Get Full Government Meeting Transcripts, Videos, & Alerts Forever!
Get email alerts on the Survey Methodology topic
No spam. Unsubscribe anytime.
Committee questions survey methods and representativeness; consultant says results are unweighted counts
Summary
Members pressed the consultant on whether analysis used weighting or AI‑driven weighting. Debbie Ward said the quantitative question results are unweighted counts and the report combines that with qualitative coding of open responses.
Several committee members raised concerns about whether the July validation survey accurately represented the city and how the consultant’s AI analysis treated open‑ended responses. Theresa asked if the AI “only count numbers of times that it comes up, or is there some kind of weighted metric,” noting that certain issues (watershed, erosion) may matter more in some neighborhoods.
Ward answered that section 3 is a quantitative analysis of unweighted question responses combined with qualitative coding of open‑ended comments: “We did not use any weighting in the questions. So this the question results are quantitative, unweighted, just counting the number of responses.” Members asked Ward to provide additional ZIP‑level household-share comparisons and to break responses out by length of residence to help interpret differences between town‑center and rural priorities.
