The things taught in schools and colleges are not an education but the means of an education. Ralph Waldo Emerson.

There is no such thing as measurement absolute; there is only measurement relative. Jeanette Winterson.

#GrowthModels and longitudinal scales

We dream about measuring cognitive status so effectively that we can monitor progress over the student’s career as confidently as we monitor changes in height, weight, and time for the 100-meters. We’re not there yet but we aren’t where we were. Partly because of Rasch. Celsius and Fahrenheit probably did not decide in their youth that their mission in life was to build thermometers; when they wanted to understand something about heat, they needed measures to do that. Educators don’t do assessment because they want to build tests; they build tests because they need measures to do assessment.

Historically, we have tried to build longitudinal scales by linking together a series of grade-level tests. I’ve tried to do it myself; sometimes I claimed to have succeeded. The big publishers often go us one better by building “bridge” forms that cover three or four grades in one fell swoop. The process requires finding, e.g., third grade items that can be given effectively to fourth graders and fourth grade items that can be given to third graders, and onward and upward. We immediately run into problems with opportunity to learn for topics that haven’t been presented and opportunity to forget with topics that haven’t been re-enforced. We often aren’t sure if we are even measuring the same aspect in adjacent grades.

Given the challenges of building longitudinal scales, perhaps we should ponder our original motivation for them. For purposes of this treatise, the following assertions will be taken as axiomatic.

1. Educational growth implies additional capability to do increasingly complex tasks.
2. Content standards that are tightly bound to grade-level instruction can be important building blocks and diagnostically useful, but they are not the goal of education.
3. Any agency will put resources into areas where it is accountable and every agency should be accountable for areas it can effect.
4. Status Model questions that Standards-based assessment was conceived to answer are about school accountability and better lesson plans, e.g., Did the students finishing third grade have what they need to succeed in fourth grade; if not, what tools were they lacking?
5. Improvement Model questions were added as annual grade-level data began to pile up in the superintendent’s office and are asking about the system’s improvement, e.g., Are the third graders this year better equipped than the third graders last year?
6. Growth Model questions are personal, Is this individual (enough) better at solving complex tasks now than last year, or last month, or last week?

Continue . . . Longitudinal Scales

Previous                                                                      Shortcuts