Concepts That You Should Know
Before you begin using JMP, you should be familiar with these concepts:
• Enter, view, edit, and manipulate data using JMP data tables.
• Select a platform from the Analyze, Graph, or DOE menus. Platforms contain interactive windows that you use to analyze data and work with graphs.
• Platforms use these windows:
‒ Launch windows where you set up and run your analysis.
‒ Report windows showing the output of your analysis.
• Report windows normally contain the following items:
‒ A graph of some type (such as a scatterplot or a chart).
‒ Specific reports that you can show or hide using the disclosure button 
.
.‒ Platform options that are located within red triangle menus 
.
.How Do I Get Started?
The general workflow in JMP is simple:
1. Get your data into JMP.
2. Select a platform and complete its launch window.
3. Explore your results and discover where your data takes you.
This workflow is described in more detail in “Understanding the JMP Workflow”.
Typically, you start your work in JMP by using graphs to visualize individual variables and relationships among your variables. Graphs make it easy to see this information, and to see the deeper questions to ask. Then you use analysis platforms to dig deeper into your problems and find solutions.
• The “Visualize Your Data” chapter shows you how to use some of the useful graphs JMP provides to look more closely at your data.
• The “The Big Picture” chapter shows you how to analyze distributions, patterns, and similar values in several platforms.
Each chapter teaches through examples. The following sections in this chapter describe data tables and general concepts for working in JMP.
Starting JMP
Start JMP in two ways:
• Double-click the JMP icon, normally found on your desktop. This starts JMP, but does not open any existing JMP files.
• Double-click an existing JMP file. This starts JMP and opens the file.
The initial view of JMP includes the Tip of the Day window and the Home Window on Windows; on Macintosh, the Tip of the Day and JMP Starter, and Home Window initially appear.
The JMP Starter window classifies actions and platforms using categories.
Figure 2.2 The JMP Starter
On the left is a list of categories. Click a category to see the features and the commands related to that category. The JMP Starter also lists JMP Pro features and platforms.
The Home Window helps you organize and access files in JMP.
Figure 2.3 The Home Window on Windows
To open the Home Window on Windows, select View > Home Window. On Macintosh, select Window > JMP Home. The Home Window includes links to the following:
• the data tables and report windows that are currently open
• files that you have opened recently
For more details about the Home Window, see the Get Started chapter in the Using JMP book.
Almost all JMP windows contain a menu bar and a toolbar. You can find most JMP features in three ways:
• using the menu bar
• using the toolbar buttons
• using the buttons on the JMP Starter window
About the Menu Bar and Toolbars
The menus and toolbars are hidden in many windows. To see them, place your mouse pointer over the blue bar under the window’s title bar. The menus in the JMP Starter window, the Home Window, and all data tables are always visible.
Using Sample Data
The examples in this book and the other JMP books use sample data tables. The default location on Windows for the sample data is:
C:/Program Files/SAS/JMP/13/Samples/Data
C:/Program Files/SAS/JMPPro/13/Samples/Data
C:/Program Files/SAS/JMPSW/13/Samples/Data
The Sample Data Index groups the data tables by category. Click a disclosure button to see a list of data tables for that category, and then click a link to open a data table.
Macintosh sample data is installed in /Library/Application Support/JMP/13/Samples/Data.
Opening a JMP Sample Data Table
1. From the Help menu, select Sample Data.
2. Open the Data Tables used in Discovering JMP list by clicking on the disclosure button next to it.
3. Click the name of the data table to use it in the examples in this book.
Sample Import Data
Use files from other applications to learn how to import data into JMP.
The default location on Windows for the sample import data is:
C:/Program Files/SAS/JMP/13/Samples/Import Data
C:/Program Files/SAS/JMPPro/13/Samples/Import Data
C:/Program Files/SAS/JMPSW/13/Samples/Import Data
Understanding Data Tables
A data table is a collection of data organized in rows and columns. It is similar to a Microsoft® Excel® spreadsheet, but with some important differences that are discussed in “How is JMP Different from Excel?”. A data table might also contain other information like notes, variables, and scripts. These supplementary items are discussed in later chapters.
Open the VA Lung Cancer data table to see the data table described here.
Figure 2.4 A Data Table
A data table contains the following parts:
Data grid
The data grid contains the data arranged in rows and columns. Generally, each row in the data grid is an observation, and the columns (also called variables) give information about the observations. In Figure 2.4, each row corresponds to a test subject, and there are twelve columns of information. Although all twelve columns cannot be shown in the data grid, the Columns panel lists them all. The information given about each test subject includes the time, cell type, treatment, and more. Each column has a header, or name. That name is not part of the table’s total count of rows.
Table panel
The table panel can contain table variables or table scripts. In Figure 2.4, there is one saved script called Model that can automatically re-create an analysis. This table also has a variable named Notes that contains information about the data. Table variables and table scripts are discussed in a later chapter.
Columns panel
The columns panel shows the total number of columns, whether any columns are selected, and a list of all the columns by name. The numbers in parentheses (12/0) show that there are twelve columns, and that no columns are selected. An icon to the left of each column name shows that column’s modeling type. Modeling types are described in “Understand Modeling Types” in the “Analyze Your Data” chapter. Icons to the right show any attributes assigned to the column. See “View or Change Column Information” in the “Work with Your Data” chapter for more information about these icons.
Rows panel
The rows panel shows the number of rows in the data table, and how many rows are selected, excluded, hidden, or labeled. In Figure 2.4, there are 137 rows in the data table.
Thumbnail links to report windows
This area shows thumbnails of all reports based on the data table. Place your mouse pointer over a thumbnail to see a larger preview of the report window. Double-click a thumbnail to bring the report window to the front.
Interacting with the data grid, which includes adding rows and columns, entering data, and editing data, is discussed in the “Work with Your Data” chapter. If you open multiple data tables, each one appears in a separate window.
Understanding the JMP Workflow
Once your data is in a data table, you can create graphs or plots, and perform analyses. All features are located in platforms, which are found primarily on the Analyze or Graph menus. They are called platforms because they do not just produce simple static results. Platform results appear in report windows, are highly interactive, and are linked to the data table and to each other.
The platforms under the Analyze and Graph menus provide a variety of analytical features and data exploration tools.
The general steps to produce a graph or analysis are as follows:
1. Open a data table.
2. Select a platform from the Graph or Analyze menu.
3. Complete the platform launch window to set up your analysis.
4. Click OK to create the report window that contains your graphs and statistical analyses.
5. Customize your report by using report options.
6. Save, export, and share your results with others.
Later chapters discuss these concepts in greater detail.
The following example shows you how to perform a simple analysis and customize it in four steps. This example uses the Companies.jmp file sample data table to show a basic analysis of the variable Profits ($M).
Step 1: Launching a Platform and Viewing Results
1. Select Help > Sample Data Library and open Companies.jmp.
2. Select Analyze > Distribution to open the Distribution launch window.
3. Select Profits ($M) in the Select Columns box and click the Y, Columns button.
The variable Profits ($M) appears in the Y, Columns role. See Figure 2.5 for the completed window.
Another way to assign variables is to click and drag columns from the Select Columns box to any of the roles boxes.
Figure 2.5 Assign Profits ($M)
4. Click OK.
The Distribution report window appears.
Figure 2.6 Distribution Report Window on Windows
The report window contains basic plots or graphs and preliminary analysis reports. The results appear in an outline format, and you can show or hide any report by clicking on the disclosure button.
Red triangle menus contain options and commands to request additional graphs and analyses at any time.
• On Windows, place your mouse pointer over the blue bar at the top of the window to see the menu bar and the toolbars.
• On Windows, click the data table button in the lower right corner to view the data table that was used to create this report. On Macintosh, click the Show Data Table button in the upper right corner of the report window.
• On Windows, click the JMP Home Window button in the lower right corner to view the Home Window. On Macintosh, select Window > JMP Home.
Step 2: Removing the Box Plot
Continue using the Distribution report that you created earlier.
1. Click the red triangle next to Profits ($M) to see a menu of report options.
2. Deselect Outlier Box Plot to turn the option off.
The outlier box plot is removed from the report window.
Figure 2.7 Removing the Outlier Box Plot
Step 3: Requesting Additional Output
Continue to use the same report window.
1. From the red triangle menu next to Profits ($M), select Test Mean.
The Test Mean window appears.
2. Enter 500 in the Specify Hypothesized Mean box.
3. Click OK.
The test for the mean is added to the report window.
Figure 2.8 Test for the Mean
Step 4: Interacting with Platform Results
All platforms produce results that are interactive. For example:
• Reports can be shown or hidden.
• Additional graphs and statistical details can be added or removed to suit your purposes.
• Platform results are connected to the data table and to each other.
For example, to close the Quantiles report, click the disclosure button next to Quantiles.
Figure 2.9 Close the Quantiles Report
Platform results are connected to the data table. The histogram in Figure 2.10 shows that a group of companies makes a much higher profit that the others. To quickly identify that group, click on the histogram bar for them. The corresponding rows in the data table are selected.
Figure 2.10 Connection between Platform Results and Data Table
In this case, the group includes only one company, and that one row is selected.
How is JMP Different from Excel?
There are a number of important differences between JMP and Excel or other spreadsheet applications.
|
Formulas
|
|
|
Excel
|
Formulas are applied to individual cells.
|
|
JMP
|
Formulas are applied only to entire columns. “Calculate Values with Formulas” in the “Work with Your Data” chapter describes how to use formulas.
|
|
Column Names
|
|
|
Excel
|
Column names are part of the grid. Numbered rows and labeled columns extend past the data. Numeric and character data reside in the same column.
|
|
JMP
|
Column names are not part of the grid. There are no rows and columns beyond the existing data. The grid is only as big as the data. A column is either numeric or character. If a column contains both character and numeric data, the entire column’s data type is character, and the numbers are treated as character data.
|
|
|
“Understand Modeling Types” in the “Analyze Your Data” chapter describes how data type influences platform results.
|
|
Tables and Worksheets
|
|
|
Excel
|
A single spreadsheet contains several tables, or worksheets.
|
|
JMP
|
JMP does not have the concept of worksheets. Each data table is a separate .jmp file and appears in a separate window.
|
|
Data Grid
|
|
|
Excel
|
Data can be located anywhere in the data grid.
|
|
JMP
|
Data always begins in row 1 and column 1.
|
|
Analysis and Graph Reports
|
|
|
Excel
|
All data, analyses, and graphs are placed inside the data grid.
|
|
JMP
|
Results appear in a separate window.
|
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.