That question sparked Janet Lewis's lifelong commitment to make social research reach out to the real world and make it a better place in which to live. David Walker reports.

While some folk sunned themselves and others got on with expanding the frontiers of knowledge, Janet Lewis spent last summer reading research assessment exercise submissions. The experience did not leave her feeling nobler or happier.

In fact, her work as a member of the ÖйúAƬ Funding Council for England's social policy and social work panel led her to question the whole basis of the RAE. "Too many factors were required to be assessed on some quite fine parameters to make collating them into one single grade a fair process, however conscientious the assessors," she says. She reserves particular criticism for the way the terms "unit of assessment" and "research active" seem to mean different things to different people; for the failure to credit team effort adequately; for poor or unclear definitions of "quality" and policy relevance; and for the failure to distinguish between different units of assessment and the different contexts in which research is done. "It is difficult to think that it is necessary to spend ?36 million to produce a funding formula. There must be a better way of spending scarce resources," she says.

Lewis's work also confirmed a belief built up during a distinguished career in social research, which she will bow out of at the end of the month when she retires as director of research for the Joseph Rowntree Foundation. She says: "Universities and the funding councils have not yet found a way of valuing the useful social knowledge that exists in social science departments/research units doing applied work." This is a source of much regret to someone who not only wants to see the world change for the better but believes that better social knowledge is a precondition for such progress.

ÖйúAƬ

The foundation is happy with much of the work it supports in British universities and with the quality of many researchers. But Lewis believes applied work does not get the backing it deserves. "There is a huge deficit in investment in ÖйúAƬ generally, and it could be argued that applied work suffers no worse than other kinds of activity. But it feels to me that applied social research is the underdog - and this is a huge waste of potential for improving people's lives."

In theory, the RAE was to take into account "outcomes". But in practice, it paid only fleeting attention to the use made of the research that her panel assessed, Lewis says. This speaks to a lifelong theme of hers - what is the point of social inquiry that is not made accessible to those able to act on the results, including those ordinary people being studied, who might have something to say about which issues must be addressed?

ÖйúAƬ

At the foundation, Lewis has tried to practise what she preaches. With ?6 million to spend each year, the foundation is not only a mainstay of research with policy relevance, it is a disseminator of relevant knowledge. If institutions could be beatified, the York-based charity would surely qualify. The legatee of the great Quaker chocolate maker is a major funder of the practical social study of housing, income and poverty, employment, young people, social care and local government. Its motto remains that of Joseph Rowntree, who left the bulk of his wealth to the three trusts bearing his name in 1904 - instead of spending money directly alleviating deprivation, find out its cause so its recurrence can be prevented. When Lewis became director of research in 1986, the foundation's annual budget was less than ?2 million. Until the recent fall in equities, it had risen to nearly ?8 million, much of which is spent on studies commissioned from universities. Yet much social research is done outside academe, as Lewis's career demonstrates.

After undergraduate study at the University of Edinburgh - because sociology was too new for the ancient Scottish university, her main subjects were economics and psychology, with some social anthropology - Lewis did a postgraduate diploma in social administration at the London School of Economics. She went on to do social work with the West London Family Service Unit and research for the social medicine unit at Guy's Hospital. In the late 1960s at Goldsmiths College, she completed a PhD examining variations in local authority health, welfare and children's services. In the 1970s, she joined Rudolf Klein at the Centre for Studies in Social Policy, which later became part of the Policy Studies Institute. Those were glory days for applied research and analysis. However, Lewis notes wryly, it is striking how little the agenda on which she was then working has advanced.

She then became a user and research manager with the Manpower Services Commission and the National Children's Bureau. Her experience of government left her keenly aware of the limits of the civil service's understanding of what constituted sound research. "Someone needed to point out that a response rate of 24 per cent for a survey was not good enough," Lewis says.

She believes that there are no intrinsic barriers between universities and the outside world. The foundation's "Findings" series, which has been widely imitated, has "shown that if you create the right conditions, researchers in universities can write in accessible ways and can present their material in ways that are 'useful'". Universities and funding councils could do more to promote communication and interaction, she says, including creating incentives and career structures to tempt good researchers out into the world.

ÖйúAƬ

In recent years, the Economic and Social Research Council has made much of its focus on "users". "In truth," Lewis says, "it is often enough for the ESRC if researchers tack on the names of one or two policy people or practitioners to confirm that they are interested in principle." She wants to nurture and develop the relationship between researchers and users, involving, say, the recipients of the public service being studied. Even if users helped frame a question, that would not diminish the role of the researcher, whose job remains converting the issue or subject into a researchable topic. But this requires time, effort, money and, crucially, more support for contract researchers.

Much social research is undertaken by people on short-term contracts, outside the university mainstream, Lewis says. "The institutional power structures of the academic world have never seen full-time research activity as a legitimate career, so they have always kept contract researchers in a subordinate position." The problem is self-perpetuating because the only people with a voice in universities and research councils are those who are tenured academics.

Compounding the problem is the lack of money for social sciences research in general. "There is so much more money in other fields. The Medical Research Council, for example, has a whole host of funding arrangements, including a variety of postgraduate fellowships and career grants. There is none of this in social science."

Lewis has long worked to obtain better support for contract researchers, to raise research standards and to promote social science research generally. In 1976, she chaired the group that led to the creation of the Social Research Association, and she has been heavily involved with the Association of Learned Societies in the Social Sciences - now the Academy.

ÖйúAƬ

Lewis says she used to believe that the way forward for applied social research was specialist research units operating outside universities. That changed when the Policy Studies Institute ceased to be able to function as an independent unit - it is now part of the University of Westminster. "Its fate demonstrated the fragility of small outfits." But if specialist research institutes do need the umbrella of a university, academic reward systems need changing. And perhaps academic values, too.

Looking back, Lewis highlights those occasions when research commissioned by the foundation did make a difference to policy and, perhaps, to people living with disability and disadvantage. It is not that research in itself changes practice, she says, it provides the foundation of knowledge that can lead to change if people other than the researchers get involved. Is that aspiration really too "committed" for academic social scientists? If it is, why do social research at all?

ÖйúAƬ