Laura H. Chapman gives more examples of the distortion and corruption of education practice amd policy by econometric language.

Students are performing on grade level if their scores on a standardized test are at or above the median on a percentile scale (1-99). On a large-scale test, a score at or near the 50th percentile (the median) will usually classify a student as proficient in the skills and subject matter on the test.

Expected growth means that gain-scores of students (on tests in a single subject, such as math or art) are staying in about the same location in a distribution from year to year—below average, average, or above average. For a large number of students, the distribution is likely to resemble a bell or normal curve.

Predicted growth is an inference about a student’s future gain-score, derived from a linear regression analysis of two or more years of that student’s gain-scores. This analysis assumes that past performance will predict future performance. Perhaps, but in education, this is a dismal assumption. It can become a self-fulfilling prophecy. The assumption is so risky that almost every corporate report begins with this caveat: Past performance does not predict future performance.

A student is said to have achieved a year’s worth of growth if his or her gain-score on a test of proficiency is equal to, or greater than, the gain-score made by a 50th percentile student. The same measure is applied to teachers. Teachers in some districts are rated highly effective only if all or most of their students have gain-scores of more than a year’s worth of growth.

References to a year’s worth of growth are fundamentally misleading because the common mental picture of a calendar year is different from a school year (typically 180 days); an instructional year (typically 172 days); and a typical accountability year (130 days from pre-test to post-test).

Academic peers are students whose test scores in a given year are the same or nearly the same. This concept permits comparisons of their gain-scores from the prior year to the current year. Students who make greater gains than their academic peers have an accelerated growth trajectory. Students who fall behind their academic peers need remedial work to keep up. The average of the gain-scores for academic peers in a teacher’s classes is typically used as a measure of the teacher’s productivity and effectiveness. This use requires a studied indifference to other influences on test scores.

A growth trajectory needs a target. Targets for learning need to be set using baseline data so the instruction offered to each student, during a known interval of time, is efficient and has a measurable impact on student learning. Meeting targets for learning is analogous to meeting a sales target or a production quota by a date certain.

Teachers and others who say they are “impacting the growth of their students” are not think-ing about the meaning of words. They are parroting econometric jargon.

Experts associated with Metametrics hope to set growth velocity standards. They describe their theoretical mapping of “aspirational trajectories toward graduation targets” in reading skills as analogous to “modifying the height, velocity, or acceleration respec-tively of a projectile launched in the physical world.” They seek greater precision in setting targets and cut scores for grade-to-grade progress in meeting the CCSS. (Williamson, G. L., Fitzgerald, J., & Stenner, A. J. (2013). The Common Core State Standards’ quanti-tative text complexity trajectory figuring out how much complexity is enough. Educational Researcher, 42(2), 59-69.).

Calibration refers to the quest for precision and consistency in measurement in the context of just-in-time delivery of a result, especially manufacturing.. In education, the term means that evaluators and other monitors have followed specifications in rating performances, presentations, processes, and products. Calibration events are training sessions intended to standardize how raters use or interpret language and to verify that rules for making judgments have been followed with fidelity. Such events are also called trainings or calibrations.

Audits are conducted to verify that calibrations are not needed, that rules have been followed, that data are free of ambiguity, and that low-inference definitions of performances and metrics are used consistently. Questions about the validity of the metrics may be ignored.

Bring to scale means that an educational policy, practice, or product is believed to merit replication in multiple locations, as in manufacturing and franchise systems for a mass market.