Why will it be a 'true' measure? Redistributing letter grades is a statistical trick to supposedly make up for unreliable assessment processes ie anomolous marking or exam design. This process does not guarantee a better correlation with absolute ability of the students, than criteria based assessment processes where distributions are allowed to vary. After all in any particular group of students the distribution of skills and knowledges is not necessarily going to follow a bell curve.

LHR=?

Are letter grades meant to be anything else but a measurement of achievement? What is the intended purpose of a letter grade?

It is actually a grade percentage rather than a letter grade. They represent completion of assignments mostly.

Hello,

As far as they are called GRADES they are a measurement of achievement!? unless I misunderstood the question.

Regards

Theodora Issa

Dear Dr. Mary Frances Rice.

Letter Grades are based on true distribution of class average and it follows Gaussian distribution curve with center as average (C). The LH of the curve C+ --> A+ whereas, RH side of the curve distributes the grades C- ---> F. If truly reflected objectives are matched with the course and evaluation process then it will be true measure.

Why will it be a 'true' measure? Redistributing letter grades is a statistical trick to supposedly make up for unreliable assessment processes ie anomolous marking or exam design. This process does not guarantee a better correlation with absolute ability of the students, than criteria based assessment processes where distributions are allowed to vary. After all in any particular group of students the distribution of skills and knowledges is not necessarily going to follow a bell curve.

Given the high degree of subjectivity associated with so much classroom assessment, it is difficult to know exactly what grades reflect. Some scholars have used this reality to reject entirely the validity of grades. Nevertheless, it seems reasonable to propose that grades have some correlation with classroom performance, even if that correlation is statistically low. On this account, using grades to indicate performance is justified, even though, as Mark noted, grades do not tell us a great deal about students' abilities.

If truly reflected objectives are matched with the course outcomes and evaluation process then it will be true measure.

I think you can justify the use of grades as one measure of achievement. But you should also have a control for variation in ability level that might be a stronger predictor of success.  How are you controlling for pre-existing differences in ability, knowledge etc that are unrelated to the intervention under study?

Peter, are you talking about measuring learning GAIN rather than absolute performance?

Good question Mark.  I guess I'm unclear what the outcome measure is.  The question refers to "achievement" and asks if grades can be the measure.  Either way there should probably be a check to see if either gains or absolute performance are correlated more strongly with some metric of incoming ability level.  Otherwise we really don't know much from the study.  We need more context to be able to answer the question accurately I think...

I agree that my question to you is not part of the parameters of the original question. I am interested in the whole topic of measuring learning as I think the poor measures we use in schools are responsible for the low performance of some students.

I agree - we don't have great measures of learning in schools.  Too often poor measures (or merely convenient ones) are used and the results are very limiting.  I am interested in learning as the closing of the gap between a current and projected identity and I think schools should become more inclusive places.  Currently schools in the US  are so narrowly focused on a few kinds of abilities that they seem to ignore many others that are valued across societies. Students who feel marginalized by this narrow focus would really benefit if we broadened definitions of what it means to learn and what is worth learning.  We really could be re-thinking learning in enormous ways.  But its sort of against the grain to talk that way these days...

Arguably, grades mean something rather than nothing, and given that we have no other system in place to designate our perception of student performance, I would say that we are, indeed, justified in using them in studies of achievement.  By way of disclaimer, I should note that I have used grades as measures of achievement throughout my teaching career, and I also use them in my some of my research.  

Now I'm aware that from a 1990s post-structualist perspective grades, like everything else, have no meaning, and given the insane grade inflation in our public schools and universities, I am sometimes tempted to agree with that perspective, even though it has become passé.  But then I reflect on one of my very early studies of student motivation, which found that the undergraduates I examined indicated that they would not work as hard in any class that was ungraded, and the temptation vanishes.

Thank you for this insight, Dr. Williams. I think you summed up the tensions I am experiencing very nicely.

James, We've been down this track before haven't we? Getting rid of poor gradIng measures doesn't mean non graded. What I want is more meaningful grades, esp in primary and secondary school. By meaningful, I mean effective measures of improvement and/or a range of informative measures graded along a continuum. I am less concerned about the measures used at university as:

l. students are already mature in their self concept, and

2. students who get to university are likely to have a positive self efficacy.

So I am not so bothered by choice of metric at university.

That those student said what they did about qrading and motivation might be cultural, in the sense of learned values. Maybe they can't imagine anything different than needing external approval. Recent work on motivation suggests that there are multiple levels of motivation ranging from amotivated through wholly extrinsic, to fully intrinsic and every shade in between. These students sound like they are still at the introjection level, where they still require some approval from others to maintain motivation. The work I am referring to is "Intrinsic and Extrinsic Motivations: Classic Definitions and New Directionsi" by Richard M. Ryan and Edward L. Deci 

Mark,

Here in the States, a driving force behind grade inflation appears to be tacit recognition that a growing percentage of our young people are unable to meet target learning outcomes (another being teachers' desire to better student evaluations).  I'm not sure what you mean by "more meaningful grades," but from an assessment perspective, such grades would be accurate measures of student performance.  If a learning outcome for 3rd grade, for example, is that all students will be able to add, subtract, multiply, and divide fractions, then a meaningful grade would indicate their ability to do so.  At the end of the term, we would give them 100 fraction problems and assess their accuracy rate.

We certainly could establish such a grading system, but any objective measure of performance would result in socially unacceptable failure rates. 

With regard to self-concept--or self-esteem--we have psychological instruments to measure this factor, and the research indicates that there is a positive correlation with academic performance.  The problem here is threefold:  First, we don't have a clear understanding of the sources of self-concept; second, there is a reciprocal relation between self-concept and academic performance; third, academic performance in this line of research has been measured through grades and/or teacher reports.  Two points warrant emphasis.  The public-school self-esteem movement in the States during the 1980s and 90s was based on providing positive feedback on student work even when it failed to meet established learning outcomes, leading many students to believe that failing work was brilliant.  Upon university matriculation, they were shocked to discover the lie.  Using grades as a variable in measures of self-concept/self-esteem takes us right back to the original question; the self-concept research is therefore quite circular in this context.

With regard the the notion of "more meaningful grades," the point of my previous post was this:  More meaningful for whom?  Grades are significantly meaningful for students.    Yes, the value they place on grades reflects their extrinsic motivation, but then the motivation research also shows that intrinsic motivation starts to decline in 4th grade, so how can we be surprised?  Moreover, we all are extrinsically motivate in one way or another, even those of us who have high levels of intrinsic motivation.  Hegel argued--convincingly, I believe--that the intersubjective nature of the human experience leads us to crave recognition.  I've proposed in some of my publications that the desire for recognition clearly has negative consequences, but it just as clearly has positive ones.  At the end of the day, does it matter, as long as individual performance is excellent? 

Nicely said James. It sounds like agreement with the key points to me. Everything I've read and seen over the last 20 years of teaching suggests that the key unintended consequences of our grading paradigm is that:

1. Lower level students decide they 'can't do school' and give up, and

2. Very capable students decide they are 'gifted' and end up not trying very hard.

Most recent research into self esteem and progress towards intrinsic motivation seems to point to specific qualities being useful.

* a reduction in competition,

* a clear link between effort and progress, and

* honesty and clarity in feedback

Positive self efficacy is necessary for younger students to maintain the quality of general education so grading that minimises comparison against a norm, yet clearly describes position along a continuum of learning is desirable UNTIL it is absolutely necessary.

Examples of times when comparison against a standard may be necessary could be:

* end of general education, when selection of specialised subjects occurs (although even this may not be needed as students who don't do well probably wouldn't choose that subject anyway)

* end of secondary education and end of university, where students are selected for future positions based on quality.

Otherwise there doesn't seem to be any value in comparative grading in early grades and we should focus on more informative and less comparative grading paradigms.

I have trialled a system that does what I describe in an experiment over 5 years and found it worked as expected. It helped change the full range of students' (and teachers') attitudes in a way that maximised their continued motivation.

Mark,

Once again, I'm pleased to find that we share many similar views.