Testing in Education: An Identity Crisis

Posted by on Feb 5, 2013 in Blog, Opinions

Testing in Education: An Identity Crisis

Are you for success of the whole? Or success of the individual? Is there a way to measure both, or just one?

Is education the filling of a bucket? Or the ignition of a flame? Obviously the amount that is in the bucket would be an easy measurement, but what about the “motivation” to increase the size of your bucket? How do you measure that?

Academia, bureaucracy, analytics, statistics, and democracy are all biased against the individual in favor of the collective, incremental good. But does collective, incremental improvement create happiness for the individual?

Are collective, incremental gains concurrent with the goal of education or is education a more personal, individual mission? In education, are we filling of the bucket or are we defining the bucket? Is it possible that conformity and standardization are extinguishing the flame of motivation as a byproduct?

This bias towards systematizing our schools is inherent in the modernization of the nation’s education. Individual differences are smoothed out by numbers and the students’ “superfluous,” individualized interests are left superfluous. The opportunities to create inspiration and ignite motivation are rarely measurable. What is the metric for inspiration?

Is this a defeat of spirit?

In psychometrics classes, large numbers and extensive sampling are necessary. Individuals are carefully conceptualized before all tests are administered, but sometimes the validity of the test isn’t well-known until the test has already been taken. Education gap wideningIn a normal high school history class, tests are content-driven and highly specific inquiries about the contents of your bucket. While in a job interview, the applicant is supposed to “stand out” and be a “special” graduate of the content-specific combine, i.e. their college and major.

Troubleshooting this dichotomy is a losing effort. These two processes of ordering achievement negate one another. Teachers are being forced into administering the contents that are allowed into the bucket, students are having the bucket poured over their heads, and the testing community is salivating over the NCLB-allocated government contracts.

The goal of hermeneutics is a humanist, idiosyncratic interpretation of one student while psychometrics assigns a model to our cognitive processes. Psychometrics includes “measurement” and cognitive “diagnosis” schemes that assert models of the average examinee’s content-specific thought processes. Human functions are implicitly algorithmic here, and the randomness and blurriness of higher-order processes, in effect the “humanity” of the individual, has been stripped because it can not be modeled with graphical analysis and Bayesian estimation. Variance cannot be eliminated from the human interactions that culminate in gaining an “education.” Algorithms cannot replace the teacher-student relationship.

Capitalism is an economic philosophy that lauds individual gains as a function of efficient supply chains, automation, and forecasting, i.e. systematization that streamlines productivity and profit for the few with the power to implement it. If we aren’t careful, the success of private equity hedging may influence the application of this perspective onto the nation’s education system.

Despite the fact that the natural order human beings evolved within, the current procedures for creating predictable results subverts spontaneity and subjugates nature into something repeatable and essentially unnatural. In contrast, babies emerge from the womb with eager eyes… learning to eat, crawl, walk, speak, run, jump, emote, share, etc. This natural urge for learning is then being confined to a classroom that is becoming more standardized… and to the child, it is rendered flavorless.

The problem is that a human being, whether you’re talking about the student or the teacher, will not go gently into that pre-packaged destiny. We must determine how to ignite their curiosity, give flavor to the content, learn from the data, and resist the temptation to eliminate the variance of the model.