In yesterday’s Atlanta Journal-Constitution an article reported that the Georgia State Department of Education knew students would fail the social studies test in huge numbers, but refused to share this information with superintendents, teachers or parents in the state. 69% of students in grades 6 and 7 failed or as the state puts it: “Does not meet expectations.” Sixty test items were on each test. 6th graders needed to get 31 correct; 7th graders needed 32 correct responses.

Apparently the test developer, CTB/McGraw-Hill “tested” out 80 questions. They administered these questions to a sampling of Georgia students who had not had instruction in the content upon which the CRT items were based. In this pilot testing experience, the students did not do very well. From my point of view, no kidding. However, a committee of teachers used the results to select 60 items that became the test.

THERE IS A BIG PROBLEM HERE.

Criterion-referenced tests (CRT) are used by the State of Georgia in their annual testing of students in mathematics, social studies, science, and other areas of the state curriculum. A CRT is a test that is referenced to a specific set of objectives or subject matter. These objectives or subject matter are defined as the Standards (in mathematics, or social studies for example), and indeed educators connect the two concepts and refer to CRTs as Standards-Based tests. To develop CRTs in mathematics, reading, social studies, or science, for example, one needs a set of Standards. These have been defined, and through a trickle down effect, states have modified national standards to create sets of state standards.

For example, in the Georgia Social Studies Standards, grade 6, students study the following: LATIN AMERICA and CANADA, EUROPE, AUSTRALIA and OCEANIA. Regardless of grade level, the content is defined in terms of four strands: history, geography, civics, and economics. Technically these strands are really additional content areas that run across Latin America, Canada, etc. Each standard is given a code number and defined. Example: SS6H4 The student will describe the important developments in Europe between 1400 CE and 1800 CE. (see the full standard in the figure below) Note the code breaker: ss=social studies; 6= grade 6; H=history strand; 4=first objective. You can go to the previous link, click of grade 6, and a PDF file will download onto your computer.

If you are 6th grade teacher responsible for teaching social studies, there are 32 Standards listed at the Georgia Department of Education Website on social studies standards. As I mentioned earlier in this post, the actual test that students were given contained 60 items which would translate into 1.8 questions per standard. I have not seen the framework that the test developers used, but one would assume that coverage of all the standards would be a prime goal of the test makers.

Even though the State claims that the tests are criterion-referenced, they appear to be using the same procedures used in norm-referenced tests. A norm referenced assessment is one where student performance or performances are compared to a larger group. Usually the larger group or ‘norm group’ is a national sample representing a wide and diverse cross-section of students. Further, the State is using a test score as the ONLY indicator of performance in the high-stakes tests such as mathematics and reading. No other student performance is taken into consideration. How can this truly be a way to assess our students and our schools?

The State uses the excuse that they are trying to develop “tougher standards.”  If you take a careful look at the standards in any field they are no different that statements of objectives, an idea that was developed first in the 19th century, and then dominated education in the 20th century.  In research done looking at international tests, tougher standards usually comes down to writing more “complicated” test items that result in confusing students, not leading us toward an understanding of their learning.

The system needs to change.