Standardized　evaluation　has　been　applied　in　the　research　and　development　(R&D)　activities　of　Chinese　universities　for　a　long　time,　which　ignores　the　technological　heterogeneity　between　these　activities.　Standardized　evaluation　has　thus　been　difficult　to　meet　the　evaluation　needs　of　the　Chinese　government,　and　classified　evaluation　is　imperative　for　the　diversified　development　of　R&D　activities　in　Chinese　universities.　Therefore,　this　paper　introduces　directional　distance　function　(DDF)　model　with　endogenous　technique　into　meta-frontier　data　envelopment　analysis　framework　to　construct　a　new　theoretical　tool　of　classified　evaluation,　and　this　new　method　is　applied　to　evaluate　the　efficiency　of　R&D　activities　of　939　Chinese　universities.　The　theoretical　contribution　is　to　unify　the　endogenous　directions　relative　to　different　frontiers,　and　then　a　new　meta-frontier　DDF　framework　is　constructed;　while　the　practical　contribution　is　to　provide　the　following　conclusions　and　implications:　(1)　classified　evaluation　and　standardized　evaluation　have　different　applicability　to　different　categories　of　Chinese　universities;　(2)　the　homogeneity　of　the　improvement　direction　is　serious　in　Chinese　universities;　(3)　different　categories　of　Chinese　universities　have　different　improvement　needs　in　different　inputs　and　outputs,　and　the　distances　between　most　universities　and　their　two　frontiers　are　not　far;　(4)　the　selections　of　both　improvement　direction　and　indicator　are　important　factors　affecting　the　evaluation　results.　Finally,　some　useful　suggestions　are　provided　to　improve　the　efficiency　of　R&D　activities　in　Chinese　universities.　©　2024　Elsevier　Ltd