2026 New Jersey Cannabis Consumer Survey Methodology

Respondents first identify all qualifying New Jersey dispensaries they personally visited within the previous 12 months.

If a respondent identifies one qualifying business, that business is used for the detailed customer-experience evaluation. If a respondent identifies two or more qualifying businesses, the survey randomly assigns one of the businesses selected by that respondent for detailed evaluation. The respondent does not manually choose which selected business receives the detailed rating.

This rule is intended to reduce the risk that respondents selectively rate only the business they liked most or disliked most. Each qualifying respondent contributes no more than one primary dispensary evaluation.

Survey responses may be recruited through channels including Cannabis Insider Hub, organic search, email, social media, direct traffic, referral traffic, Root 22 Dispensary-controlled channels, other participating-business channels, and other documented sources. The survey does not require that every participating business recruit respondents. Actual recruitment sources are tracked and reported with the final methodology/results.

Where technically available, the survey captures referring source, UTM source, UTM medium, UTM campaign, and a direct/unknown classification. If Root 22 Dispensary or another participating business contributes a large share of responses, that fact must be considered and disclosed.

A response is incomplete for primary analysis if the participant does not complete the required eligibility questions or the required assigned-business rating questions. Incomplete responses do not count toward the 400-response study minimum or the 30-response business minimum, though they may be retained internally for quality-control reporting.

Reasonable safeguards (rate limiting, bot-protection signals, an anonymous response ID, timestamp, duplicate-submission indicators, completion status, UTM/referral source, and abnormal-pattern detection) are used to flag duplicate, automated or suspicious submissions. Flagged responses are not automatically deleted solely because they are flagged, and any exclusion uses consistent review criteria. A response is never excluded merely because it contains a negative rating or unfavorable opinion.

A response may be excluded from the analytical dataset only when objective review establishes that it fails eligibility, is materially incomplete, is a confirmed duplicate, is confirmed automated/bot activity, contains data that makes it impossible to interpret reliably, or meets another pre-established fraud/integrity exclusion criterion. The reason for any exclusion is retained internally, and the number of excluded submissions (28 excluded out of 428 total submissions) and principal exclusion reasons are summarized with the final methodology.

The system makes the following fields exportable for internal analysis: submission date/time, anonymous response ID, ZIP/geographic response, dispensaries personally visited, randomly assigned dispensary, rating for each primary category, direct comparative answer where applicable, referral/source information, UTM information, completion status, duplicate/suspicious flag, and analytical inclusion/exclusion status. Unnecessary sensitive or personally identifying information is not collected.