To group rows in a JMP data table, select the rows you want to keep together, then choose Rows > Group. Grouping organizes selected rows as a row group; it does not combine their data into one observation or calculate a summary. If you want to group every record that shares a category, first sort the table so each category occupies a contiguous block, then group each block separately.
What You’ll Need
- A JMP data table open with the rows you want to organize.
- A clear idea of which rows belong together. If you are grouping by a column value, check that the values are consistent and sort the table by that column first.
- Permission to make changes to the table. If the table is important, save a copy before reorganizing it.
Step-by-Step
- Open and inspect the data table. Open the table in JMP and check the row numbers, column names, and values. Confirm that the rows you intend to group are present and that you are working in the right table. Grouping is a table-organization action, so it does not require a formula, a modeling platform, or a special column type.
- Decide what “group” means for this task. The Rows > Group command groups rows you select; it does not automatically collect matching values from a column. For example, to keep all observations for a particular store together, first arrange the table so that the store’s rows are next to each other. If your actual goal is to calculate totals or averages by store, use a summarizing workflow such as Tables > Summary instead. That produces a summarized table rather than merely grouping original rows.
- Put each intended group in a contiguous block. When groups are based on a value, sort the table by the relevant column before selecting rows. Click that column’s header, then use the appropriate sort command from the Tables menu or the column’s context menu; the exact menu wording can vary by JMP version. Check the resulting order before continuing. For instance, if a category appears in several separated parts of the table, grouping only one section will not group all records with that category.
- Select the rows to group. Click the row number at the left edge of the first row in the block. To select a continuous range, hold Shift and click the last row number in that range. To add or remove individual rows from a selection, use Ctrl-click on Windows; on Mac, use Command-click. If the selection is difficult to see, pause and verify which row numbers are highlighted before you proceed.
- Create the row group. With the intended rows selected, choose Rows > Group from JMP’s menu bar. On a Mac, the menu bar is at the top of the screen; the command is still under Rows. JMP applies the grouping to the selected rows. Look at the row-header area for the group’s visual indicator or outline, and check that the selected rows are represented as one group rather than as separate selections.
- Repeat for other blocks. If you need a group for each category, select and group each contiguous block in turn. Confirm the boundaries before each command. Do not assume that selecting one row from each category will make JMP group all rows with matching values; arrange the rows and select the full intended block instead.
- Check the result and save. Inspect the table to confirm that the groups contain the right rows and that no row was accidentally left out or included. If the table provides a group outline control, use it to expand or collapse a group and verify its contents. Save the table when you are satisfied. If the result is wrong, use Rows > Ungroup on the affected group and create it again with the corrected selection.
For a concrete example, suppose a table contains records for North, South, and West regions, and the rows are mixed together. Sort by the region column. Select the consecutive North rows by their row numbers and choose Rows > Group. Repeat for South and West if you want each region to have its own group. The original observations remain in the table; the operation organizes them rather than replacing them with a single regional record.
Common Mistakes
- Expecting automatic grouping by a column. Rows > Group acts on the current selection. It does not scan a column and group every matching value for you. Sort first and select each block, or use a data-summary method if you need calculated results by category.
- Selecting row cells instead of row numbers. Clicking in a data cell selects or edits a cell; it may not select the whole row. Start from the numbered row header at the left, and confirm the full row selection is visible.
- Grouping a partial category. A category can have multiple records, and it is easy to miss a row at the beginning or end of a block. Check the first and last row numbers and scan the category values before grouping.
- Assuming a group is a summary or a filter. A row group does not calculate a mean, count, or total, and it does not remove other records from the table. Use a Summary table or a data filter when those are the outcomes you need.
- Sorting without checking the selected data. Sorting changes row order, so verify the order and selection before grouping. If the table has related columns, make sure you are sorting the whole table as intended rather than inadvertently changing only a column’s values.
Tips
- Keep a consistent order. Sort on the grouping column before making groups, and use the same sort order for every group. This makes group boundaries easier to check and the table easier to scan later.
- Use a small test first. If you have many rows or are unfamiliar with row grouping, try the command on a small, easy-to-recognize block. Confirm the result, then continue with the remaining blocks.
- Distinguish row groups from row states. Selecting, excluding, hiding, coloring, and labeling rows are different actions. Grouping is for organizing rows together; it is not a substitute for excluding records from an analysis or marking them with a color.
- Keep the source data safe. Save a separate copy before making extensive structural changes, particularly when the table is shared or supports a report. A clear table name can also help distinguish the organized version from the original.
- Use ungrouping for a correction. If a boundary is wrong, ungroup the affected rows, correct the selection, and group them again. This is more reliable than trying to repair an incorrectly defined block by selecting additional rows without checking how the existing group is structured.
Troubleshooting
Rows > Group is unavailable or has no visible effect. Check that the data table is active and that one or more row numbers are selected. Clicking a cell alone may not select a row. If the table is read-only or you do not have permission to change it, make an editable copy and try again. Also inspect the row headers after the command; the group marker can be subtle in a large table.
The wrong rows ended up together. Choose Rows > Ungroup for the incorrect group, then verify the sort order and select the correct row-number range. If you grouped by a category, compare the first and last values in the block and check for blank or inconsistent category labels that may have split the records into unexpected runs.
Rows with the same value are still in separate places. Grouping does not search the full table for matching values. Sort by the intended category first, then select and group each contiguous run. If you need a single output row per category, create a summary instead of a row group.
The table looks different after sorting. Sorting changes the row order, not the underlying values, but the new order can make a table unfamiliar. Check that you used a table-wide sort appropriate to your task and that linked columns still match their records. If the order is not what you want, sort again using the desired column and direction before creating groups.
You cannot tell whether grouping worked. Look closely at the left row-header margin for the group outline or related control, and try expanding or collapsing it if available. If the appearance remains unclear, make a small test group in a copy of the table and compare the row-header display. JMP’s visual details can differ somewhat between releases, but the grouping command remains under Rows in the data table.
Frequently Asked Questions
Does grouping rows combine them into one row?
No. It organizes selected observations as a group; it does not merge their values or discard the individual rows. Use Tables > Summary if you need one calculated record per category.
Can JMP group rows automatically based on a column?
Rows > Group works on the selected rows rather than automatically finding matching values. Sort by the column, select each contiguous block, and group it. For calculated group-level output, make a summary table instead.
Can I group rows that are not next to each other?
For a clear, predictable row group, arrange the intended rows together and select them as a contiguous block. If matching records are scattered, sort first. This also makes it easier to verify exactly which observations belong in the group.
Will row groups change my analysis results?
Grouping itself is an organizational action, not a calculation. It does not change the values in the rows. However, selection, exclusion, or hiding can affect what is displayed or used by some JMP analyses, so do not confuse those row states with grouping.
How do I remove a row group?
Select or activate the group as appropriate and choose Rows > Ungroup. Check the row headers afterward to confirm that the group has been removed. If other groups are present, take care to ungroup only the one you intend to change.
Is grouping the same in JMP on Windows and Mac?
The basic command is Rows > Group in both. On Mac, menus appear in the system menu bar at the top of the screen, and Command generally replaces Ctrl for modifier-key selection. Menu placement and visual indicators can vary slightly by JMP release.
Related
- Conditional Formatting When a Cell Is Blank in Excel
- Balloon Charges in RS3: Logs, Routes and Costs Explained
- Balancing Chemical Equations in 8th Grade: How It Works
Try it: the Date Difference Calculator: days, months, years and weekdays between two dates, with the DATEDIF formulas.
More in How-to basics
- How to Show Negative Numbers in Brackets in ExcelTo show negative numbers in brackets in Excel without changing their values, select the cells, press Ctrl+1, choose…
- Add a Shared Mailbox to New Outlook: Setup and FixesTo add a shared mailbox in New Outlook for Windows, right-click your account name in the folder pane,…
- How to Turn On Circular Reference Iteration in ExcelIn Excel for Microsoft 365 on Windows, turn on circular-reference calculation by going to File > Options >…
- Adding a Dropdown in Google SheetsTo add a dropdown in Google Sheets in a web browser, select the cell or cells, choose Insert…
- Merge 2 Cells in Excel: A Step-by-Step GuideTo merge two cells in Excel, select the adjacent cells, then choose Home > Merge & Center ▼…
- Print Excel Sheets With Gridlines: Settings and FixesYes. In Excel for Microsoft 365 on Windows, select the worksheet, then choose Page Layout > Sheet Options…
Sources: Function behaviour and menu paths: Microsoft Support documentation for Excel. Method & sources
Spotted an error in this guide? Let us know.