Understanding Benchmarks – The Mechanics

Definition

In its simplest definition, a benchmark is a standard against which to measure or compare an outcome. It is visually represented in Scorelink as a single horizontal column divided evenly into 4 colours (red, yellow, blue and green). When a benchmark is created for a test, it looks up the test score range and assigns the lowest possible test value at the start of the benchmark (beginning of the red band) and the highest possible test value at the end of the benchmark (end of the green band).

It then automatically divides the score range by 4 and assigns a quarter of the total value to each of the 4 colour segments.

So as an example, if a test had a top score of 100, the default benchmark would look like this:

Scorelink Colours

In Scorelink, colours always have the same meaning.

Knowing what the colours represent, means that we can assign numerical values for each colour to define risk, caution, benchmark and above benchmark for targeted groups of students. The default benchmark where each colour falls neatly into quarters is very rarely, if ever accurate, so therefore it is up to each individual site to determine what the number ranges for each colour should be.

In order to assign these, the user/administrator needs to know the test, and within it, what scores represent risk, excellence etc. so they can correctly apply them to the benchmark. This will only need to be done once and if done accurately, will provide valuable and clear information about a student’s performance at any point in their test history.

How Many Benchmarks do I need?

As few as possible. You only need a different benchmark for every time a different standard is required. For example, only Year 3, 5, 7 students do NAPLAN tests, and we do require different standards for each of those year levels so therefore we need to create 3 benchmarks – one for
Year 3, Year 5 and Year 7.

The question to ask is “Does Band 5 in NAPLAN Maths mean the same for a Year 3 student as it does for a Year 5 or Year 7 student?” If the answer is no, then you will need to create 3 different benchmarks to cater for that difference.

Let’s ask the same question for say AC English. “Does an “A” grade mean the same for a Year 3 student as it does for a Year 5 or Year 6 student?” In this case the answer is Yes because in the AC an “A” always means the student is achieving at maximum regardless of which year level they are in. Therefore, in this case we only need to create one benchmark and apply it to all students.

Sometimes we may require different standards at different times of the year. In Running Records for example, we may have differing expectations of our students each term. So, let’s say that we intend to test Running Records for Years R-3. Let’s ask our first question. “Does Reader 12 (for example) mean the same for a Reception as it does for a Year 1 or Year 2 student?” The answer is No so therefore we require 3 different benchmarks.
In addition we need to ask “Do we expect only one standard for the whole year or do we expect a different standard at different times of the year?”, and if it is the latter, we ask “Do the standards change every semester or every term?” Assuming its every term, we will need a total of twelve benchmarks:

Reception Term 1, Reception Term 2…………… Year 3 Term3, Year 3 Term 4.

Configuring Benchmarks

Once you have created the correct number of benchmarks you need to configure them (set the number scores for each colour range). To simplify the process, follow these steps.

1. 1. Click VIEW on the test to work with.
   2. Click on EDIT alongside the first benchmark to configure. This will display the Edit Benchmark screen. The name of the Benchmark will be displayed in the Name field. 
   3. With the exception of the first and last values (which are greyed out as they cannot be edited), every other value can be manually changed. However, we only need to change the values of three cells to complete the benchmark-setting task. These are the two blue values and the one red value.
   4. BLUE VALUES. Because the two blue values represent the minimum and maximum scores that a student can attain to be inside the benchmark, there are two questions one needs to ask to arrive at these values. (in this case we are setting the benchmark for Reception Semester 1)
      Q1: “What is the LOWEST score (reader) that a Reception student can attain by the end of Semester 1 but still be within the benchmark?”
      Let’s say it is Reader 4. We need to place the number 4.00 in the left hand blue cell.
      Q2: “What is the HIGHEST score that a Reception student can attain by the end of Semester 1 but still be within the benchmark?”
      Let’s say it is Reader 7. We need to place the number 7.99 in the right hand blue cell. All values placed in the right-hand cells should end with .99.
   5. RED VALUE. To get the value for the right hand RED cell the question to ask is
      Q3: “What is the HIGHEST score (reader) that a Reception student can attain by the end of Semester 1 but still be at risk?”
      Let’s say it was Reader 2. Place the value 2.99 in the RH RED cell.
      The rest of the values will now re-flow themselves around these three manual entries.
   
   6. Click “Save Changes” to save these settings.

Tagged: benchmarks