<result> <column_name>Sleep_Duration_Hours</column_name> <family>additive</family> <subtype>real_additive</subtype> <confidence>0.62</confidence> <requires_user_confirmation>false</requires_user_confirmation> <sum_meaningful>true</sum_meaningful> <signed>false</signed> <reducers> <primary_reducer>sum</primary_reducer> <valid_reducers>sum, min, max, mean, median, count, distribution</valid_reducers> <excluded_reducers>product (not a ratio), geomean (not a ratio)</excluded_reducers> <average_semantics>arithmetic_mean</average_semantics> </reducers> <inner_outer_measure> <inner_measure_reducer>sum</inner_measure_reducer> <outer_measure_reducer>sum</outer_measure_reducer> </inner_outer_measure> <multiplicative_partner_derivable>false</multiplicative_partner_derivable> <reasoning> The table is a DIMENSION with no sequence axis, so multiplicative is implausible. The data agrees: the mean and median are about 6.7, all values are positive, the range is 3 to 10.5, nothing is near 1.0, and log symmetry gain is negative. The name token "duration" and the physical unit (hours) point to a real_additive quantity, and the gate leans real_additive for physical quantities on DIMENSION tables. Confidence is moderate because the column is a bounded per-entity typical level, which could be read as intensive_additive, where mean would be the natural rollup. The mean reducer is kept in the valid list so that reading still works. </reasoning> <table_type_conflict>false</table_type_conflict> </result>