Matrix Analysis.
A matrix question asks the same set of choices about several statements at once — rate each travel type, score each ad channel, agree/disagree with each claim. That extra dimension is what makes matrix the most powerful question type in Analyse, and the Table Slots are the key to it: a matrix view has a Statements slot, a Choices slot and a Filters slot, and where you put each one decides the story the table tells. (New to Analyse? Start with the Analysis Overview.)
Open a matrix view and the defaults are:
- Statements = Rows, Choices = Columns
- Values = Cell % — each cell is its group’s share of the people asked
- Chart = Stacked Bars
The example here is the demo dashboard’s "Q39. Please indicate your level of experience with or interest in each of the following travel types" — eight travel types, each answered Have done / Have not done / I am not sure (n=907):
One glance tells the story: 78% have done a domestic road trip, but only 31% an international cruise.
Focus on one statement — the ready-made crosstab
The whole-matrix view can’t show you who is behind each number. To get segments in, collapse the Statements slot to the statement you care about: the filters step into the freed rows automatically, your segments become the rows, and the choices stay across the top — a ready-made crosstab. (Equally valid from the other direction: set the Filters slot to Rows, and the statements step aside by collapsing to a single statement you can then change.) Either way, the chip under the chart names the statement in focus, and the + under the filter chips adds the segments you care about:
Now the story has segments in it: 81% of males have done a domestic road trip vs 74% of females; Millennials (80%) out-travel Baby Boomers (76%). Step the Statements slot through statement after statement to run the same crosstab across the whole matrix, and set Statements back to Rows to return to the full statements × choices view.
Collapse a choice for the one-screen overview
The Choices slot works the same way — its menu lists Rows, Columns, then every choice. Collapse it to one choice — say "Have done" — and the table becomes statements × filters, showing that single answer for every statement and every segment at once:
That’s the whole travel market on one screen — every travel type, every segment, one number per cell. Tip: with filters as the chart series, switch the chart type to Grouped Bars — segments overlap, so stacked bars would add percentages that were never meant to be added.
Prefilters & cross filters
Everything from Prefilters & Cross Filters applies to matrix views too: + Prefilter above the chart narrows the whole view to one audience (watch n drop in the top corner), and the cross-filter chips you add can go to one view, selected views or all views via the Add to menu.
Score & average the scale
When the choices are an ordered scale, a single average per statement often beats a percentage breakdown. The demo’s "Q49. How annoying do you find advertisements in the following channels?" rates ten channels on a five-point scale — perfect for scoring:
- Click the ⚙ on the Choices slot.
- Score the scale 1–5, and set anything that shouldn’t count — here "N/A" — to exclude.
- Tick Combine Weighted, name the combined row ("Average"), and click Done.
Then set the Values slot to Average and every statement collapses to one score:
Streaming video and social media ads annoy most (3.6 out of 5); outdoor media least (2.9). And the crosstab trick still works on averages: put Filters in the Rows, pick a statement in the Statements slot, and the segments compare on one line — Free-to-air TV ads average 3.60 for Regional vs 3.47 for Metro respondents.
Statistics
Tick Subset Significance Test in the dashboard Settings (⚙) to highlight cells green or red against the Total, at a 95% confidence level by default — sig testing applies when filters are in your rows or columns, on Row % or Column % values. Margin of Error shows how reliable each row’s base is. Details: Significance Testing and Margin of Error.
Tidy the matrix itself
Merge scale points into Top 2 Box, hide statements you’re not presenting, exclude N/A, or reorder either dimension — all behind the ⚙ on the Choices and Statements slots, as described in Configure Choices, Statements & Loops. (Remember: choices get the full toolkit; statements can be relabelled, hidden and reordered.) The Sort slot ranks statements by result automatically — note that merging or hiding items returns Sort to Manual, so re-apply it afterwards. And when it all looks right, click Save — switching dashboards discards unsaved changes.
FAQs
Where did the Target slot go?
Its job moved into the dimension slots: targeting a statement is now the Statements slot set to that statement; targeting a choice is the Choices slot set to that choice. The chip under the chart always names what’s collapsed. Full tour: Table Slots.
Why did my Statements collapse when I moved Filters into the Rows?
By design — a table only has two axes, so giving the filters an occupied axis makes the current occupant step aside. Set Statements back to Rows whenever you want the whole matrix again.
Cell % or Row % for a matrix?
Cell % (the default for new tables) is each group’s share of the people asked, and it survives transposing and multi-select statements. Row % makes each row read as its own 100% — natural for single-select scales. See Values for the full decision guide.
Can I compare more than one statement at a time by segment?
Collapse the Choices slot to one answer instead — you get every statement × every segment for that answer on one screen.
Where did "Add Global" go?
Retired — the Add to menu now offers Current View, Selected Views or Global, and chips stay ordinary local chips. See Prefilters & Cross Filters.