Countybudget simulator drew 14,800 page views and 294 submissions but design limits skew results

Sedgwick County staff presented early analytics from the county's first online budget simulation, saying it reached about 14,809 page views and produced 294 completed submissions during a May launch that was later extended by a week.

Dr. David, who led the presentation at the June staff meeting, told commissioners the tool recorded a total of roughly 974 hours and 55 minutes of time on the simulation and showed a median on-page time around four minutes. "This is very, preliminary," he said, and described the data as a starting point for internal improvements and for the recommended budget stage.

The nut of the presentation: the simulation achieved substantially more reach than in-person outreach but produced a relatively small number of full submissions. Among submitters, county staff reported that 71% identified as residential property owners, 22% as non-owners and about 6% as both residential and commercial owners. Ownership responses at exit were close to 70% owners and 16% renters. Age and education among submitters skewed toward middle-age and higher education: 30-49-year-olds made up about 44% of submitters, 18-29-year-olds about 17%, and about 34% reported a graduate degree. Dr. David described that last figure as a likely skew in the first round: "definitely, this is kind of skewed towards people with higher education."

Staff emphasized several technical design choices that affected results. The simulation began with a $3,200,000 deficit; participants could only change property tax revenue and spending categories constrained to roughly +/-2.5% for many departmental subcategories. Dr. David said the tax control was structured with a total adjustable range up to $7,000,000 and that each tax increment in the interface represented roughly one-eighth of a mill. Because the tool required submitters to balance the budget before they could post comments, some users who could not reach a balanced configuration did not complete submissions. That behavior likely biased comment and submission samples toward users who were able to accept tax increases or make particular spending trade-offs.

Commissioners and staff also discussed what the simulation did and did not show. A commissioner recommended clearer presentation of the tax assumptions behind the starting deficit. Staff said the tool used the current forecast assumption of a 28.701 mill levy and a 5% assessed-value increase. Commissioners and staff agreed that an explicit slide or FAQ explaining how assessed-value changes and mill-levy choices affect revenue would be useful because many users do not follow the full budget process.

Presenters also described how embedded videos affected engagement. Dr. David and commissioners noted that videos explaining EMS/911 and property taxes appeared to influence how users allocated funds in the simulator. "You saw how much money was given to EMS and 911 when they got to that process," a commissioner said, noting that explanatory content can shape participant choices.

Staff listed other limitations and next steps: demographic fields were optional (so some users left them blank), the tool currently allowed only property-tax changes on the revenue side, some department subcategories were not balanced so a 2.5% cut produced widely different dollar impacts between large and small budget lines, and weekend timing did not produce higher participation as staff had hoped. Dr. David said about 150-200 of the 294 submitters left an email address and that he could share raw comments and the comment archive with commissioners for review.

Why it matters: county staff framed the simulator primarily as an engagement and education tool that helps residents see trade-offs in a large county budget; commissioners said the tool extended reach well beyond prior in-person forums but cautioned against treating the submission sample as statistically representative.

Next steps noted in the meeting included: adding clearer FAQ material and a property-tax impact summary in the interface, rebalancing subcategory granularities so percent changes translate to comparable dollar effects, reconsidering when and how the simulator is publicized, and assessing whether a paid random-sample survey would be worth the additional cost to reduce demographic skew. Dr. David said he plans further analysis as the county moves from recommended to adopted budget stages and welcomed commissioners' questions and suggested design changes.