Glossary
A reference guide to the terms you’ll meet in Data Wrangler. Terms in bold are labels you’ll see on screen.
Advanced Filter
Section titled “Advanced Filter”A temporary filter on Data View, built from conditions joined by AND/OR, or written as SQL. It isn’t saved, so it’s cleared when you switch tabs or leave the page. See Data View.
Ask EVE
Section titled “Ask EVE”An AI assistant that answers questions about one data store. You need Agent access to use it. See Ask EVE.
Case / record
Section titled “Case / record”One row in a data store, such as one survey response. The screens say “case” in delete, restore and conversion messages. See Edit & delete records.
Codeframe
Section titled “Codeframe”A list of numeric codes and their labels, for example 1 = “Yes”, 2 = “No”. See Keys, codeframes & multi-response.
Compute
Section titled “Compute”Calculates a new variable from ones you already have, such as a total, an average across several questions, cleaned-up text or an age from a date of birth. It works inside one data store and runs once. The new variable is marked COMPUTED. See Compute variables.
Connector
Section titled “Connector”A line between two nodes in a wrangle. Connectors set the order the steps run in. See also Node. See Build a wrangle.
Data store
Section titled “Data store”A table of data held in the platform. Rows are records and columns are variables. Most other HumanListening apps read their data from data stores. See Getting started.
Data Store Settings
Section titled “Data Store Settings”The dialog behind Settings on a data store’s header, for owners and editors. Choose which deletion notifications are sent and who’s subscribed, and check survey and data-scrubbing rules. See Data store settings.
Data View / Variable View
Section titled “Data View / Variable View”The two tabs of a data store page. Data View shows the records; Variable View shows the variables that describe them. See Getting started and Data View. See Manage variables.
Draft / Published
Section titled “Draft / Published”A Draft data store can be freely restructured but can’t be used by other apps. Publishing makes it available and locks its structure. Edit Schema unlocks it again (owner only). See Getting started. See also Publish & manage a data store.
The last row of a Recode. It decides what happens to records that match none of your rules: by default, they’re left blank. See Recode variables.
Export definition
Section titled “Export definition”A saved set of export settings you can reuse. See Saved & scheduled exports.
Live / Test
Section titled “Live / Test”Whether a survey response is real (Live) or a test entry (Test). See Survey data stores.
Manage Actions
Section titled “Manage Actions”The list of every Recode and Compute in a data store, with its status and run history. Open it from Variable View to run one again or reopen it. See Recode variables.
Hides a field’s values in the viewer. The values are still stored, and they’re still included in exports. Editors can reveal masked values, and every reveal is recorded. See also Wipe and Mask + retention. See Personal information.
Mask + retention
Section titled “Mask + retention”Masks a field now, then erases each record’s value once it’s older than a set period. See Personal information.
Multi-response variable
Section titled “Multi-response variable”A group of Yes/No child variables that together represent one “choose all that apply” question. See Keys, codeframes & multi-response.
One step in a wrangle. A Reader node pulls data in, a Writer node saves it out, and microservice nodes (such as Recode, Join and Weighting) transform it in between. See also Connector. See Build a wrangle.
Personal information
Section titled “Personal information”A field holding details that could identify a person. You choose how each field is handled: masked, wiped, kept for a set time, or marked as not personal information. See Personal information.
Primary key
Section titled “Primary key”The variable (or combination of variables) that uniquely identifies each record. It must be unique and never blank. Imports and updates use it to match records. See Keys, codeframes & multi-response.
Recode
Section titled “Recode”Creates a new version of a variable by turning its values into codes, scores or text, for example 0–10 scores into Detractor, Passive and Promoter. There are two kinds. Recode in Variable View works inside one data store and runs once (see Recode variables). A wrangle’s Recode step runs every time the wrangle runs (see Recode data).
Recycle Bin (records)
Section titled “Recycle Bin (records)”Where deleted records are held for 30 days before they’re removed for good. This is separate from the Recycle Bin on the home list, which holds whole data stores and other content. See Edit & delete records.
Reference Table
Section titled “Reference Table”A lookup table for mapping one set of values to another. The wrangle Recode step shows a Reference Table option, but it isn’t available at present. See Recode data.
Request
Section titled “Request”A background job on a data store, such as a delete, restore, Live/Test conversion or wipe. You can track requests under Requests on Data View. See Edit & delete records.
RIM weighting
Section titled “RIM weighting”Adjusts each record’s weight so that several variables match their target proportions at the same time. Weighting on one variable gives simple weighting. See Weight data.
A packaged capability for Ask EVE, with example prompts to get you started. See Ask EVE.
A tag on a variable created by a Recode or a Compute. It means a source variable has changed since it last ran, so the values may be out of date. Reopen it and run it again. See Recode variables.
Staged change
Section titled “Staged change”A change in Variable View that isn’t applied until you click SAVE. See Manage variables.
Suggested
Section titled “Suggested”A field the import flagged as possibly holding personal information. It’s a suggestion only and doesn’t stop the import. See Personal information.
Syntax (Recode)
Section titled “Syntax (Recode)”A tab in the Recode window that shows the whole Recode as lines of text, starting RECODE VERSION 1 and ending END. Edits there update Rule Builder, and the other way around. See Recode syntax.
Survey data store
Section titled “Survey data store”A data store created and filled automatically from a Survey Shaper survey. Its survey-owned variables can’t be edited in Data Wrangler. See Survey data stores.
Tokens
Section titled “Tokens”The unit used to measure Ask EVE usage against daily and monthly limits. See Ask EVE.
Variable
Section titled “Variable”One column in a data store. It has a name, which is fixed, lowercase and used by the system, and a label, which is the friendly text you see on screen. See Manage variables.
Variable type
Section titled “Variable type”What kind of values a variable holds: Text, Long Text, Integer, Number, Boolean, Date, Date Time, Time, Guid or Multi-Response. (The Add Variable dialog spells Date Time as DateTime.) See Manage variables.
Permanently erases a field’s values, and keeps erasing any new values daily. It can’t be undone. See Personal information.
Wrangle
Section titled “Wrangle”“The definition of one or more steps used to manipulate data and create a new Data Store.” A wrangle must be published before it produces any output. See Getting started. See Build a wrangle.