Use of stanines in the STAR Reading Test

STAR results are reported as scale scores. If you have used PATs, you will be familiar with how scale scores work. Scale scores are not dependent on a particular year group, test or time of year. Scores can be compared across classes, year level cohorts and across other demographics such as gender and ethnic group. We recommend using scale scores to monitor progress for both individuals and groups of students.

Stanines are useful when you want to get a sense of how a student's scale score compares with a nationally representative sample of students in a year level. This is sometimes called a normative comparison. When making a normative comparison, teachers need to select the stanines for the most appropriate year level reference sample and record which one they have used. This will depend on the time of year the student was tested. Because the reference data for STAR was collected in March, it is not always appropriate to use the reference sample for the same year group as the student.

 

At the start of the year (Term 1)

If students are tested at the start of the year, then stanines for their year level should be used. For example, if a Year 3 student is assessed in February, then teachers should record the appropriate stanine for the Year 3 reference sample and note that is a Year 3 stanine.

 

At the end of the year (Term 4)

When students sit a STAR test in Term 4 their scale scores should be compared with the reference sample for the next year level up. This is because the reference data refer to students who did the test in Term 1. By the end of the year, students are more like students who are just starting the next year level up than students who are just starting their own year level. So, if a Year 3 student is tested in Term 4, then stanines for the Year 4 reference group should be used. It is important to record that they are Year 4 stanines.

 

In the middle of the year

Making stanine comparisons in the middle of the year is less straightforward. This is because there is no obvious reference group. If you want to make a normative comparison in Term 2 or 3 it is probably best to record the stanine for the student's year level but to note that this will tend to overestimate how well they are doing compared to peers at the same stage of schooling. So, for instance, if a Year 3 student is tested in Term 2 then the Year 3 stanines should be used. Again, it is important to record that they are Year 3 stanines and to note they tend to make the students look a little better against their year level than if they were tested in Term 1.

 

Where can I find stanines for year levels not recorded in the Teacher Manual?

The Teacher Manual provides stanines for two or three year levels for each test form. Stanines for all year levels for each test form can be found on NZCER Assist. You will be able to download the relevant PDF document from the lefthand side of the screen under the heading "STAR 2nd edition - Stanine table for all years". Contact our Assessment Services team if you require assistance. 

 

What are the advantages and disadvantages of using stanines?

 

Advantages

  • Stanines let you compare your own students' results with a similar national representative groups of students.
  • Stanines form an "interval" variable which means that the amount of progress represented by going from the middle of one stanine band to the middle of the next stanine band is the same across all stanines.
  • Because they form an interval variable, stanines can be averaged. An average stanine for a class or a year level cohort is a meaningful variable. This is not true for percentiles.

Disadvantages

  • Stanines are always tied to a particular year level and time of year. If your students have not done their STAR tests at the same time of year as students who formed the nationally representative sample, it is much more difficult to use stanines meaningfully.
  • Stanines give us only nine bands with which to describe the full range of achievement within a year level. This means that stanines can only provide a rather coarse measure of achievement.