<result> <column_name>Education: Some college</column_name> <family>additive</family> <subtype>intensive_additive</subtype> <confidence>0.85</confidence> <requires_user_confirmation>false</requires_user_confirmation> <sum_meaningful>false</sum_meaningful> <signed>false</signed> <reducers> <primary_reducer>mean</primary_reducer> <valid_reducers>mean, min, max, median, distribution</valid_reducers> <excluded_reducers>sum (summing percentages across entities is meaningless), product (not a ratio of change)</excluded_reducers> <average_semantics>arithmetic_mean</average_semantics> </reducers> <inner_outer_measure> <inner_measure_reducer>min</inner_measure_reducer> <outer_measure_reducer>max</outer_measure_reducer> </inner_outer_measure> <multiplicative_partner_derivable>false</multiplicative_partner_derivable> <reasoning> DIMENSION table with no sequence axis strongly priors against multiplicative. The column contains percentage values (0.55% to 1.64%) representing the proportion of a population with some college education at each entity. These are intensive demographic attributes - snapshot levels that describe a characteristic at a point, not extensive quantities that accumulate. Summing "1.17% + 1.27%" across two cities is meaningless; the natural rollup is a weighted mean. The percentage notation indicates a proportion/share, which is an absolute level measure (intensive_additive), not a ratio of sequential change (which would be ratio_multiplicative). </reasoning> <table_type_conflict>false</table_type_conflict> </result>