The Primary Mathematics Research Project: Towards Evidence-Based Educational Development in South Africa
The Primary Mathematics Research Project: Towards Evidence-Based Educational Development in South Africa
The Primary Mathematics Research Project: Towards Evidence-Based Educational Development in South Africa
RESEARCH PROJECT
TOWARDS EVIDENCE-BASED
EDUCATIONAL DEVELOPMENT
IN SOUTH AFRICA
Inability to calculate
Phase I: Lesson Observation
‘If you are asked 15 times four its meaning is that you are
counting 15 four times. To do this you must first expand the
sum.’
Writes on board: 15x4 = (10x4)+(5x4) = (10+10+10+10) +
(5+5+5+5) = (1111111111+1111111111 –four times) +
(11111+11111 –four times) Counts units = (40)+(20) ‘When
you add 40 and 20 what do you get?’ First group to answer
60 is asked to show workings on board: pupil comes to
board and writes (1111111111 – four times) + (111111111
– twice). Counts individual units and writes = 60
Each group is given a sum. Workings of first group to
report: 23x7 = (20x7)+(3x7) = (11111111111111111111 +
1111111111111111111 –seven times) + (111+ 111 –seven
times). Counts units = 160 [correct answer is 161]
Phase I: Conclusions
The fundamental causes of poor performance are:
a failure to extend the ability of learners from counting to
true calculating. All more complex mathematics depends on
an understanding of place value within the base-10 number
system, the ability to readily perform basic calculations and
see numeric relationships.
this problem is caused by the application of ineffective
learning practices in classrooms resulting in the virtual
disappearance of memorization, consistent drill and regular
extensive practice of learned content
learners are not being given the opportunity to develop
the neural pathways and structures required for the
development of higher order cognitive competencies in
mathematics
Phase I: Conclusions
Closely associated with these causes has been the virtual
abolition of the concept of a national or provincial syllabus
of study combined with textbooks designed to give effect to
this syllabus.
ge
d
le
ra
ra
ol
ve
ve
tr
co
co
on
%
nc
U
50
80
Learner Performance on NSE Items: Controlled =
80% (11 weeks) of coverage from here on
Grade 4 (12 items) Grade 6 (16 items)
Uncontrolled Controlled Uncontrolled Controlled
Project
Baseline 23.17 21.58 20.19 18.50
Post Test 30.00 36.83 25.38 32.12
Change +6.83 +15.25 +5.19 +13.62
% increase on baseline +29.50 +70.67 +25.70 +73.62
Control
Baseline 23.42 23.42 17.75 17.75
Post Test 17.92 17.92 19.75 19.75
Change -5.50 -5.50 +2.0 +2.0
% increase on baseline -23.48 -23.48 +11.27 +11.27
Impact: % points +12.33 +20.75 +3.19 +11.62
Impact: % increase +52.98 +94.15 +14.43 +62.35
Learner performance in the Four Operations
Grade 4 Grade 6
Uncontrolled Controlled Uncontrolled Controlled
Add
% points +7.2 +9.4 +18.2 +23.4
% of baseline +16.43 +20.0 +42.18 +54.86
Subtract
% points +9.00 +11.2 +17.4 +25.0
% of baseline +32.90 +34.66 +87.91 +121.32
Multiply
% points +9.0 +13.0 +11.8 +18.6
% of baseline +117.09 +184.12 +378.57 +814.29
Divide
% points +12.6 +16.4 +10.2 +13.8
% of baseline +92.24 +104.94 +128.15 +156.01
Grade 6: Performance against Assessment
Standards for Grades 4 and 5: Post Test Only
Uncontrolled Controlled
Project 38.5 45.1
Positive
27.6
Impact
45.7 No Impact
Negative
26.8 Impact
PMRP: Distribution of Impact: Schools
Both graphics are based on data uncontrolled for coverage
Grade 4 Grade 6
5 0 5
5
Positive
impact
No impact
Negative
impact
90
95
Once the data are controlled for coverage, all 100% of the
schools recorded significant impact
PMRP: Distribution of Impact: Individual
Learners
In these graphics, scores are drawn from the combined
group of Grade 4 and 6 learners.
76.9
89.2
Frequency of Problem-Solving Methods:
% of Learners
Grade 4 Grade 6
70 70
60 60
50 50
% o f le a r n e r s
% o f lea rn ers
40 40
30 30
20 20
10
10
0
0
Counting Counting >50% Numbers>50% Calculations
Counting Counting >50% Numbers>50% Calculations
Project Type of method
Type of method
Control
Frequency of Problem-Solving Methods:
% of all Attempts
Grade 4 Grade 6
90
90
80 80
70 70
% o f a ll w o rk in g s
% o f a ll w o r k in g s
60 60
50 50
40 40
30 30
20 20
10 10
0 0
Unit counting Rep. Operations Calculations
Unit counting Rep. Operations Calculations
Project
Type of method
Control Type of method
Problem-Solving: Scans from Learner Scripts
These examples are both drawn from Grade 6 scripts.
Control Project
Problem-Solving: Scans from Learner Scripts
The scans on the left typify methods used in control
schools: 4x7 and 5x67. The scan on the right is from
a project school: 856x45 and 8 681x37.
Problem-Solving: Scans from Learner Scripts
Learners from project schools are able to correctly answer
questions from Part 2 (ESA) and Part 1 (NSE), respectively
Assessment: Scans from Learner Scripts
An illustration of the enormous differences in ability in
Grade 6 learners; both were passed as competent at the end
of Grade 5.
Scans from Teacher research Diaries
Teachers were asked to keep a research diary as they worked
through the PMRP programme
Scans from Teacher research Diaries
This practice helped convince teachers they were part of the
research team and provided useful formative information
Photographs from the Field
A learner practices the solving of many division problems
in one lesson using whole number calculations
Photographs from the Field
Direct instruction: the teacher explicates place value
Photographs from the Field
Group work: winner of mental arithmetic game
Working as individuals
Photographs from the Field
Working in pairs