Brilliant dissection of Ball's argumentative dodges here. The "guided discovery" rebranding is such a slippery move because it lets defenders claim whatever instructional elements they need in the moment while avoiding any real accountability. I've seen similar dynamics inschool board meetings where administrators retreat to vaguer language the second anyone asks for measurabel outcomes. The parallel to balanced literacy's unfalsifiability is spot-on.
I don't think that educational 'researchers' are able to control for parental intervention - which is highly likely to be correlated with student capabilities. Speaking from personal experience, when I was not satisfied with the math teaching approach my kids were exposed to - I supplemented their math education - they had to learn it the teacher's way and dad's way - and dad could and would give more homework than the teacher: Enough to make sure that they learned it my way. Mean dad.
My kids say that they will do the same with their kids if necessary.
But I don't think I am at all unique. I assume most STEM educated parents will do the same - and therefore, regardless of the ineffectiveness of the school program, you will have a subset of kids who master the material (perhaps in spite of the teaching methodology). Ditto reading and other subjects - although non STEM educated parents might be a bit weak on math supplementation.
But the educational researchers seem to assume that the kids learned because of their approach, not potentially in spite of it. The kids whose parents did not supplement the teaching may of course suffer severely.
With bad educational doctrine the children of less educated parents have much reduced opportunity to learn more and improve their chances in life.
This is why rigorous standards for educational research (i.e., the WWC) hinge primarily upon the use of randomized control trials, which were well represented in our original petition. In such designs, confounds like those that you describe are nullified as they are likely to occur equally between the groups examined. If children's parents in a control condition are supplementing as a result of concerns with the curriculum, randomization ensures a likely equal amount of supplementation occurs when testing the experimental method. For very large quasi-experimental studies such as Project Follow Through, matching occurred to minimize this threat, and the overall size of the sample and size of the effect would unlikely shrink to 0 even considering this.
Accepted. You are going to pick up population averages for the instructional changes, but let us assume that you have instruction type A, and instruction type B. Let us consider the case where demanding parents consider A to be acceptable and B to be inadequate. Demanding parents are likely to supplement for B, but not A. The inadequacy of B will not show as much among the children of the demanding parents as it will among the children of the accepting parents who have not supplemented. I have seen people defending the adequacy of programs based upon the fact that the better students were mastering the material (quite likely due to parental involvement).
But if there is reason to suspect systematic parental supplementation in education you are going to need to consider the changes in learning among the students who are not getting supplemental education, as the comparison at the capable end will be systematically biased by the supplementation.
Brilliant dissection of Ball's argumentative dodges here. The "guided discovery" rebranding is such a slippery move because it lets defenders claim whatever instructional elements they need in the moment while avoiding any real accountability. I've seen similar dynamics inschool board meetings where administrators retreat to vaguer language the second anyone asks for measurabel outcomes. The parallel to balanced literacy's unfalsifiability is spot-on.
I don't think that educational 'researchers' are able to control for parental intervention - which is highly likely to be correlated with student capabilities. Speaking from personal experience, when I was not satisfied with the math teaching approach my kids were exposed to - I supplemented their math education - they had to learn it the teacher's way and dad's way - and dad could and would give more homework than the teacher: Enough to make sure that they learned it my way. Mean dad.
My kids say that they will do the same with their kids if necessary.
But I don't think I am at all unique. I assume most STEM educated parents will do the same - and therefore, regardless of the ineffectiveness of the school program, you will have a subset of kids who master the material (perhaps in spite of the teaching methodology). Ditto reading and other subjects - although non STEM educated parents might be a bit weak on math supplementation.
But the educational researchers seem to assume that the kids learned because of their approach, not potentially in spite of it. The kids whose parents did not supplement the teaching may of course suffer severely.
With bad educational doctrine the children of less educated parents have much reduced opportunity to learn more and improve their chances in life.
This is why rigorous standards for educational research (i.e., the WWC) hinge primarily upon the use of randomized control trials, which were well represented in our original petition. In such designs, confounds like those that you describe are nullified as they are likely to occur equally between the groups examined. If children's parents in a control condition are supplementing as a result of concerns with the curriculum, randomization ensures a likely equal amount of supplementation occurs when testing the experimental method. For very large quasi-experimental studies such as Project Follow Through, matching occurred to minimize this threat, and the overall size of the sample and size of the effect would unlikely shrink to 0 even considering this.
Accepted. You are going to pick up population averages for the instructional changes, but let us assume that you have instruction type A, and instruction type B. Let us consider the case where demanding parents consider A to be acceptable and B to be inadequate. Demanding parents are likely to supplement for B, but not A. The inadequacy of B will not show as much among the children of the demanding parents as it will among the children of the accepting parents who have not supplemented. I have seen people defending the adequacy of programs based upon the fact that the better students were mastering the material (quite likely due to parental involvement).
But if there is reason to suspect systematic parental supplementation in education you are going to need to consider the changes in learning among the students who are not getting supplemental education, as the comparison at the capable end will be systematically biased by the supplementation.
You should have an "Executive Summary" of your full-fledged argument/rebuttal. TLDR always applies...
How depressing. What is it about education that lets this kind of thing keep happening?