Monitoring Achievement and Progress in Independent Schools

1
Monitoring Achievement and Progress in
Independent Schools
Peter Hendry CEM Consultant

• Running Your School on Data
• January 2011

Peter.Hendry_at_cem.dur.ac.uk
2
Running your school on data

• CEM data includes
• Baseline test acquired ability data (IPRs)
• Predictive data including chances graphs
• Value-added data
• Attitudinal data
• Curriculum assessments (Insight)
• PARIS software programmes

3
CEM data is used for, e.g.

• Teachers to help learners
• Curriculum and staffing decisions
• Target setting and monitoring pupil progress
• Inspection evidence
• Self-evaluation
• Monitoring changes over time such as pupil
  ability intake profiles and VAD
• Asking the question can we do better?
• i.e. the data is used to aid and to support
  professional judgement

4
with due consideration to

• Ethos and tradition of my school
• Accountability
• Parental expectations
• Staff training in use of data and ability to cope
  with data (data overload)
• Integrating the data into school procedures,
  storage, retrieval, distribution and access
  (..........policy?)
• i.e. doing our best to help every pupil to at
  least achieve, if not exceed, their potential

5
So much data................................!
Some key questions

• Which data do I need AND which data do I not
  need?
• What does the data mean and what doesn't the data
  mean?
• Who is the data for? Who does not need to have
  some of the data?

6
1. What data do I need? A key first step!

• e.g. using MidYIS GCSE predictive data for target
  setting
• Based on whole MidYIS cohort?
• Based on the Independent Sector?
• Based on the schools prior value-added?
• Point and/or grade predictions? (e.g. 6.2/grade
  B)
• The chances graphs?

7
subject predictions all MidYIS cohort
6.4
6.2
5.7
MidYIS baseline score120
8
subject predictions Independent Sector
7.1
7.0
6.4
MidYIS baseline score 104
9
Comparing MidYIS predictions
10
Prediction 6.3 A/B
11
Prediction 6.9 A (Independent Sector)
12
Prediction 7.4 A/A
13
Discussion setting the targets...........

• Which type of predictive data would you use to
  set the targets, and why?
• Would your students be involved as part of the
  target setting process?
• Would parents be informed about the process and
  outcome?
• How would you ensure that HoDs were involved to
  ensure overview the process?

14
2. What does the data mean?

• e.g. value-added data
• The difference between raw and standardised
  residuals
• The use of confidence limits to distinguish
  between average and statistically significant
  data, and to allow for small entry subjects
• Can zero or negative value-added be acceptable?

15
Raw residual Bar chart (MIDYIS and YELLIS only)
16
Standardised Residual Bar Chart
99.7 confidence limit
95 confidence limit
ANY VALUE IN THE INNER SHADED AREA IS WITHIN THE
EXPECTED RANGE AND IS THEREFORE CONSIDERED TO BE
AVERAGE
17
Raw residual Bar chart (MIDYIS and YELLIS only)
18
Standardised Residual Bar Chart
99.7 confidence limit
95 confidence limit
ANY VALUE IN THE INNER SHADED AREA IS CONSIDERED
TO BE AVERAGE VALUE ADDED
19
Standardised Residual Bar Chart
20
3. CEM data who is it for? For example

• SMT/SLT
• HoDs
• Subject teachers
• Form/House Tutors
• Head of Learning/Year
• Parents
• Pupils
• Governors

21
Case study Alis value-added data

• Four sets of VAD are available!
• From average GCSE baseline
• all Alis cohort
• type of Institution (Independent Sector)
• syllabus
• From the baseline test
• all Alis cohort

22
School SPC Chart with confidence limits Baseline
Av. GCSE
Institution
All Alis Cohort
Syllabus
23
Using PARIS software CABT Baseline
Whole School
From your perspective, which set of VAD would you
use for the different user groups? (Governors,
HoDs, Parents, SMT/SLT...)
24
A key question is therefore....... Is it
possible to keep adding value at each key stage?
Typical Av. GCSE Subject Raw Residuals
25
Typical Av. GCSE Subject Standardised Residuals
A2 Intake Profiles typical average GCSEs for
students in each subject
26
2010 A2
2009 A2
27
2009
2010
28

• USE ONE YEARS DATA WITH CAUTION!
• Better to use three years data as patterns over
  time are more significant.