Twenty personalities, Michel Rocard, Daniel Pennac, Richard Descoings, Marcel Rufo and researchers as Axel Kahn and Eric Maurin, a colleague, among others, have recently joined an appeal launched in September by AFEV aimed at eliminating the use of notes in elementary school. The main arguments are well known, the notes would discourage students, undermine their confidence in them and the conclusion also striking to note how a school cooperation rather than a school competition (see here ).
As a teacher, my view is actually shared personal assessment must be part of the route of a pupil or student, because it takes many elements it needs to be within the absolute respect and possibly all of these comrades. Moreover, this assessment does not necessarily reduce to a note, but can take many forms. I also understand the arguments of Eric Maurin, which probably gives a more sociological reading of the problem, based on factual evidence that is indisputable. As an economist, I must say that incentives have an effect, and the whole issue is whether the incentives from the rating is more effective or less effective for student achievement that the lack of system incentive or another incentive scheme.
Moreover, by the way, striking the notes is also an issue of public policy, insofar as the effectiveness of education policies, when it is measured - is often gauged through the variation of performance Student against those famous notes. Obviously, this approach to evaluation by the notes can be only very partial, and it was said long ago that the effectiveness of education reforms should be judged on multiple criteria other than performance terms of notes, for example through the impact on the general condition and behavior of children and adolescents (health, tendency to addiction, etc.).. I quote including Bowles and Gintis Obsborne in 2001 in the American Economic Review about the limitations of the notes (test scores) for evaluation performance in school and the need to go beyond:
"(Economists) Need Broader Indicators of School Success, Including Measures based On The contribution of schooling to behavioral and personality traits"
It also appears that the relationship between grades in schooling and income earned by people later became active is relatively thin .
From a behavioral perspective, the question that is asked is whether the impact of information that can get people on their performance current levels of effort to come and future performance. Or, put differently, what is the incentive effect of these famous notes on the level of effort in the work of students?
This issue of the impact of feedback is at the heart of the work on the economics of human resources (I do not know better translated "personal economics," but said it may be just the economy of staff), which is little or my specialty. The empirical literature, especially experimental, has been important in recent years on this issue. I already
mentioned in this ticket problems there might be trying to establish a system of incentives in a context of education, the perverse effect is that extrinsic motivations substitute for intrinsic motivation and the Total net effect could ultimately be negative. Mutatis mutandis, the effect of motivation related to the notes could oust the taste of staff work and effort.
particular, the question of efficiency scores (feedback in general) can be approached from two aspects: the fact that the level of my note gives me information Absolute on my performance, and secondly, the relative level of my grade compared to other students, if this information is public. If it is public, it can take many forms, my grade can be compared to an average possibly coupled with a standard deviation provided by the school, show up to a perfect knowledge of the distribution of ratings.
From an economic standpoint, feedback on past performance can affect my performance standard either directly, as past and present performance are substitutes or complements in the utility function of the agents, or indirectly by revealing to the individual information on the performance of its effect (signaling effect).
This controversy is timely, since the literature on this issue is booming in the experimental field. Recently, Azmat and Irriberi, 2010 published the results of two studies that focus particularly on the impact of feedback in part on performance, but also on the welfare assessed. They focus in particular on the impact of feeback on the performance of individuals. They observed a significant effect on scores on the final performance of students, and therefore their level of motivation, on the occasion of a natural experiment. They confirm these results in this paper , but also highlight the negative effects on the well being of students below the average of a measurement system relative to an absolute measurement system.
However, the study's most interesting because the nearest of the question is that of Todd Cherry & Larry Ellis, in an article published in 2005 in the International Review of Economics Education . These authors compare the impact of an absolute evaluation system (my my grade or rank on a scale of 1 to 10 for example) compared to a rating system on. In the evaluation system on my score (my final) is determined on the performance of other students. For example, if I answer more questions than 90% of my class, I get an A. If I say better than 75% of my class, I get a B, etc.. This system is particularly strong competition in front of students.
is precisely the effectiveness of any scheme of assessment on the final performance students they wish to measure. As they have 4 classes in an introductory course in macroeconomics, they set up for two of these groups on an assessment (rank order grading) and the other two absolute evaluation (Criterion-referenced grading). The chart below shows the distribution of scores finally obtained in the examination according to the two incentive schemes by Notes:
Source: Cherry & Ellis, 2005
As it is difficult to draw a clear conclusion, these authors make an econometric analysis of the score through the treatment as an explanatory variable "rank order grading 'treated as a variable dummy. The results are quite explicit: the average score of students is better in the scheme which provides an assessment on the student rather than an absolute assessment. They are cautious, however, and insist that this kind of rating system is not necessarily appropriate if the goal is to promote cooperation among students, the evaluation system is absolute, in these conditions, probably better.
To conclude this post, this is not the rating itself appears to be a problem but how this notation is used by teachers and students to compare themselves to others. That can backfire. However, the effects of competition are not always bad (I'm sorry to remind so obvious, but many get carried away in a big discussion about the perverse effects of competition, forgetting some of his indisputable merits). Finally, I borrow from this week's Telerama conclusion that I found beautiful. A French teacher told a student collapsed by its note of 7 / 20:
"You are not what 7 / 20. This 7 / 20 is what your work was worth last week between 8 am and noon. Made the difference. "
is one who has all inclusive pedagogy. More reason not to break the thermometer too fast and think of a weighted potential consequences of the abolition of the notes.
