Weber argues that values shouldn't be used when conducting research (e.g. no leading questions) but are needed when choosing topics, data and how findings from the research are used (considering practical, ethical and moral implications of the work).