Data processing with JMP - Short videos on how to operate each process - (Japanese)

Introduction to this video.

Main items

• Turn raw data into analytical data
• What to do on the JMP side when you load data into JMP
• Benefits of data processing with JMP
• Data processing introduced in this video
• Main data used in this seminar
• Demonstration of Exploratory Data in JMP

Uses : After importing a spreadsheet such as Excel into JMP, check that the column scale is as expected and change the scale if necessary.

Continuous scale: Numerical measurements, quantitative variables (continuous)

Nominal scale: Qualitative variables with numbers or letters that have no ordering (categorical)

Ordinal scale: Numeric or character qualitative variables with order meaning (categorical)

As a general rule, numeric columns are converted to a continuous scale and strings are converted to a nominal scale.

Action : Click the icon to the left of the column name in the column panel (center left of the data table) to change the scale.

use : Changes the order in which categories (levels) appear in JMP reports (tables, graphs, etc.).

operate：

1. In the data table, select the column for which you want to change the order of categories, and then select Column > Column Info.
2. From the [Column Properties] on the bottom left, select [Value Display Order]
3. Change the order of values ​​in the Custom Order section

use ：Displays labels for plot points on the graph. If you set a column label, the value of that column will be displayed as the label. If you set a row label, the label will be displayed on the graph for the plot points in that row.

operate：

Column Label : Select the column you want to label and select [Column] > [With Label/Without Label].

Row Label : Select the relevant row and select Row > With/Without Label

Note : To remove the label setting, perform the same operation as above (no label).

use : In a JMP data table, you can group several columns together so that when specifying columns in an analysis or graphing platform, you can specify all the variables in that group at once.

Operate : Select all the columns you want to group and choose Column > Group Columns .

supplement : If you want to ungroup a group, select the group and choose Columns > Ungroup Columns .

use : For JMP data tables, use this to exclude rows not needed in the analysis or to subset data that meets certain conditions.

operate：

Hide and Exclude Rows

With the rows you want to exclude selected, select Rows > Hide and Exclude

Subset

1. Select Table > Subset
2. Select the radio button [Filtered rows] and set the filter conditions.

use : Hides specific columns in a JMP data table. For example, hides columns that are not needed for analysis in a data set with many columns.

Hide: Do not display the column in the data table.

Exclude: Do not include the column in the analysis (it will be removed from the platform column selection options).

operate：

Do not display

Select the column in question and select Column > Hide/Unhide

exclude

Select the relevant column and select [Column] > [Exclude/Don't Exclude]

use : Mainly used for categorical variables (nominal scale, ordinal scale) to change or merge category values. Can also be used to extract specific characters (numeric values) from a string.

operate : Select the relevant column, select [Column] > [Recode] , and specify the new value (the value to be recoded) for the old value (the current value).

use : Creates a formula for a JMP data table.

operate：

1. Select the appropriate formula from the right-click menu

Select the column to be used for the formula, right-click the column name, and select [Create New Formula].

2. How to create a formula in the formula editor

Select the column for which you want to create a formula, and then select Column > Formula.

⇒ The formula editor will start, so create the formula.

use : Set cutpoints for continuous columns and create categorical variables.

Example: For a column containing age, create a categorical column that divides the age into teens, twenties, etc.

operate：

1. Select the continuous column and choose Column > Utilities > Create Binning Formula.
2. Set a cut point and select Create Calculated Column .

use : In JMP data tables, you can set properties (attributes) for columns, such as the display order of values ​​and value labels. You can copy the properties set for one column and paste them into another column to apply the same properties to that column.

operate：

• Select the column, right-click the column name and select Copy Column Properties .
• Select the columns you want to copy, right-click the column name and select Paste Column Properties .

use : Row attributes (exclude, hide, label, etc.) can be copied to a data table column and pasted wherever needed.

For example, when there are multiple exclusion conditions for one data table, those exclusion conditions can be saved in the data table and applied when needed.

operate：

Create a destination column

Change the data type to Row State.

Copy current row status to columns

Click the column icon and select Copy from row attributes .

Apply information in a row state column as row states in a data table

Click the column icon and select Copy to Row Attributes .

use ：Check for missing data. In JMP, you can check missing data in detail mainly in the [Check for Missing Data] or [Show Missing Data Patterns] platforms.

operate：

Check for missing values : Select Analyze > Screening > Check for Missing Values ​​and select the columns you want to check for missing values.

Show Missing Value Patterns : Select Table > Show Missing Value Patterns and select the column you want to check for missing values.

use ：Check for outliers. In JMP, there are two ways to check: visually using box plots and scatter plots, and by using the [Check for outliers] platform.

operate：

Check for outliers : Select Analyze > Screening > Check for outliers and select the column (continuous) you want to check for outliers.

use : Select columns that have duplicate values ​​for the selected column.

operate : Select the column you want to examine and choose [Rows] > [Select Rows] > [Select Duplicate Rows] .