Politi also questioned the CAMA method, which often uses home sales in Lake Placid and other nearby towns to estimate market values in Keene, a town where very few homes sell each year.
Mallison assured the committee analysis through CAMA made reliable adjustments when comparing sales from outside towns.
"The CAMA models that they build actually take into practice that there are differences between Lake Placid and Elizabethtown," he said.
Scozzafava questioned the number of sales needed to make a reliable analysis, arguing that the use of CAMA led to an unnecessarily low equalization rate in Moriah.
According to Mallison, a sales ratio study requires less than 100 sales as samples while CAMA analysis usually requires about 300. CAMA was required in the case of Moriah, he said, because raised assessments were skewed toward recently sold homes.
"Our model, from a statistical basis, doesn't work with a limited number of sales," said Mallison, "and that's difficult in a rural area. Our job is to be statistically accurate."
Mallison said his agency continues to work on ways to make the property tax system more fair statewide and that they would at least reevaluate the methods used to determine equalization in areas with very few sales.
"I will go back to the office and have a serious discussion about what our statistical analysis are," said Mallison. "We want to be able to reflect what the local officials see as fair in the rural areas."