Output

The Output element is declared in .r.yaml and defines what gets written to the spreadsheet. Its companion Output option is declared in .a.yaml and controls the Save checkbox in the UI. Together they form the save-to-spreadsheet feature: the option in .a.yaml captures the user’s intent, while this element specifies what the column looks like and how it is populated.

Note: The Output element writes to the spreadsheet, not the results panel. Values only appear when the user has ticked the corresponding Save checkbox.

YAML Properties

Property	Type	Default	Description
`name`	string	—	Internal name. Must match the `Output` option name in `.a.yaml`. Accessed in R as `self$results$name`.
`type`	string	—	Must be `Output`.
`title`	string	—	Display label shown in the Save section of the analysis UI.
`varTitle`	string	—	Column header written to the spreadsheet.
`varDescription`	string	—	Tooltip or description for the column in the spreadsheet.
`measureType`	string	`continuous`	Variable type for the new column. One of `continuous`, `nominal`, `ordinal`, or `id`.
`clearWith`	list	`*`	List of option names whose change invalidates this output. Defaults to `*` (cleared by any option change).
`items`	integer or expression	`1`	Number of output columns to produce. Defaults to `1`. For multi-column output, set to the number of columns or an R expression evaluated against the options (e.g. `(length(factors))`).

R Methods

All methods are accessed via self$results$name, where name matches the name property in .r.yaml.

Method	Description
`setRowNums(rowNums)`	Sets which spreadsheet rows to write to. Must be `rownames(data)`, not `1:nrow(data)` — see important note below.
`setValues(values, index, key)`	Sets the vector of values for the column. For a single output (`items: 1`) pass a plain vector. For multi-output use `index` (integer position) or `key` (name) to target a specific column.
`isFilled(key)`	Returns `TRUE` if this output has already been populated in the current run.
`isNotFilled(key)`	Returns `TRUE` if this output has not yet been populated. Use as a guard to avoid redundant computation.
`set(keys, titles, descriptions, measureTypes)`	Configures all columns at once. Each argument is a vector with one entry per column. Convenient for dynamic multi-column output where the number of columns depends on the data or options.
`setTitle(title, index, key)`	Overrides `varTitle` at runtime for a single column.
`setDescription(desc, index, key)`	Overrides `varDescription` at runtime for a single column.

Note on key and index: These parameters are only relevant when items > 1 (multi-column output). For the default single-output case (items: 1) you can omit them entirely — e.g. self$results$myOutput$setValues(myVector) is sufficient.

Important

setRowNums must receive rownames(), not 1:n

When jamovi applies row filters or handles missing values, it drops rows from the cleaned data frame but preserves the original row names. If you pass 1:nrow(data) instead of rownames(data), the values are written to the wrong spreadsheet rows — silently, with no error. Always use:

self$results$residsOV$setRowNums(rownames(data))

Examples

1. Define in `.r.yaml`

- name: residsOV
  type: Output
  title: Residuals
  varTitle: Residuals
  varDescription: Ordinary residuals from the fitted model
  measureType: continuous
  clearWith:
    - dep
    - factors

2. Populate in `.b.R`

if (self$options$residsOV && self$results$residsOV$isNotFilled()) {
    self$results$residsOV$setRowNums(rownames(data))
    self$results$residsOV$setValues(residuals(model))
}

Here data is the cleaned data frame (after na.omit()) and model is the fitted model — both computed earlier in .run(). See the Output Variables tutorial for a full worked example.

3. Configure multiple columns with `set()`

When the number of output columns depends on the data or options (items set to an expression), use set() to define all columns in one call rather than overriding each title and description individually:

keys         <- seq_len(nModels)
titles       <- vapply(keys, function(k) paste('Residuals', k), '')
descriptions <- vapply(keys, function(k) paste('Residuals of model', k), '')
measureTypes <- rep('continuous', nModels)

self$results$residsOV$set(keys, titles, descriptions, measureTypes)

Each argument is a vector with one entry per column. After set(), populate each column by passing its key or index to setValues().