TA1-2

1. You can type commands in the Session window. To do so, you must first enable the command language:

   • Click in the Session window
   • Click “Editor,” “Enable commands”

   This will produce a “MTB>” prompt in the Session window. At this prompt, you can then type desired commands.

2. Make a sequence of selections beginning in the toolbar menu. For example, in this chapter, we produce a graph known as a boxplot. To create this graph, you would click “Graph” and then select “Boxplot.” In this appendix, such a sequence of selections will be presented as Graph ➔ Boxplot. Once the sequence of selections has been made, dialog and/or option boxes will be encountered that allow you to indicate which variable(s) will be part of the analysis, along with other information. If further help is needed, you can click the “Help” button that appears with every dialog box. Once all appropriate information is provided, click the “OK” button to get the desired output.

1. With pretabulated frequencies, the spreadsheet should have two columns with a column name in the top row: one column should have the category names, and the other column should have the total counts of each category.
2. Select all cells, including the column names.
3. Click the “Insert” tab and click “PivotTable” in the Tables group.
4. Click “PivotChart” (this will create a “PivotTable Field List” box).
5. Select the field(s) presented to you by clicking a checkmark next to the name(s).
6. A bar graph will be created automatically.

1. Follow the steps for making a bar graph.
2. Change the created bar graph into a pie chart by clicking the “Design” tab, then click “Change Chart Type” in the “Type” group, then select the “Pie” chart type.

1. Create a bar graph, as described above.
2. You will find in the spreadsheet a PivotTable report made up of two columns: (1) a column labeled “Row Labels” and (2) a column with the frequencies. Highlight the contents of the report (that is, the cells with the category names and the cells with the frequencies).
3. Click the “Data” tab and then click “Sort” in the “Sort & Filter” group.
4. Choose the “Descending (Z to A)” option and select the column associated with the frequency numbers in the menu box found immediately below the option.
5. To convert the counts into percent, click the field name found in the “ Values” section, select the “Value Field Setting” option, click the “Show Values As” tab, then finally select “% of total” from the Show Values As menu and click “OK.”

1. Select “Histogram” in the Data Analysis menu box and click “OK.”
2. Enter the cell range of the data into the “Input Range” box. If you want Excel to automatically select the classes, leave the “Bin Range” box empty.
3. Place a checkmark next to the “Chart Output” option. Click “OK.”

1. Click and drag the mouse to highlight the cell range of the data you wish to time plot (include the column name if you wish it to appear as a chart label).
2. With the cell range highlighted, click the “Insert” tab and then click “Line” in the “Charts” group.
3. Within the 2-D Line choices, you can choose whether to have data symbols at the data values or not.

1. Analyze ➔ Distribution
2. Select variables of interest and click “Y” to cast variables into that role.
3. Click “OK.”

1. Graph ➔ Graph Builder
2. Drag a nominal or ordinal variable of interest to the X axis.
3. Click the bar chart icon in the toolbar.

1. Click the red triangle next to the variable name.
2. Select Order by ➔ Count descending.

1. Analyze ➔ Quality and Process ➔ Pareto Plot
2. Select nominal or ordinal variable of interest, and click “Y, Cause” to cast variables into that role.
3. Click “OK.”
4. Click the red triangle and select “Pie Chart.”

1. Graph ➔ Graph Builder
2. Drag a nominal or ordinal variable of interest to the X axis.
3. Click the pie chart in the toolbar.

1. Analyze ➔ Distribution
2. Select continuous variables of interest and click “Y” to cast variables into that role.
3. Click “OK.”
4. If you wish to change the automatically selected classes, use the Graph ➔ Graph Builder option; drag the variable of interest to the X axis, then select the histogram icon. Click on a value on the axis, then change the minimum, maximum, and/or the bin width (Increment) and click “OK.”

1. Analyze ➔ Distribution
2. Select continuous variables of interest and click “Y” to cast variables into that role.
3. Click “OK.”
4. Click the red triangle next to a variable’s name and select “Stem and Leaf.”

1. Analyze ➔ Modeling ➔ Time Series
2. Select outcome variable, and click “Y, Time Series” to enter that variable.
3. If a time variable is available, enter it into “X, Time ID” (if you do not specify a time variable, JMP will order and label the time plot by row).
4. Click “OK.”

1. Graph ➔ Graph Builder
2. Drag the time variable to the X axis.
3. Drag a continuous outcome variable to the Y axis.
4. Click the line chart (next to the bar chart icon) in the toolbar.

1. Graph ➔ Bar Chart

2. If the frequencies have been pretabulated, select “Values from a table” from the Bars represent menu.

   If the frequencies have not been tabulated, select “Counts of unique values” from the Bars represent menu. Select “Simple” for the type of bar graph, then click “OK.”

3. For pretabulated frequencies, click-in the data column into the “Graph” variables box and click-in the column that has the names of the categories into the “Categorical” variables box.

   If the frequencies have not been pretabulated, click the column that has data on the categorical names that need to be counted into the “Categorical” variables box.

4. Click “OK.”

1. Follow the instructions for a bar chart.
2. Click “Chart Options.”
3. Check the box next to Decreasing Y.

1. Graph ➔ Pie Chart

2. If the frequencies have been pretabulated, select the “Chart values” from a table option.

   If the frequencies have not been pretabulated, select “Chart counts” of unique values option.

3. If the frequencies have been pretabulated, click-in the data column into the Summary variables box, and click the column that has the names of the categories into the Categorical variables box.

   If the frequencies have not been pretabulated, click-in the column that has data on the categorical names that need to be counted into the Categorical variables box.

4. If you wish to have the pie slices labeled by categorical names and have percents reported (as in Figure 1.1(b) in your text), click “Label,” then click the “Slice Labels” tab and place checkmarks next to the desired labels.

1. Graph ➔ Histogram
2. Select “Simple” for the type of histogram, then click “OK.”
3. Click-in the data column into the “Graph Variables” box and then click “OK.”
4. If you wish to change the automatically selected classes, double-click on the horizontal axis to make the “Edit Scale” box appear. Now, click the “Binning” tab and then choose the “Midpoint/Cutpoint positions” option found in the “Interval Definition” section. Depending on whether you choose the Interval type as “Midpoint” or “Cutpoint,” you then give the desired values of the midpoints (that is, the middle values of the classes) or the cutpoints (that is, lower and upper values of the classes).

1. Graph ➔ Stem-and-Leaf
2. Click-in the data column into the “Graph Variables” box and then click “OK.”

1. Graph ➔ Time Series Plot
2. Select “Simple” for the type of time series plot, then click “OK.”
3. Click-in the data column into the Series box.

1. Analyze ➔ Descriptive Statistics ➔ Frequencies
2. Select the variable of interest on the left, then click the right arrow to move the variable to the right.
3. Click “Charts” and select “Bar Chart,” then click “Continue.”
4. Click “OK.”

1. Analyze ➔ Descriptive Statistics ➔ Frequencies
2. Select the variable of interest on the left, then click the right arrow to move the variable to the right.
3. Click “Charts” and select “Pie Chart,” and select “Percentages” under “Chart Values.” Click “Continue.”
4. Click “OK.”

1. Analyze ➔ Quality Control ➔ Pareto Chart
2. Select “Simple” and click “Define.”

3. Select the variable to plot, then click the arrow next to “Category Axis” to move the variable to the category axis section.

   TA1-12

4. To add a title to your plot, click the “Titles . . . ” button. A new box will appear to enter titles or subtitles. Click “Continue” to close this dialog box.
5. Click “OK.”

1. Analyze ➔ Descriptive Statistics ➔ Frequencies
2. Select the variable of interest on the left, then click the right arrow to move the variable to the right.
3. Click “Charts” and select “Histogram.” Click “Continue” and “OK.”
4. To change the binning (bar scaling), double-click in the graph for the “Chart Editor,” then click in a bar of the graph for “Properties.” Select the “Binning” tab. Move the radio button to “Custom” and enter either a number of bars (bins) or a bin width. To change the maximum or minimum X value, click on “X” in the tool bar, and then the “Scale” tab. Uncheck the box under “Automatic” next to the value you want to change and enter the new value. Click “Apply” and “Close.”

1. Analyze ➔ Descriptive Statistics ➔ Explore
2. Select the variable of interest on the left, then click the right arrow next to “Dependent List” to move the variable to that section.
3. Click “OK.”

1. Analyze ➔ Forecasting ➔ Sequence Chart
2. Select the variable of interest on the left, then click the right arrow next to “Variables” to move the variable to that section.
3. If you have a variable identifying time, select it and click the right arrow next to “Time Axis Labels.”
4. Click “OK.”

1. Graph ➔ Legacy Dialogs ➔ Scatter/Dot
2. Select “Simple Scatter” and click “Define.”
3. Select outcome variable and click the right arrow next to “Y Axis.”
4. Select time variable and click the right arrow next to “X Axis.”
5. Click “OK.”
6. Double-click the scatterplot in the output window to open the editor.
7. In the toolbar, select the “Interpolation Line.”
8. Close the editor to finalize the graph.

1. Graphics ➔ Bar Chart With Summarized Data
2. For “Labels,” select the column identifying groups.
3. For “Heights,” select the frequency variable.
4. Add a title and X and Y axis labels, if desired.
5. Click “Calculate.”

1. Graphics ➔ Bar Chart With Raw Data
2. For “Sample,” select the column of interest; to avoid many “short” bars, you can enter a value in Cutoff that will gather all categories with frequencies less than the specified value into an “Other” category.
3. Add a title and X and Y axis labels, if desired.
4. Click “Calculate.”

1. Graphics ➔ Pie Chart With Summarized Data
2. For “Labels,” select the column identifying groups.
3. For “Sizes,” select the frequency variable.
4. Add a title, if desired.
5. Click “Calculate.”

1. Graphics ➔ Pie Chart With Raw Data
2. For “Sample,” select the column of interest; to avoid many small slices, you can enter a value in Cutoff that will gather all categories with frequencies less than the specified value into an “Other” category.
3. Add a title and X and Y axis labels, if desired.
4. Click “Calculate.”

1. Graphics ➔ Histogram
2. For “Sample,” select the column of interest.
3. If desired, specify number of bins, bin width, start point, title, and axis labels.
4. Click “Calculate.”

1. Graphics ➔ Stem and Leaf
2. For “Sample,” select the column of interest.
3. If desired, enter a title.
4. Click “Calculate.”

1. Graphics ➔ Scatter Plot
2. For “X,” enter time variable which must be numeric, such as day, year, or an index (1, 2, …, n).
3. For “Y,” select the variable of interest.
4. In the Parameters section, change “Points” to “Lines” or “Both.”
5. Click “Calculate.”

1. Press to select “Edit.”
2. In L1 enter sequential values (1, 2, 3…) up to as many categories you have.
3. Enter the values associated with each category in L2.
4. Press , then set the Xmin and Xmax to match the values in L1, and adjust Ymin and Ymax to be an appropriate range for your Y variable.
5. Press = STAT PLOT, turn the plot “On” if needed by using the and pressing to move the highlight. Select the histogram .
6. Select L1 for Xlist and L2 for Freq.
7. Press .

1. Press = STAT PLOT, and select the histogram.
2. Select a plot (press to select Plot 1).
3. Enter the name of the list that contains the data by pressing , and so on.
4. Get an initial histogram by pressing .
5. Adjust the windowing (if needed) using . Reset Xmin, Xmax, Xscl (the bar width), and Ymax as needed.
6. Press .

1. Press to enter the list editor.
2. In L1 enter time values (or an index from 1 to n).
3. In L2 enter data for outcome variable.
4. Press = STAT PLOT and select the connected scatterplot
5. Select L1 for Xlist, and L2 for Ylist.
6. Press .

1. Select “Descriptive Statistics” in the Data Analysis menu box and click “OK.”
2. Enter the cell range of the data into the Input Range box.
3. Place a checkmark next to the “Chart Output” option. Click “OK.”
4. If you wish to compute the first and third quartiles, click an empty cell in the spreadsheet and then proceed to the statistical function menu as described in the Excel overview section of this Appendix. Scroll down the list of functions and double-click on the QUARTILE function choice. In the “Array” box, input the cell range of the data. In the Quartile box, input the value “1” to get the first quartile or the value “3” to get the third quartile and then click “OK.”

1. Analyze ➔ Distribution
2. Select variable(s) of interest and click “Y” to cast into that role.
3. Click “OK.” This gives the mean, standard deviation, and five-number summary. For other statistics, click the red triangle next to “Summary Statistics” and check the ones you want.

1. Analyze ➔ Distribution
2. Select variable(s) of interest and click “Y” to cast into that role.
3. Click “OK.”

1. Graph ➔ Graph Builder
2. Drag a continuous variable of interest to the X or Y axis.
3. Click the boxplot icon in the toolbar.

1. Stat ➔ Basic Statistics ➔ Display Descriptive Statistics
2. Click in the data column(s) for which you want numerical summaries into the Variables box.
3. Click the Statistics button to select additional options, then click “OK” to close the pop-up.
4. Click “OK” to have the summaries reported in the Session window.

1. Graph ➔ Boxplot
2. For a single variable, select “One Y Simple,” then click “OK.” If you have multiple boxplots that you want to display together, select “Multiple Y’s Simple” (if the variables are in separate columns), or “One Y With Groups” (if there is a column with the group identifier). Click “OK.”
3. Click in the data column(s) for which you want to construct boxplots into the Graph variables box. Click “OK.”

1. Analyze ➔ Descriptive Statistics ➔ Explore
2. Select the variable of interest on the left, then click the right arrow next to “Dependent List” to move the variable to that section.
3. Click “OK.”

1. Analyze ➔ Descriptive Statistics ➔ Explore
2. Select the variable of interest on the left, then click the right arrow next to “Dependent List” to move the variable to that section.
3. Click “OK.”

1. Statistics ➔ Descriptive Statistics
2. Place a check box next to each variable of interest.
3. If you want percentiles other than the standard five-number summary, enter those in the box.
4. Click “Calculate.”

1. Graphics ➔ Box Plot
2. Check the box next to each variable you want to visualize.
3. Enter a title and X and Y axis labels, if desired.
4. Click “Calculate.”

1. Enter the data in a list, using to access the list editor.
2. Press , and to CALC. Select option 1:1-Var Stats. This will add the command to the home screen.
3. Enter L1 (as or the list with your data) after “1-Var Stats” and press enter.
4. Press the down arrow to reveal the five-number summary.

1. Press = STAT PLOT and select the box plot or . The first does not identify potential outliers using the 1.5IQR criteria; the second does.
2. Enter L1 (as ) or the list with your data as Xlist.
3. Press .

1. Click an empty cell in the spreadsheet, then Formulas ➔ More Functions, Statistical. Scroll down to find NORM.DIST.
2. To find the area to the left of the point of interest for any Normal distribution, enter the value, the mean (), the standard deviation (), and “1” for Cumulative. Click “OK.”
3. For a standard Normal (Z) distribution, use NORM.S. DIST. Enter the z-score and “1” for Cumulative. Click “OK.”
4. To do backward Normal calculations, use NORM.S.INV for standard Normal (enter the probability to the left of the point, that is, ; for any Normal distribution, use NORM.INV and enter the probability to the left of the point, the mean (), and the standard deviation (). Click “OK.”

1. Help ➔ Sample Data ➔ Teaching Scripts (in Teaching Resources) ➔ Interactive Teaching Modules ➔ Distribution Calculator
2. Enter a mean and standard deviation if not standard Normal and click “Reset Distribution.”
3. Select the type of calculation desired (input values to obtain a probability or enter a probability to obtain a value—the inverse calculation),
4. Enter the value(s) type to the right, enter the value(s) of interest, and press the “Enter” key to see shading on distribution.

1. Analyze ➔ Distribution
2. Select variables of interest and click “Y” to cast variables into that role.
3. Click “OK.”
4. Click the red triangle next to the variable name, then select “Normal Quantile Plot.”

1. Graph ➔ Probability Distribution Plot
2. Select “View Probability” and click “OK.”
3. Enter values for the mean and/or standard deviation.
4. Click the “Shaded Area” tab. If you want to find the area under the curve associated with a specified value, select the “ Value” option. You can choose to find the area to the left or right of that specified value or even between two values by clicking the appropriate picture.
5. Enter the specified value(s) in the “ value” box. Click “OK.”

1. Calc ➔ Probability Distributions ➔ Normal
2. Choose the “Cumulative probability” option if you wish to find the area to the left of a specified value. Choose the “Inverse cumulative probability” option if you wish to find the value associated with a specified area to the left of that value.
3. Select the “Input constant” option.
4. Enter the specified value of or or enter the specified area. Click “OK” to find the results reported in the Session window.

1. Graph ➔ Probability Plot
2. Select “single” (for only one variable) and “OK.”
3. Click in the data column of interest into the Variable box and then click “OK.”

1. Transform ➔ Compute Variable
2. In “Target Variable” enter a name for the new variable.
3. To find a cumulative probability, select “CDF and Noncentral” then “CDF.normal” (this will enter the formula in the editor). Enter the desired parameters in to the formula (quantity to look up, mean, and standard deviation). Click “OK.”
4. To find a variable value given a cumulative probability, select “Inverse DF” and then “IDF.Normal.” Enter the parameters and click “OK.”
5. A new column will be added to the dataset containing the result.

1. Analyze ➔ Descriptive Statistics ➔ Q-Q Plots
2. Select the variable of interest on the left, then click the right arrow to move the variable to that section.
3. Click “OK.”

1. Distribution Calculator ➔ Normal
2. Enter the mean and standard deviation (or accept defaults for standard normal),
3. To look up proportions, use the probability tab; to look up a variable value given a cumulative proportion, , use the quantile tab.
4. After entering value, click “Calculate.”

1. Graphics ➔ QQ Plot
2. For “Sample,” select the column of interest.
3. Click “Calculate.”

1. Press ➔ 2:normalcdf( and press (or just press).
2. To look up proportions from a Normal distribution with a mean other than 0, and standard deviation other than 1, enter the mean and standard deviation after entering the range. For example, normcdf(90,110,100,15) will look up the proportion between 90 and 110 for a Normal distribution with a mean of 100 and a standard deviation of 15. If you are using standard Normal, there is no need to enter the mean and standard deviation.
3. To use ShadeNorm, you need to size the window. An appropriate size is for the X axis. Y sizing will depend on the distribution. For standard Normal, will work well. The parameters for the command (low value, high value, , ) are the same.

1. Press ➔ 3:invNorm(and press (or just press).
2. Enter the proportion desired, a comma, the mean, the standard deviation and press . If you are using standard Normal, there is no need to enter the mean and standard deviation. For example, invNorm(.50) will return 0 (at , half of the distribution is to the left).

1. Enter the data in a list.
2. Press = STAT PLOT, select one of the three plots, and select the Quantile plot, .
3. Enter L1 (as , or the list with your data) as Xlist.
4. Select a data axis (your text uses the Y axis) and a mark type for each data point.
5. Press .