Skip to content

Home / Statistical Tools / Data Selection

Data Selection

The window every analysis opens with. It asks two things: which cells hold your data, and how that data is arranged. Answer them and press Next > to reach the tool's own dialog.

It is titled Data Selection and it carries a Help button, a Cancel button and Next >.

The controls

Control What it does
Selected Range: the cells to analyse. Type an address or select it in the worksheet
Data Orientation Data in Columns, Data in Rows or One Group Data
Dataset Name: the name for the single dataset, and it appears only for One Group Data
First Row/Column is Header the first row, or the first column for data in rows, holds the names rather than data
Exclude Data in Hidden Cells hidden rows and columns are left out of the analysis

A small button beside the range box shrinks the window, so it stops covering the cells you are trying to select. Its tooltip is Shrink/Expand form. Press it again to bring the window back.

You can select while the window is open. It does not block Excel, so click and drag in the worksheet and the range box follows.

The three orientations

Data in Columns. Each column is one dataset and each row is one observation. This is the usual layout and the one most tools expect.

Data in Rows. Each row is one dataset and each column is one observation. The same data, turned on its side.

One Group Data. The whole selection is a single dataset, however many rows and columns it covers, and it gets the one name you type in Dataset Name:. Use it when the values are all measurements of the same thing and their arrangement on the sheet means nothing.

What is already filled in when the window opens

Quantum XL guesses all of it from what you had selected, and every guess is one you can change.

Setting How it is guessed
Data Orientation more columns than rows gives Data in Rows; anything else gives Data in Columns. A selection inside an Excel Table always gives Data in Columns, whatever its shape
First Row/Column is Header ticked when the first cell holds text, cleared when it holds a number. A Table read as columns is always ticked
Exclude Data in Hidden Cells ticked
Dataset Name: empty, with Dataset shown as the placeholder

Select a single cell and the range grows to fit the data around it. Inside an Excel Table it becomes that whole table; outside one it becomes the block of filled cells surrounding the cell, the same region Ctrl+Shift+8 selects in Excel.

Select an entire worksheet, or entire columns, and it trims to the used cells rather than handing the analysis a million empty rows.

Selecting more than one block

Hold Ctrl and select several separate blocks and the analysis reads them all. They have to line up, though, and the window says so if they do not:

Reading data in columns, every block must have the same number of rows and start on the same row:

(Data in Columns) All non-contiguous areas must have the same number of rows.

(Data in Columns) All non-contiguous areas must start at the same row.

(Data in Columns) Non-contiguous areas cannot overlap.

Reading data in rows, the same three rules apply to columns:

(Data in Rows) All non-contiguous areas must have the same number of columns.

(Data in Rows) All non-contiguous areas must start at the same column.

(Data in Rows) Non-contiguous areas cannot overlap.

One Group Data has none of these rules. The blocks can be any shape anywhere, because the whole selection is one dataset and their arrangement carries no meaning.

When you have selected an Excel Table

Quantum XL notices, and asks about it rather than deciding for you. Up to three questions, each in its own message box:

It asks When
Your selection includes both an Excel Table and non-table region, would you like to include both? the selection straddles the edge of a table
The Excel table has headers, would you like to use them as the headers for your data? the table has a header row
Typically, when analyzing data in a table the data orientation is in Columns. Are you sure you want to analyze this in rows? you have set Data in Rows on a table

The third is a question and not a refusal. Answer yes and the analysis runs in rows.

What it will not accept

The message What to do
Please select a range of cells for the analysis. nothing is selected. Select the data
The selection contains merged cells at address. Please unmerge cells before continuing. the address is given, so go to that cell and unmerge it. A merged cell cannot be read as data, and the analysis stops here rather than guessing which of the merged cells the value belongs to
Cannot omit hidden rows when first row contains headers. The header row is hidden. your header row is hidden and you have asked to exclude hidden cells, which would throw the names away. Unhide the row, or clear First Row/Column is Header, or clear Exclude Data in Hidden Cells
Cannot omit hidden columns when first column contains headers. The header column is hidden. the same, for data in rows
(Data in Columns) You must select n or more columns for this analysis. some tools need more than one dataset. A scatter plot needs two, and the T² chart needs two variables
(Data in Rows) You must select n or more rows for this analysis. the same, read the other way

One Group Data is never refused for having too few datasets, because it is always exactly one.

The Dataset Name

It applies to One Group Data only, and two rules apply to it:

  • These characters are not allowed: : \ / ? * [ ]. The message is The characters :\/?*[] cannot be used as the name of a dataset.
  • Thirty characters at most. The message names both lengths: The maximum length of the dataset name is 30 characters, the current length is n characters.

Leaving it blank is allowed. The analysis then names the dataset itself.

What happens to the names

With First Row/Column is Header ticked, the first row (or column) supplies the dataset names and takes no part in the calculation.

With it cleared, Quantum XL generates the names from the position instead, so a chart still has something to label its series with.

What each tool does with the data next

This window says which cells and how they are arranged. It does not say which column does what. That comes next, on the tool's own Data tab, where a column becomes a measurement, a subgroup identifier, a GroupBy column or an X-axis label.

See Also