Skills Check-ins uses an Online Bayesian IRT model to diagnose students and calculate their proficiency across skills. It predicts the probability of a student getting a question correct based on their current mastery versus the difficulty of the question they are attempting. The engine uses this prediction to select the next best question to present to a student and to calculate proficiency.

In addition, the engine also uses a curriculum knowledge graph to maximise the efficiency of the diagnostic check-ins. This means that students only have to answer a small set of questions to allow Skills Check-ins to predict their proficiency across a large number of skills.

When calculating proficiency, the model looks at all of the questions that are related to a particular outcome, and the probability of the student getting each of those questions correct.

Skills Check-ins reports proficiency at each grade level. This proficiency is calculated as described above, based on the probability of a student getting all of the questions mapped into all of those grade standards correct. e.g. a student with a Grade 8 proficiency of 1 is a student that has a high probability of answering all easy, medium and hard questions correct in Grade 8 standards.

Skills Check-ins also reports an overall grade level. This level is a very simple calculation that sums up the proficiencies for all grades in the curriculum. e.g. a student with a Grade 3 proficiency of 1 and Grade 4 proficiency of 0.8 would be placed at Grade Level 3.8.

Growth is calculated as the grade level at the end of the timeframe minus the grade level at the start of the timeframe, e.g. if a student had a grade level of 3.1 in September and a grade level of 3.4 in November, then the growth is 0.3 grade levels.

Growth estimates compare the grade level that a student had at the start of the period, and at the end of the period. For example, if at the start of the period, the student’s grade level was 7.1 and at the end of the period it was 7.2, then the student’s growth is 0.1.

Teachers can use the Growth report to identify the grade level growth for every student over a period of time, and their projected growth rate. e.g. if a student had a grade level of 3.1 in September and a grade level of 3.4 in November, then the growth is 0.3 grade levels. Given the student has grown 0.3 grade level in 3 months, their projected growth over the next 10 months is 1 full grade level.

Can a student show negative growth?

Yes, a student may also show negative growth if proficiency drops on any skill that they had previously demonstrated. This means the grade level at the end of the period is lower than at the start of the period. e.g. if a student had a grade level of 7.5 in September and a grade level of 7.3 in November, then the growth is -0.2 grade levels.

The expected growth for one year of school is one grade level. So the expected growth on Skills Check-ins is 1.

The aim of Skills Check-ins is to determine the student’s frontier of knowledge. As students might be at different levels, during the first diagnostic check-in we test all years below and one above, in order to catch students that might be ahead. The grade cap is removed for growth check-ins so that we can maximise growth captured, so students will occasionally get a question from multiple grade levels above. Students should skip any questions that they don't know.

If a student answers a question incorrectly that they had previously answered correctly, their green ticks/ checks drop to yellow or empty boxes in the skills report, as it may indicate they are no longer proficient at that skill and would benefit from practice. The green tick/ check will reappear once a student answers the question again in a subsequent check-in.