and type the data into the first cell directly below the heading. Press Enter again, type the data, and repeat as needed. Rows within JMP are consecutively numbered as observations or cases. (See Figure 2.18.)
Figure 2.18 Add a New Column
4. To create another column, double-click on the next column’s heading and enter the data as you did before.
If it is more practical for you to enter a series of data for each row as you build your data table, set up all of your column headings first and then use the Tab key to move from the left columns to the right. When each column has been filled, the Tab key moves down to the beginning of the next row.
| Note |
| JMP will recognize the type of data you are entering and assign a data type to the column, either numeric or character. It also assigns an icon next to the columns (or variables) in the box on the left. These icons are the modeling type and are discussed in Section 2.3. |
2.2 The JMP Data Table
The JMP Data Table looks very much like any spreadsheet. (See Figure 2.19.) In JMP, column headings indicate variables (what you have measured or counted), and rows indicate individual cases or observations. JMP requires your data to be structured in this way. If it is not, JMP can help you reformat your data. (See Section 2.4.)
Figure 2.19 The JMP Data Table
Data Table refers to the spreadsheet-like grid where your data resides.
The data grid can contain any number of columns (your variables) or rows (observations or cases). In this sense, we refer to data within the JMP data table as structured data.
In addition to the data grid, notice the three panels to the left of the data table. These panels contain information about your data (metadata). They provide vital information about your data as well as options to streamline and save your analyses.
The first and upper-most panel contains the name of the data table. (See Figure 2.20.) This panel stores references, notes, and/or scripts. Scripts enable you to save, automate, and customize analyses. If you perform a regular analysis or scheduled task, you will want to learn more about JMP scripts. (See the JMP Scripting Guide at Help JMP Documentation Library JMP Scripting Guide.)
Figure 2.20 The Table Panel
The Columns panel (see Figure 2.21) is where your column names (or variables) appear. Each column has an icon in front of it.
Figure 2.21 The Columns Panel
These icons correspond to the modeling type of the data in each column. As discussed in the next section, this is vitally important. JMP produces only the graphs or statistics that are appropriate for a column’s modeling type. In most cases, you can change the modeling type by simply clicking on the icon and selecting another appropriate type. The first number in parentheses represents the total number of column names or variables in the data table. The second number represents any of those columns that are currently selected or highlighted.
The bottom panel is the Rows panel. (See Figure 2.22.) The Rows panel indicates how many rows (observations) are in your data table. This panel also indicates the number of selected, hidden, or excluded rows, if any.
Figure 2.22 The Rows Panel
When rows are hidden, the observations are not included in graphs. When rows are excluded, they are not included in analyses. This row state is effective when you want to see or analyze a subset of your data. You can also both hide and exclude specific rows, which effectively removes the row(s) from your analyses and graphs, but not from your data table. Section 2.5 provides more information about row states including hiding and excluding rows.
| Note |
| Multiple data tables can be open at any time, but only one active data table can be analyzed at a time. If you have multiple data tables open within JMP and you want to switch to another open data table, go to the Home Window (see Figure 2.23) and select the desired data table under Windows List. |
Figure 2.23 The Home Window
A special type of data table is shape files, which are used to create thematic maps. These data tables consist of two tables including a “Name” and corresponding “Boundary” table. These are stored in the Maps folder: C: Program Files SAS JMP 15 Maps. Section 2.7 covers some of the basics about shape files.
| Note |
| There is no practical limit on the size of the data table that you can analyze. However, because JMP runs in your computer’s local memory, the amount of RAM that you have determines the upper size limit of your data table. Your computer should be equipped with at least twice as much memory as the size of the data table. Thus, if you have just 8 GB of RAM, you can analyze a 4GB data table or about a 10-variable data set with 4 million rows! More details about JMP system requirements can be found at www.jmp.com. |
2.3 Data and Modeling Types
One of JMP’s great features is the ability to produce graphs and statistics that make sense for the data that you are analyzing. This feature assumes that your data is correctly classified in the data table. So, what do we mean by data type and modeling type? Let’s define a few terms.
Data
refers to any values placed within a cell of a JMP data grid. Examples include numeric and/or text descriptions: 3.6, $2500, Female, Somewhat Likely, or 7/7/19.
Data type
refers to the nature of the data. The data type is usually either numeric (numbers) or character (often words and letters but sometimes also numbers). Other special purpose data types include expression (used for images and matrices) and row state.
Modeling type
refers to how the data within a column should be used in an analysis or a graph. JMP uses three distinct and primary modeling types: continuous, nominal, and ordinal. (JMP also includes three additional special purpose modeling types: Multiple response, Unstructured text, and None. Since these are less common, we will only summarize them in the note at the end of this section.)
● Continuous data (also referred to as quantitative, ratio, or interval scale data) takes a numeric form and is often thought of as some type of measurement. For example, home selling prices, income earned, costs per square foot, and dates are all examples of continuous data. As a rule of thumb, continuous data can be used in calculations. For example, calculating the average cost per square foot would be meaningful.
● Nominal data is categorical data (also referred to as qualitative, discrete, count, or attribute data) and can take on either a character or numeric form. Nominal data fits into categories or groups such as car type, gender, department, and sales territory and also includes indicator variables such as yes/no or 0/1. In nominal data, it is helpful to count the frequency of the occurrence of values, but otherwise, nominal data is not used in calculations. For example, calculating the average car type would not be meaningful.
● Ordinal data is categorical data that has an inherent order or hierarchy. For instance, Likert scales (such as levels of satisfaction) in a survey and grade levels in school (freshman, sophomore, junior, senior) are examples of ordinal data. That is, they represent categories that have some sequence or order that should be retained in any analysis. Ordinal data is less common than continuous and nominal data, but there