The interview is a central tool in our evaluation and selection of candidates for employment and (particularly at Oxford) admission. It should concern us greatly, therefore, that psychological research has consistently and repeatedly shown that interviewing is an unreliable indicator of candidates’ future performance, and is also particularly vivid and influential compared to other, more reliable, sources of evidence. Involving interviews in the selection process is, unless carefully managed, nothing more than an expensive source of vivid noise. Since we don’t intentionally wish to randomise our selection process, it is important that we are aware of the potential problems with interviews, so that we can take appropriate measures to avoid them.