Regression
Fitted line plot
Amount of Tip
Tip = - 0.347 + 0.1486 Bill
17.5 S 3.26581
R-Sq 68.6%
R-Sq(adj) 60.7%
15.0
12.5
Tip
10.0
7.5
5.0
30 40 50 60 70 80 90 100 110
Bill
20
Select:
Editor
Add
Footnote
OK
Amount of Tip
Tip = - 0.347 + 0.1486 Bill
17.5 S 3.26581
R-Sq 68.6%
R-Sq(adj) 60.7%
15.0
12.5
Tip
10.0
7.5
5.0
30 40 50 60 70 80 90 100 110
Bill
YOUR NAME HERE
Prediction Intervals
Stat
Regression
Fitted line plot
Options
Select the box labeled Display prediction interval
Enter 95 (not .95) in the confidence level box
OK
OK
Amount of Tip
Tip = - 0.347 + 0.1486 Bill
30 Regression
95% PI
S 3.26581
R-Sq 68.6%
20 R-Sq(adj) 60.7%
10
Tip
-10
30 40 50 60 70 80 90 100 110
Bill
Note: For the assignment you will only have to print out this final graph.
22
To find the prediction interval for a specific value:
Stat
Regression
Regression (again)
Response: Tip
Predictor: Bill
Options
Prediction intervals for new observations:
87
Confidence level: 95
OK
OK (again)
In the Session window you will find that the 95% prediction interval for the tip
on a bill of $87 is: $8.44 to $16.73
In the assignment, find this information for both values asked for, erase the rest and print.
23
GRAPHING
CALCULATORS
24
Key Sequence:
Example 1:
Actors: 32 51 33 61 35 45 55 39 76 37 42
40 32 60 38 56 48 48 40 43 62 43
42 44 41 56 39 46 31 47
Actresses: 80 26 41 21 61 38 49 33 74 30 33
41 31 35 41 42 37 26 34 34 35 26
61 60 34 24 30 37 31 27
After entering the last number you can return to the Home (calculator) screen:
QUIT which is
25
SORTING A LIST
Sorting a list is not needed for performing calculations.
It will make reading the list and finding the mode easier.
To view the lists return to the list editor through the following sequence:
You will then see the lists are in ascending (increasing) order:
26
ClrAllLists
You may also clear all the lists at one time by using the ClrAllLists function.
From the Home screen you mudst first find the Catalog – on the 0 button – then find
the command ClrAllLists:
Note: Use this command carefully – it clears EVERY list in the calculator. If you
wish to save one or more lists then use the routine on the previous page to clear the
individual lists that you do not want.
28
1-VARIABLE STATISTICS
Enter the Best Actress data from page 23 of these notes in list L1.
The down arrow on the bottom line of the screen indicates that there is more data
below, access this using the down arrow keys:
Note: These last five numbers are the numbers needed for a 5-number summary
Note: The calculator will NOT find the mode – you must go back to the List
editor and scan the list yourself to find the mode.
29
Enter the Class Midpoints in list L1 and the Frequencies in list L2.
We then use the 1-Var stats function as above, but attaching the frequency list.
Note that the Midpoint list is first then the Frequency list:
Note: This arrow indicates there is more information below – use the down
arrow key to view.
30
For a basic box and whisker diagram first turn Plot 1 On then select the 5th small
diagram under Type:
31
Draw the boxplot diagram with:
MULTIPLE BOXPLOTS
From the Home screen, and with the boxplot diagram from the previous page still
activated:
Note: For a modified boxplot, under Type: select the 4th small diagram. Note
that the calculator cannot distinguish between mild and extreme outliers.
PROBABILITY
Factorials:
The factorial symbol is found under the Math – PRB menu.
To calculate 5! and 8!
Enter the number 5 on the HOME screen, then attach the factorial symbol:
Enter the number 24, then attach the Permutation symbol, then enter the number 4:
Enter the number 24, then attach the Combination symbol, then enter the number 4:
35
Answers are displayed on the Home screen:
BINOMIAL PROBABILITY
Given that n = 12 and the probability of success is 0.8:
n = 12 and p = .8
The binomial function is found under the Dist menu on the key.
Individual Probabilities:
To find the probability that x = 5:
Results are:
In the problem on the previous page, find the probability of at most 7 successes:
Results are:
You must first enter the binomial function in the calculator’s Y= editor:
This establishes the initial value as starting at 0 and the step value also as 1.
Results are:
We will round off probabilities to the 3rd decimal place. Translating the above table
and rounding off gives the following results:
x P(x)
0 0+
1 0+
2 0+
3 0+
4 0.001
5 0.003
6 0.016
7 0.053
8 0.133
9 0.236
10 0.283
11 0.206
12 0.069
40
POISSON PROBABILITY
For this example p = .929.
The Poission function is found under the Distr menu (right below the binomial
functions)
Results are:
Results are:
41
Functions needed for these topics are found under the DISTR menu which is
Answer:
Since this example has no upper limit, and the calculator function requires an upper
limit we must enter a number that is “big enough” that the calculator thinks it is
infinite.
Any number larger than 10 standard deviations will be “big enough” however, to
establish a habit, we will use the largest exact number the calculator can use, which is
1 followed by 99 zero’s. In scientific notation on the calculator that is displayed as
E99.
Results are:
Note that the button on the calculator case shows EE but the screen shows E.
44
Finding Percentiles
To find a z-score corresponding to a particular area (to the left of that z-score):
Example: P35:
Results are:
We will be simply adding the mean and standard deviation to the functions used for
the normal distribution.
P( x < 130):
Results are:
P( x > 131.5):
Results are:
CONFIDENCE
INTERVALS
48
CONFIDENCE INTERVAL
FOR 1 PROPORTION
This problem uses the 1-PropZInt function found under the STAT – TETS menu.
The 99% confidence interval for the proportion is: 63.1% < p < 70.2%
49
Find the critical value for the z-distribution for a 96% confidence level:
Results are:
Confidence Intervals
We will use the function named ZInterval located in the STAT – TESTS menu.
Results are:
So the 95% confidence interval for the mean is $61,606 < μ < $72,794
Critical Values
Critical values use the invT( function located under the DISTR menu. However, this
function is only available on “newer” TI-84 calculators. It is not available on the TI-83
or on older TI-84 calculators. If you do not have this function you will have to use the
tables in the textbook to find critical values.
Results are:
Confidence Intervals
We will use the function named TInterval located in the STAT – TESTS menu.
Results are:
The Genetics and IVF Institute conducted a clinical trial of its methods for gender
selection of babies. Two methods were tested: XSORT to increase the chances of a
male baby and YSORT to increase the chances of a male baby. Of 325 couples using
the XSORT method 295 had girls. Of 51 couples using the YSORT method 39 had
boys. Construct a 95% confidence interval for the difference between the proportion of
girls and the proportion of boys.
The 95% confidence for the difference between the proportion of girls and the
proportion of boys is:
Example:
A study was conducted on the effects of marijuana use on college students. The
following results were obtained:
Items correctly sorted by light users: n = 64 = 53.3 s = 3.6
Items correctly sorted by heavy users: n = 64 = 51.3 s = 4.5
Using subscripts of 1 for light users and subscripts of 2 for heavy users:
We will construct a confidence interval for the differences between the pairs.
x 2 5 7 1 1
y 2 3 9 4 0
Move over the list L3 and move the cursor UP to highlight the name L3:
Note: To access an internal list in the calculator we use the Data selection
on the Inpt: line and the list name on the next line.
HYPOTHESIS TEST
1 PROPORTION
We will use the function 1-PropZTest found under the STAT – TEST menu.
Using the gender selection data on page 54 of these notes, test the claim that the
probability of a male baby is greater than 0.5 with the YSORT method.
Ho: p ≤ .5
Ha: p > .5 claim
You will then fill in the required information in the following screen:
Results are:
Ho: μ = .8535
Ha: μ ≠ .8535
Results are:
Given n = 21 = 132 s = 12 test the claim that the mean is less than 7 years.
The hypotheses:
Ho: μ=7
Ha: μ<7 claim
Results are:
Ho: p1 = p2
Ha: p1 > p2
Results are:
Ho: μ1 = μ2
Ha: μ1 ≠ μ2
x 9 6 11 11 3 5
y 13 12 14 10 4 12
Test the claim that there is no difference between the two data sets.
Ho: μd = 0 claim
Ha: μd > 0
Move over the list L3 and move the cursor UP to highlight the name L3:
Note: To access an internal list in the calculator we use the Data selection
on the Inpt: line and the list name on the next line.
LINEAR REGRESSION
DIAGNOSTICS
Before starting linear regression you must activate something called the Diagnostics
which you will find in the Catalog.
You will only have to do this ONCE. It will remain activated even when you turn your
calculator off. (if you change batteries you will have to activate it again)
SCATTER PLOT
NOTE: Before proceeding go to Y= and clear out any functions there. Then return
to the HOME screen.
68
Result is:
REGRESSION EQUATION
Result is:
The linear regression equation is: y = 2.86 x + 76.74 (to 2 decimal places).
69
We will now graph the linear regression equation on top of the scatter plot.
The following sequence will place the entire regression equation into the Y= location
(from the HOME screen):
Result is:
You can then find predicted y-values from the regression equation (using the value
function under the CALC menu on the TRACE button).
Note: When you are finished with your scatter plots go back and
deactivate the STAT PLOT function. The scatter plots may interfere with the normal
graphing function.
70
χ2 TEST
Test the following data for independence:
To perform a χ2 test you must first enter the numbers into a matrix. A matrix is a rectangular array of
numbers. The dimensions of a matrix are (the number of rows) x (the number of columns). The above
problem will need a 2 x 4 matrix.
Note: You do NOT have to clear the matrix first – you will replace anything already there.
The calculator places the expected values in matrix B. To view the expected values:
Use the arrow keys to scroll the screen to the right to see values off the screen.
72
ANOVA
Analysis of Variance is performed using the ANOVA function found under the STAT –
TESTS menu.
Subcompact 55 47 59 49 42
Compact 57 57 46 54 51
Midsize 45 53 49 51 46
Full-size 44 45 39 58 44