Are bibliometric measures consistent with scientists’ perceptions? The case of interdisciplinarity in research