My experience with benchmark testing kicked off during my first week as a school psychology intern. These were the days before the terms RTI or MTSS were in existence, and AIMSweb was not yet developed. Nonetheless, my school psychology internship supervisor had it all figured out. Fresh out in the field, I remember the thrill as I faced the reality that what I had learned from books and during classes was finally coming into play!
These were the earlier days when we used Curriculum-Based Measures pulled directly from the curriculum that students were instructed from. Growth or lack of growth was somewhat clouded due to the varying difficulty levels of the measures from each benchmark screening or monitoring occurrence. This is where I discovered the importance of having multiple forms that are similar in difficulty level — crucial in both benchmark testing and progress monitoring. One key lesson that I learned: The growth or lack of growth should not be attributed to variability in form difficulty.
Again, thinking back, we were in the earlier phase of technology usage. We had software programs that we could complete calculations and help with documentation of data, but there was nothing specifically developed to plot and analyze our screening and monitoring data both at individual and group levels. I found myself assisting teachers for countless hours both plotting data and analyzing trends by hand. The paper wore thin across the year as I erased and recreated trends each time data was collected. This was not an efficient way for us all to spend our time. I learned: Benchmark testing should provide accurate results in an efficient manner. It should also lend itself to varying levels of interpretation.
We had all of our data collected with administration and scoring fidelity in mind. We had the data nicely calculated and printed on graphs for teachers to view and meetings to take place. What we didn’t have was a robust norming sample to compare our results to. We relied on research studies that gave us a reference to determine what scores might be to show overall achievement. This was helpful, but led me to question what was actually the best comparison to make. I learned: When benchmark testing, it is always important to have a reliable source with which to compare benchmark expectations. You need the research and science to help you determine student needs, and to determine the effect of core instruction and individual or small group interventions.
By the spring of my internship year I had learned a lot. I learned what we currently had access to and began to think of what would make things better. I started to dream of the day when we would have simpler administration, more efficient data analysis, and measures with references to good representative norms or expectations. Now looking back over the past 20+years, I can truly say that dreams do come true!