Thanks Ursula. Hopefully we'll get some more responses.

Well, I would suggest that this approach is rooted in the reality that school, like society itself, operates as a sorting mechanism. I also would argue that there is value in such sorting. (If I remember correctly Pinker argued that sorting is part of our evolutionary heritage; if he didn't, he should have.) We don't have to think either long or hard to see the advantages in assessing performance. When taking my car in for service, do I want a superior mechanic to do the work or a mediocre one? The mediocre mechanic may have improved significantly during training, but he never reached the superior level. Also, I question the proposition that cohort assessment favors the extroverted in all situations. Introverts tend to excel academically, although their discomfort in social settings can be an obstacle in certain areas.

James, I accept your points but ask if you would have a different view if I focused on lower age groups, say up to 15 years old. Prior to this age children are still developing resilience and beliefs about themselves in different contexts, such as school. My point is that many of these students, mostly by accident of birth, socioeconomic status etc, are positioned by many factors to believe themselves to be failures at school and so disengage earlier from education than they might gave otherwise. This reinforces their views unfairly so that they do not gave a chance to break out of their status, and society misses out on their possible input. At later stages, performance measures are OK because students have already developed their belief systems. Does that influence your viewpoint?

Mark,

That's a tough question because the realities you listed are largely social in nature, as you and I have discussed several times in the past. I don't believe that our schools have the ability or even the potential to affect these social issues, but I do believe that teachers have the potential to inspire even our most disadvantaged students in ways that will help--or perhaps "allow" is a better term--them to excel academically. When they succeed, the self-perception of the students will be based on their performance when facing challenging tasks, not on some artificial measures designed to elevated their self-esteem. Here in the States, we embarked on the Self-Esteem Movement in the early 1980s, with painful results. Students know that performance is what counts; what many do not know is their own potential.

James, I am aware of the issues with self esteem and have researched modern ideas of self efficacy and resilience as different ways of approaching this issue. I dont think schools or teachers can fix the problem by being supportive because the issue is social and systemic. If they could, the problem would have been fixed. Teachers are nothing if not supportive. I think though that changes to the system can reduce the negatives. It isn't necessary to build students' self esteem, just not knock it down too early in an educational context. It intrigues me that educators seem to accept that because performance measures are the way it has been done and is done, they will always be done in all situations. A more nuanced approach that is not completely out there like Summerhill or OBE, which I thought had potential, is surely possible, given the right political and academic support. Regarding your comment about self esteem being developed by performance on challenging tasks, that is exactly the point. Too many students' must get a fail performance from an early age because of the nature of performance tasks,so that they never actually develop an honest understanding of what they can do, merely that they cannot do anything. This attitude is the one that remains with them and blocks their continuing improvement.

It is possible to to reframe performance measures as 'graded ability', That is what was supposed to happen in Queensland Australia, however these measure have too much historical baggage and parents, students and teachers all position the performance measures as pass, pass well or fail. This is why I believe a change to the structure of educational attainment measures is needed.

PS your use of the term 'allow' is useful. I try to get teachers to ALLOW students to make their own mistakes so they learn from them rather than MAKE them either right and take away the learning, or suffer because they were wrong. Too many teachers want to MAKE students suffer as if that is needed for learning (despite psychological research to the contrary) ALLOWING them to recognize their own consequences is far more valuable.

In a sense the current system MAKES them suffer the emotionally negative consequences of failure. It would be better if the system ALLOWED them to see the consequences of disengagement in an emotionally neutral way.

In the end I am trying to create a new social attitude towards assessment. One that is emotionally neutral, at least for younger age groups, so that students develop the mindset that ‘assessment is a way to know what I can do and how well’, rather than the current mindset that ‘assessment sorts the ones who can do school from the ones who can’t do school’. That this latter mindset predominates has been evident anecdotally over my 30 years of teaching this ages group. Unfortunately, until now, I didn’t think to do proper research on it. Far too many students have said to me in so many ways that it is ‘better to not try than to fail’, which speaks of lack of self efficacy and resilience in this context.

Yes, exactly. Low expectations from teachers regarding students, especially those from low SES backgrounds, is a huge problem with regard to addressing the social issues that we both recognize. I believe that the first step is to have better teachers. Many years ago, I chaired a state-wide committee charged with developing proposals to improve public education in California. Our top two proposals were: (1) raise admission standards for credential candidates in schools of education that match those for other graduate school and professional school standards; (2) establish salary scales that matched those of other professions, such as law and medicine. Our proposals were blocked immediately by--guess who?--the teachers' union. Go figure.

It is true that unions can be part of the problem. They are as limited by social constructs as anyone else. I applaud your ideas for improving teachers.No matter what I believe about assessment, better teachers make a difference.

My 2 key points are though that:

1. This issue impacts inequitably on low SES students because they more commonly have a skillset deficit on entry to school. That this is true is evident in Australia in indigenous kids and in America in black kids via PIS testing - drawn from another RG question.

2. It is very difficult for a teacher, no matter how excellent to change a student's perception of themselves as a learner, once formed in the light of previous results.

James and Marc, interesting discussion. I would like to add only one point. People or students are graded individually; learning, however, takes place collectively in social contexts. This insight, not new at all, however, is mostly forgotten in our grading systems - as practices of grading all over the world illustrate. But if one takes the insight seriously it becomes obvious that grading is the medium and result of grading practices on different interlaced levels, the level of interaction in class, the school level, the level of school district or the organizational field of learning and so on up to the level of the dominant culture. The social meaning of grading is constituted via these socially interrelated practices. Scripts of what teachers, students etc. do (or have to do) and what are appropriate ways of actions are for example the medium and result of these practices, practices of grading included. Thus, the social meaning of grading is socially constituted. And the meaning of grading include social understandings of how people should be selected, included or excluded, and positioned in relation to their future chances in life (or working life) in and through grading practices. The socially constituted understandings of grading, thus as Foucault would have pointed out, produce our way of dealing with grading. The interplay of different levels of grading practices can, I think, by the way quite nicely be seen if one looks at the example of Summerhill, the example you mentioned in your discussion. Grading in the famous Summerhill experiemnt can be done quite differently than in the normal school system. However, as one could say with reference to Durkheim: Obviously you can act otherwise as socially expected but you wil have to face the consequences. However I should quickly add, grading systems and grading practices are not simply given. They are produced and reproduced by more or less powerful agents, students, teachers, unions, politicians and so on in and through their socially embedded grading practices in time-space. Grading systems and grading practices are, thereby, prolonged or even sometimes changed - nowadays mostly with the involvement of organizations who coordinate activities togethe with others in networks.

I would also suggest that our students are thoroughly socialized when it comes to assessment. One of my early research projects was a study of student motivation in composition classes. A key item on the student questionnaire asked respondents to indicate the level of work they would invest in a graded v. an ungraded class. Nearly 70% indicated that they would work less hard in an ungraded class. Written comments from respondents indicated that the majority viewed grades as the equivalent of a paycheck for the work they perform.

James & Arnold, I accept what you say completely. From Arnold's response, I am asking 'can socially produced expectations & readings of assessment practices be isolated in one context ie in schools?', I reiterate that I am specifically focussing on children up to about 15 years of age who will be much more variable in educational resilience & self efficacy than those who have succeeded enough to enter university, such as those in James' study. James, your point about 'paycheck' is interesting because many of the students I am talking about have little belief in their capacity to influence their paycheck with effort. Many of them come from families that have multiple generations of unemployment.

I will admit that I already have some evidence that it is possible to create an isolation of beliefs to within a school setting as that was the basis for my action research 15 years ago. The problem is that I have found it extremely difficult to gain acceptance of even He possibility of this idea beyond the school as in the Summerhill experience, what has been shown to work with fewer negative effects appears to be less acceptable than the current paradigm even with its negative Social consequences, I guess it is an example of Kuhn's description of the process of paradigm shift,

PS. James, with respect to the issue of graded/ ungraded.I accept the truth of that. What else would one expect, If something is perceived as no value including not interesting to you. ( I note that 30% said they would attend). I dealt with that in my project by describing engagement in the work as 'everything is a chance to give you credit. If you don't engage, I can't give you any credit. If you do engage, I can qive you credit for any improvement for which I can find evidence'. In other words I described engagement in work as an opportunity to provide evidence of what they could do, and hence influence the final result.Naturally I then had to provide evidence that I truly did what I said I would,

Mark, my answers to your questions are:

First, I would expect that socially produced expectations, views and understandings of assessment practices of different settings in a common field of activity like education are always interlaced to some degree. However, this includes that they might differ structurally, that means, the rules of signification, and legitimation as well as the ways of using ressources in assessment practices can develop some kind of system specifity. Therefore, for example the cognitive and normative orientations of actors in one system, e.g. classroom interactions, might differ, say, from the orientations agents of the school administration or the school district follow. Thus, the expectations, views and understandings of assessment practices can differ to some degree. at the same time. At the same time, they should, that would be my assumption, show some basic similarities between systems in a common field

Second, I would start with the assumption that children up to 15 years are knowledgeable agents too. They - due to, for example, their experiences of the school system and the ways how one talks about the school system and the practices of assessment more generally - do not only have knowledge about the school ássessment practices; they are also able to practically use this knowledge in their doings. Their knowledge, however, I would expect, might be, as I said, mostly practical knowledge in the Wittgensteinean sense. That means, the pupils know how to use their knowledge of assessment practices in their doings but are only to some extent able to discursively articulate this knowledge and use this knowledge discursively.