The Washington Times
January 2, 2001

We Need Teaching, Not Additional Testing

By Michael Farris
The Washington Times
January 2, 2001

Let us consider a hypothetical situation.

Massive numbers of children are sick. We can tell by taking their temperature. And every few years, a national medical panel releases a new report on children’s temperatures. Every time, the temperatures are unacceptably high.

After each report, the panel of experts says, “We don’t have current-enough temperature reports to make good recommendations. We need more opportunities to take the children’s temperatures nationwide. Then we could figure out what is wrong with the children.”

We would think that they are unqualified to make decisions about children’s health. That is surely what virtually all Americans would conclude. We would ask: “Why don’t local doctors examine each child and prescribe individualized treatment?” We would agree that individual treatment, not more temperature taking, would be the right course of action.

Now let us consider the real situation. Here we see that the common-sense approach in the hypothetical evades the people who continually evaluate our national educational progress.

In early December, yet another federal education panel moaned and groaned about the dismal results of America’s public schools. Wisconsin Gov. Tommy Thompson and former Maine Gov. John McKernan, who both served on the panel, urged that there be more frequent national testing paid for by the federal government.

“How can we make recommendations on old data?” they asked. The data in the study was three years old at the time of their evaluation.

Let’s step away from the environment of a national panel and think through the schooling process a little closer to home starting in the classroom.

When Johnny turns in a math assignment and he only gets 75 percent right, there is immediate feedback that should alert the classroom teacher that there is a problem. Johnny needs help on this assignment, and he needs it now. Johnny will not be helped by an annual national test anymore than he is by the current testing schedule. Johnny needs help on a daily basis. That is the role of the teaching profession.

Standardized testing is not really designed to help Johnny. Standardized testing really tells you how a teacher compares to the other teachers in the building. It can tell you how a school is doing compared to other schools in the system. Standardized testing can tell you how districts compare within a state and how states compare to one another.

What do you learn when you compare classrooms? You have the potential of determining whether one teacher is falling below the standards of the school. Such comparisons are fraught with danger because no two groups of students are exactly alike. But if a building has three third-grade teachers, and the principal gives the teachers roughly equal groups of students in size and natural ability, the principal can make a reasoned conclusion that one teacher is not cutting it if year after year the students in that classroom fall behind.

What do you learn when you compare school buildings within a district? You can make some general assessment of the effectiveness of the principal. Again, there are many variances between schools because of socioeconomic factors and other issues, but on a local level a school superintendent can allow for those factors and make some assessment of which principals are truly effective.

But it is absolutely impossible to make any sense out of a statement that Mississippi schools test lower than North Dakota schools. There are two reasons this is true. There are few universal practices in classrooms in Mississippi that would distinguish them from classrooms in North Dakota. Most researchers would believe that such disparities are only understandable from a socioeconomic perspective. There are just too many variables within a state to make almost any other generalizations about the schools in that state.

We learn virtually nothing useful from national testing because the solutions are never national in character. We have known for a long time that spending per child is no guarantee of success. We cannot find national solutions to education problems even if we test the children every single day.

If children are constantly sick, take them to a doctor to give them individual attention. Individual attention produces educational success. Ask the home-schoolers. National testing produces a pile of meaningless statistics and frustrated social planners.

Education in America will never succeed on the economic principles of the Soviet Union. Centralized planners could not produce enough refrigerators in the USSR no matter how many times they measured the temperatures inside their units. American education planners will never find the magic bullet that centralized bureaucrats seek no matter how often they test the children.

Its time to give up the utopian dreams.

We need more teaching, not more testing.

Michael Farris is the father of 10 home-schooled children and chairman of the
Home School Legal Defense Association

Copyright 2001 News World Communications, Inc. Reprinted with permission of The Washington Times. Visit our web site at