{
    "table_name": "AI_Usage_and_Impact_on_Students_and_Professionals",
    "table_type_tags": [
        {
            "tag": "dimension",
            "confidence": 0.9
        }
    ],
    "primary_type": "dimension",
    "insufficient_evidence": false,
    "entity": {
        "columns": [
            "User_ID"
        ],
        "unique_per_row": true
    },
    "time_axis": null,
    "reasoning": "User_ID (Uniqueness 0.960, 288 distinct of 300 rows) names the respondent. The few repeats (top value count 2, e.g. U0116 and U0059 appear as fully identical duplicate rows) are duplicate survey rows, not repeated observations. No event timestamp or sequence column exists; the only datetime columns are provenance columns, which are ignored. Each row is a one-off survey profile of a user (demographics, AI tool, scores), so there is no action or sampling axis → dimension."
}