Blog

As nice as it might be to have consistent proficiency standards across states, what’s shocking is that many states don’t even have consistent proficiency expectations across grades.  Remember, we’re not talking about academic content, which naturally varies across grades, but the relative difficulty of achieving proficiency (i.e., a “passing score”).  The figure below illustrates the problem, showing the difficulty of the reading and math proficiency standards in California for grades three through eight.

In this figure (adapted from NWEA’s most recent California Linking study), proficiency standards are expressed as NWEA normative percentile ranks, with higher numbers indicating greater difficulty.  In mathematics, for example, the third grade proficiency standard falls at percentile rank 39, meaning that roughly 61% of a national sample of third graders could meet the standard.  At eighth grade, the proficiency standard climbs to 61, a standard that only 39% of a national sample of eighth graders could meet.  Put another way, the typical third grader would find it only moderately difficult to meet the math standards in California, while the typical eighth grader would have a much harder time.

Why is this important?  Because under NCLB,  schools are evaluated on what percentage of students at each grade attain proficiency, yet no consideration or allowance is made for differences in proficiency standards.  Imagine a school district in which 70-80% of the elementary students are meeting proficiency standards, but only 20-30% of middle school students are meeting standards.  If the middle school proficiency rates are too low for federal accountability standards, these schools may face sanctions, including forced reallocations of funds in order to boost student performance.

But consider the figure below, a preview from an upcoming Kingsbury Center data report.  The orange lines represent the proficiency rates of a nationally representative sample of students, showing the percentages who meet the California math and reading standards.  In math, for example, 65% of third and fourth graders meet standards, but these rates decrease gradually to a low of 43% at eighth grade.  The story for reading is a little bit different, with proficiency rates starting at 47%, increasing dramatically, and then decreasing by eighth grade to 53%.  These are the kinds of data commonly seen by school administrators, with relatively high rates of proficiency at elementary grades, and lower rates during middle school.

The blue lines in the figure show what the proficiency rates would look like if all grades were held to the same standard…in this case, whichever grade had the most difficult proficiency standard.  As can be seen in the case of California, if students in every grade were tested using consistent proficiency standards, there would be almost no differences in proficiency rate across grades.  In other words, the middle school drop off is entirely an artifact of inconsistent proficiency standards across grades.  Resources that might be spent trying to bring the middle school students up to elementary school performance levels are being wasted, since these students are doing just as well as the elementary students.  One might even say that the problem is not poor performance by middle school students, but expectations that are too low at the lower grades.

Creating Common Core academic content standards is certainly a step in the right direction towards creating a more consistent, more transparent school accountability system.  But the proficiency standards are the real story, not the content standards.  Until we establish consistency across grades and across states, the federal accountability system will never be consistent or transparent.  And meanwhile, school reform decisions will continue to be made using faulty and misleading information about student performance.