Lievesley, who heads the college’s school of social science and public policy, told delegates at the International Journal of Market Research’s Research Methods Forum today: “Particularly when I’m talking to statistical audiences, quality is interpreted in a very narrow way, with regard to validity and reliability… I want to argue that quality is about a whole range of dimensions”.

One of the key aspects of quality, she said, is utility – being able to put the data to some sort of use. She spoke of going “undercover” in the NHS to look at how information is collected and used. “I was flabbergasted at the extent to which people at all levels in the NHS had no idea why the information was being collected,” she said. “It was very worrying that information was seen as being owned only at the top of the organisation.”

Lievesley also warned that standardisation, while valuable in ensuring compatibility over time, can become a barrier to innovation and harm other aspects of quality. The conservatism of public sector researchers was later defended by Richard Bartholomew of the Department for Education, who set out reasons why they tend to be less willing than their commercial counterparts to move away from traditional research approaches such as random probability sampling and “the obsession with the final report”.

Lievesley said excessive performance monitoring in the public sector has become a threat to data integrity. “The danger with a measurement culture is that excessive attention is given to what can be easily measured, at the expense of what is difficult or impossible to measure,” she said.

In answer to a question about resistance to data collection in the public sector because of the sheer amount of it going on, Lievesley said: “My experience is that if you engage with the suppliers of data and they get to influence what you’re collecting and use it, and you don’t have a top-down approach, then actually people don’t see it as a burden.”

The key, she believes, is to involve people in discussions about what information is collected and how it is used. “It is not enough that data are trustworthy,” she said; “they must also be trusted”.