There are different things to know and multiple ways of knowing. Our world is not to be understood by either quantitative methods of inquiry nor by the qualitative alone but neither can both together in the best of mixed methods studies lead us to the holy grail of consensus as to what works. Rigour of design and application will, hopefully however, take us a great deal closer. In the opening paper in this volume the holy grail in question is leadership, or more specifically principalship, a subject on which Jim Spillane and colleagues are amply qualified to comment.

Randomised control trials in this field are, as the authors, comment, rare. Therefore, perhaps, a suitable case for treatment. And they are particularly difficult in the ‘real world’ in which human and sociopolitical factors have a horrible penchant for disturbing the best laid plans and efforts at controlled treatments. So, frustrated experimenters end up with overlapping distributions between the treatment and the control groups.

Indeed rare in a field such as leadership and professional development, RCT requires highly complex research designs and analytic strategies, providing the optimum level of evidence to allow the informed reader to agree or disagree with the findings without being sidetracked by the limitations of the data or the weakness of research design.

Data from multiple sources, write Spillane and colleagues, both qualitative and quantitative, ‘allowed us to unpack core constructs such as principals’ knowledge and practice and also informed us of possible threats to validity’. And, they add, ‘While quantitative research often assumes a single underlying truth, qualitative research finds its roots in an interpretive framework that allows for multiple ways of understanding the same phenomenon’. Creating a team of researchers from different traditions and expertise qualitative approaches to data collection meant that qualitative data were never allowed to be the poor relation of quantitative approaches, nor we might add, the other way around. There are different things to know and multiple ways of knowing.

Data, their uses, misuses and abuses is the theme of the second paper in this volume by Mitello and colleagues. Where is the connect, or disconnect, ask the authors, between ‘needs’ (one might say policy imperatives) of school districts and the kind of data that are seen as useful and informative for schools themselves? The nature of summative data provided to schools, write the authors, are often too late to be of use in teachers’ practice, and how meaningful are they in a formative sense?

The authors analysed three formative assessment systems in three schools: ‘Ryders District Developed Assessment System,’ ‘NWEA’s MAP’ and ‘ATT’s Galileo’ and found that Ryders gave no fit with district’s needs; MAP gave only partial fit and Galileo gave a good fit. Their finding that assessments systems in order to fit districts’ needs must be based on the purpose of the district. No surprises there perhaps.

The common sense message is that you can only measure what you set out to measure. The ‘real world’ of inquiry and practice (as we learn from Spillane et al) is not where policy makers live, and the purposes and uses of data at school and classroom level do not always coincide with the purposes and uses by policy makers, politicians and supernational bodies such as the OECD, whose comparative tables provide an unending source of contention and debate. In simple terms, on the one hand, data on student outcomes may serve to control, praise, celebrate or punish schools and on the other to improve teaching and learning. Can the same data be used to serve two masters? There is little evidence internationally to support such a proposition. There is little evidence to suggest that improvement in any meaningful sense grows out of a press for accountability, yet, data intelligently deployed for formative purposes can lead naturally into accountability. Successful and evidence rich schools are generally keen to tell their stories, drawing on a rich mix of qualitative and quantitative data.

Do the district or national data, league tables or accountability measures help teachers to help their students develop reading or writing skills? Following the previous papers Bankcroft’s article casts further doubt on that proposition. Although Bankcroft’s paper focuses on schools in the United States, the tension between teachers’ professional judgements and the demands for standardised test scores is again underscored. It is a growing epidemic internationally and the virus is catching. Even Denmark, widely celebrated as the most democratic, teacher friendly and student friendly of systems is subject to increasing political pressure to leave no child left behind, or perhaps more aptly, no child left untested.

‘No Child Left Behind’, the most appealing of slogans conceals within it a pernicious effect, especially for children and teachers in highly challenging schools where effects often lie beyond their control. The production of numerous state standards for curriculum and instruction followed by high stake tests is a volatile mix, one outcome of which is the closing of schools and/or teacher dismissals—the reward for sub-standard improvement in measured student outcomes. In the study reported here the author was interested in finding out how those standards and ‘benchmark’ tests would influence teaching and learning in classrooms: The study was a multi-year ethnographic study that set out to discuss a rationalistic approach against a neo-institutionalist approach to school reforms.

The study in a low-income school showed that teachers found it difficult to get students up to grade with hundreds of skills represented in the standards because they were not calibrated to teachers’ professional judgements. The tests were viewed as interruptions to classroom teaching and an inadequate means of measuring the progress of those particular students. The data from the tests did not, and cannot by their very nature, give any guidance to what might be done to improve poor performance. It is a familiar story and only surprising in that policy makers seem unable, or unwilling, to realise that such high stakes policies are inevitably self defeating and that the only creativity they inspire is new ways of cheating.

Is it coincidental that the United States is at the very bottom of OECD’s PISA league table for child well being, just behind the U.K. which shares many of its high stakes policies? While the Nordic countries generally do well on measures of child well being, social capital and (with some variability among them) on achievement. In recent years they have begun to feel the pressure of impatient policy and political pressure to compete on a malformed playing field. The final paper in this volume which comes from Sweden adds a postscript to the indicators of student well being. Well being is to some extent a function of who makes decisions that affect your life and the perceived legitimacy and impact of those decisions. The study examined how students in Swedish schools viewed decision-making, the legitimacy of decisions made and process by which those decisions were made and mandated. Three processes examined were student councils, by students via referendum and by the teaching staff. The design of the study was made up of a series of scenario-style experiments where students were exposed to a questionnaire describing various decision-making scenarios.

The themes for decision-making in the questionnaires were: whether to ban religious symbols in schools; whether teachers had the right to confiscate mobile telephones and how to spend money from student fundraising activities. These issues were chosen on the ground that they all had an affect on student well-being.

Given the long tradition for emphasising the importance of student councils in the Swedish educational policy the results surprised the authors as students were most likely to accept decisions made via student referenda, with decisions made in student council in second place and teacher/staff decisions least valued. Where this leads is somewhat inconclusive as it does not touch on the very epicentre of what ultimately matters most. There would appear to room for a further study in which the issues put to the students were more closely related to curriculum, assessment, teaching and learning—the more significant arbitrators of student well being which, as the preceding papers suggest, is always vulnerable to the anxieties and ill-informed judgements of short term policy.

There are, after all, different things to know and multiple ways of knowing.