Computational Application for Early Detection of Dyslexia in Pre

Download Report

Transcript Computational Application for Early Detection of Dyslexia in Pre

Computational Application for
Early Detection of Dyslexia in
Pre-school Children
Background
 Aim of overall project
 Aim of current phase
 Phonological Deficit Hypothesis
• The most commonly used tests in Ireland at present:
•The Dyslexia Screening Test (6.6 to 16.5 years)
•The Bangor Dyslexia Test (test not normed)
•Genetic Factor (Rack, 1999)
Dyslexic
Father
NonDyslexic
Father
Dyslexic
Mother
NonDyslexic
Mother
Males
0.447
0.050
0.361
0.056
Females
0.206
0.012
0.207
0.015
Developing the test
 Early/Intermediate/Late Consonants (Smit
& Hand, 1997)
 38 test items based on 19 phonemes
 Selecting words and images
 Pre-pilot the test – initial idea of validity
and reliability
 Technical aspects
Test Administration
 8 pre-schools
 104 children (+ 20 at pre-pilot stage)
 Age range: 33 to 69 months
 Locations: rural Wicklow to urban City
Centre
 20 questions per child
 Mean score: 11.89; Mode: 11
Test Analysis
Histogram
20
Frequency
15
10
5
Mean =11.89
Std. Dev. =4.32
N =104
0
0
5
10
Total_1s
15
20
Point-Biserial Test
 An item analysis test
 Establishes a correlation between a
right/wrong score a child receives in a
given item and the total score that a child
gets in the overall test
 Ranges from +1 to –1
 Varma: Point Biserial of at least .15
Test Analysis contd.
 Low Point-Biserial Scores Removed:
Histogram
12
10
Frequency
8
6
4
2
Mean =9.5
Std. Dev. =4.063
N =104
0
0
5
10
Proper_total_correct
15
20
Cronbach’s α
 A split-half measure of the reliability of the
test items
•Cronbach's
Alpha
.830
•Number of
Items
17
Early/Intermediate/Late
Consonants
 Developmental Sequence as recognised by
Smit&Hand (1997)
 Early Consonants: Mean = 32
 Intermediate Consonants: Mean = 33.8
 Late Consonants: Mean = 32.58
 No significant difference between groups
Location
Mean
Std.
Deviation
N
Location
Location A
8.67
6
3.670
Location B
11.00
9
4.031
Location C
8.67
12
3.312
Location D
8.83
12
3.950
Location E
6.64
14
2.706
•Phonological training?
Location F
10.07
28
4.472
•Age variation?
Location G
13.07
14
2.999
Location H
7.67
9
3.317
Total
9.50
104
4.063
•Significant variation in scores
•Parental interest?
Age-Sensitive Test
Mean Score
Standard Deviation
Group 1
8.05
3.268
Group 2
11.90
3.919
Group 3
12.35
4.499
Group 4
13.30
3.975
Group 5
13.54
3.833
Achievement
 30 ‘good items’ in terms of reliability
 Discriminates between children on their
success at a phoneme-based test
 Age sensitive
 Correct pitch – format of test
Futher Development
 Further pilot-test
– More care in getting a representative sample
– Try on a group who are known to have
dyslexia
 Further statistical analysis
– Z-scores
– Multiple Regression