Step 2: Create a JSON file for each assessment tool¶

Each <measurement_tool>.tsv file must be accompanied by a <measurement_tool>.json file that describes its contents in a human- and machine-readable way. The .json file contains:

General questionnaire metadata (MeasurementToolMetadata)
Column-level descriptions, one entry per column in the .tsv

General questionnaire metadata¶

Use the MeasurementToolMetadata key to describe the questionnaire as a whole. Provide a Description (the full name and/or common abbreviation) and, where available, a TermURL.

"MeasurementToolMetadata": {
    "Description": "Beck Depression Inventory",
    "TermURL": "snomed:273306008"
}

What is a TermURL?

TermURLs are persistent identifiers for terms in a standardised vocabulary. They are important when questionnaires have multiple names (e.g. "Progressive Matrices Test", "APM", "Raven's Progressive Matrices" all refer to the same instrument).

The ANC supports two vocabularies:

SNOMED — e.g. snomed:273306008 for the Beck Depression Inventory
Cognitive Atlas — e.g. cogatlas:trm_526af65b16c82 for the Stanford-Binet Intelligence Scales

If your questionnaire is not found in either vocabulary, you can omit the TermURL.

Participant ID column¶

Every <measurement_tool>.tsv includes a participant_id column. Annotate it as follows:

"participant_id": {
    "Description": "A participant ID",
    "Annotations": {
        "IsAbout": {
            "Label": "Subject Unique Identifier",
            "TermURL": "nb:ParticipantID"
        },
        "Identifies": "participant"
    }
}

Item-level metadata¶

Add one entry per questionnaire item (column) in the .tsv. The structure depends on the response type:

Discrete response values (e.g. Likert scale)¶

Provide LongName (the full item text), Description, and Levels (a mapping from numeric value to label).

"tas[tas01]": {
    "LongName": "Wenn ich weine, weiß ich immer warum",
    "Description": "Description for tas[tas01]",
    "Levels": {
        "1": "trifft gar nicht zu",
        "2": "trifft eher nicht zu",
        "3": "trifft teilweise zu/teilweise nicht zu",
        "4": "trifft eher zu",
        "5": "trifft voll und ganz zu",
        "6": "Keine Antwort"
    }
}

Continuous or free-text responses¶

For items where responses are not discrete (e.g. a numeric rating on a continuous scale, or an open text field), only LongName and Description are required:

"age_at_testing": {
    "LongName": "Age at time of testing",
    "Description": "Participant age in years at the time of questionnaire administration"
}

Contact¶

If you have questions about annotating a specific questionnaire, contact your assigned data steward.