```xml <result> <column_name>overall_pct</column_name> <family>uncertain</family> <subtype>none</subtype> <confidence>0.0</confidence> <requires_user_confirmation>true</requires_user_confirmation> <sum_meaningful>false</sum_meaningful> <signed>false</signed> <reducers> <primary_reducer>none</primary_reducer> <valid_reducers>none</valid_reducers> <excluded_reducers>all - column is entirely null</excluded_reducers> <average_semantics>n/a</average_semantics> </reducers> <inner_outer_measure> <inner_measure_reducer>n/a</inner_measure_reducer> <outer_measure_reducer>n/a</outer_measure_reducer> </inner_outer_measure> <multiplicative_partner_derivable>false</multiplicative_partner_derivable> <reasoning> The column is entirely null (null_fraction=1.0) with dtype VARCHAR, providing zero data evidence for classification. The name "overall_pct" suggests a percentage/ratio (multiplicative family), but on a DIMENSION table with no sequence axis, multiplicative is implausible per the table-type gate. Without any non-null values, no statistical signals exist to override the naming hint or confirm any family. Classification is impossible; human review required to determine if this is a data quality issue or an unused column. </reasoning> <table_type_conflict>false</table_type_conflict> </result> ```