Do all of these statistics, committees, readers, and writers make the PSSA a fair and valid test?

The hundred million dollar question! Though many statistics are used to ensure that test questions are as statistically fair as our mathematics system will allow, subjectivity by “expert readers” and committees will inevitably occur. (After all, we are all human beings.) When reviewing the results of the 2006 PSSA in the Technical Analysis of grades 5, 8, and 11, there are differences in the scores of male versus female as well as differences in the scores that were collected from the different subgroups. For example, students that were Asian had different scores from those students that were White. (DRC Tech Report 5, 8, 11). In looking at these diverse scores, a troubling question crosses one’s mind. Are the differences in scores of the different groups based on a true educational disadvantage of one group or are some of the lower scores, particularly in the Analysis and Interpretation of Fiction, a manifestation of how students from diverse backgrounds interpret questions? Must the government insist that all of our students view these works in the same way? Arguments can be made for and against this question.

To answer in a less thought provoking manner, the PDE commissioned Human Resources Research Organization (HumRRO) to study the validity of the PSSA. Basically, HumRRO reported that though average scores differ between the different groups of students (white versus black), this difference is also seen in other standardized tests that are comparable to the PSSA. (Thacker, Dickinson, Koger) Additionally, students from a higher socioeconomic class have higher scores on the PSSA as well as other standardized tests. Also, those students who perform well on the PSSA will also perform well on the SAT and university placement exams. (Thacker, 2004) (Sinclair, Thacker 2005).

The question now becomes, if our students do well on the SAT, will they succeed in college? The College Board would say an emphatic “YES”, but evidence shows that the relationship between the academic SAT and the prediction of freshman grades is rather weak. (Sinclair, Thacker 2005) In fact, between the introduction of the essay questions on the 2005 SAT and the SAT tests general failure to make accurate predictions about college success, many colleges have ceased to require the SAT for admissions or at least decreased the importance of the test in the acceptance process. Other factors such as grades and class rank are actually a stronger predictor freshman success. (Perez, 2002)

So, in other words, a student can get good grades, score low on the PSSA and SAT and still do well in college or any other future endeavor. This fact supports what is common knowledge for most of us (but apparently not for our government): a motivated student can learn and succeed whether or not they can sit for a couple of hours and take a test that is supposedly fair and unbiased.