EGRA SENEGAL

1
EGRA SENEGAL
Senegalese Primary School Students Learning to
Read in French and in Wolof
Momar Samb, National Institute for Education
Development, Senegal, Liliane Sprenger-Charolles,
CNRS Paris Descartes, France

• Early Grade Reading AssessmentSecond Workshop,
  March 12-14, 2008Washington, DC

2
A. METHOD

• With Senegalese and Gambian officials, during a
  workshop organized in Dakar by RTI and the World
  Bank (April 2007)
• the initial EGRA protocol was modified and
  adapted in French and English
• the French version was translated into Wolof
• these new versions of the EGRA protocol were
  pretested in several schools
• After the pretesting process, the EGRA protocol
  was once again modified
• Tasks used in Senegal (and Gambia)
• 1 reading and 1 listening comprehension task
• 4 tasks that assess accuracy and speed in reading
  (1 minute tests)
• letters
• isolated words
• words in context
• invented words (pseudowords)
• 1 spelling task, including an assessment of word
  spelling
• 2 phonemic awareness tasks phoneme
  identification and counting
• Surveys
• Questions to students about their cultural and
  linguistic environment, and socioeconomic status
  (SES).
• Questions to teachers about their practices of
  instruction (results not analyzed)

3

• Participants Senegalese students learning to
  read in
• French 502 children (1st, 2nd and 3rd graders)
• Wolof 186 children (only 1st and 3rd graders)
• Two main analyses were performed
• To assess the significance of the differences due
  to the main factors assumed to have an incidence
  on reading acquisition (e.g., socio-economic
  status, SES) we have compared (ANOVAs and
  t-tests) the results of
• children from the lower SES category versus those
  from the higher category
• boys versus girls,
• children speaking only Wolof at home versus those
  speaking also French,
• children with at least one literate parent versus
  those with no literate parent,
• children learning to read in their mother tongue
  (Wolof) versus in French,
• and those learning to read in French versus in
  English (Gambian study).
• To determine the main predictors of reading
  level, regression analyses were carried out.

4
B. RESULTS

• 1. Percentage of first graders with scores at the
  floor level ( 0)

• Very high percentage of first graders with scores
  at the floor level
• for the word reading and spelling tasks,
• especially for children who were learning to read
  in Wolof
• For the listening comprehension task,
• especially for children who were learning to read
  in French

5

• A problem for the reading comprehension task 5
  questions on the text used to assess
  word-in-context reading.
• A child had to read at least 20 words of this
  text to obtain a reading comprehension score of
  1.
• About 50 of the Senegalese children were unable
  to correctly read more than 5 words of this text
  in 1 minute.
• Some of these children achieved a reading
  comprehension score between 1 and 5 (5 being the
  maximum).
• Problem due to the procedure used for the
  word-in-context task
• The students were allowed to read the text for at
  least 2 minutes, but only words correctly read in
  1 minute were coded
• Not possible to determine the total number of
  words read correctly.
• When the child was unable to read a word, the
  examiner provided it
• In cases where the examiner provided a large
  number of words, the task became a mixture of
  reading and listening comprehension.
• Children can have guessed on answers, especially
  for yes/no questions
• Therefore, for the regression analyses, we have
  only examined the reading comprehension scores
  for children who were able to read at least 20
  words in 1 minute (around 30 in each group)
• However, for comparisons within and between the
  groups, we have taken into account the original
  data
• these results, thus, should be considered with
  caution

6
2. Effects of SES, Gender, Parents literacy
status

• SES Significant only among children learning to
  read in Wolof
• GENDER Significant only among children learning
  to read in French
• Girls outperformed boys in word reading and
  spelling
• PARENTS LITERACY STATUS
• Significant only among the children learning to
  read in French
• HOME LANGUAGE
• Children learning to read in French and speaking
  at home only Wolof or French Wolof
• Significant differences, except for spelling.

7
3a. Effect of the language in which children are
learning to Read Wolof vs French (Grades 1 and 3)
mean and Standard Deviation

• No differences between the groups in terms of age
  and SES.
• Children learning to read in Wolof scored higher
  for two tasks that assessed
• spoken language (listening comprehension and
  phoneme counting)
• Children learning to read in French scored higher
  
• for two tasks that assessed written language
  (orthographic skills).

8
3b. Effect of the language in which children are
learning to Read English vs French (Grades 1 to
3)
mean and Standard Deviation

• The SES of Gambian children was lower than that
  of Senegalese children.
• Even after the effect of SES was taken into
  account, the scores of the Senegalese
• children were higher than those of the Gambian
  children
• Except for the listening comprehension task.

9
4. Prediction of Isolated Word and Word in
context reading and reading comprehension
Cells in grey Variables not taken into account
10
C. SUMMARY

• 1. Predictors of reading skills

• Isolated-word and word-in-context reading for
  both groups
• most of the variance in these two tasks was
  explained by the variables entered in the model,
• and the sole unique predictor was pseudoword
  reading,
• not phonemic awareness, nor letter knowledge,
  WHY?
• Significance of decoding skills for reading
  acquisition
• Reading comprehension for both groups
• only a small part of the variance was explained
  by the variables entered in the model,
• and the sole unique predictor was listening
  comprehension,
• not word reading, except word-in-context reading
  for children learning to read in Wolof.
• This result might be due to the fact that only
  children able to read at least 20 words of the
  text in 1 min. were included in the analyses
  (30)
• The reading comprehension task should be modified

11
2a. Effect of the language in which the children
are learning to read English versus French

• For the reading tasks, the Senegalese children
  surpassed the Gambian children.
• This result confirms those reported in the
  literature,
• indicating that the degree of consistency of
  Grapheme-phoneme correspondences affects the
  acquisition of reading skills
• see Seymour et al., 2003 Sprenger-Charolles,
  2003 Sprenger-Charolles et al., 2006 Ziegler
  Goswami, 2005.

12
2b. Wolof versus French

• In some tasks involving written language,
  children learning to read in French surpassed
  those learning to read in Wolof.
• This result may be due to the fact that written
  Wolof is not as developed as written French.
  Children learning to read in Wolof were thus less
  exposed to written materials in this language
  than those learning to read in French.
• This interpretation is reinforced by the fact
  that
• there were more floor effects in reading-spelling
  tasks among children learning to read in Wolof
  than among those learning to read in French
• there were no difference in reading-spelling
  scores due to the parents literacy status among
  children learning to read in Wolof,
• while for children learning to read in French,
  those with at least one literate parent scored
  higher in almost all reading-spelling task

13
D. Implications for educational policy

• This study highlights the fact that reading
  acquisition depends on the degree of
  Grapheme-Phoneme Correspondence (GPC)
  consistency.
• Given the importance of mastering GPC to learn to
  read,
• teachers must be aware of the GPC rules in the
  language in which they are teaching children how
  to read.
• To master GPC, children must be able to
  discriminate the phonemes of the language in
  which they learn to read,
• Thus, teachers must know the phonological system
• of the language in which they are teaching
• and that of the childrens mother tongue.
• This knowledge might help them understand the
  possible interferences between phonological
  systems that may impede
• not only the acquisition of a new spoken
  language,
• but also reading acquisition in that language.
• At least, the main characteristics of the
  phonological and orthographic systems of the
  languages concerned must be provided to the local
  teams

14

• This study also highlights the difficulties
  related to establishing educational policy.
• The fact that children learning to read in their
  mother tongue (Wolof) achieved better results
  than those learning to read in French
• only for tasks involving spoken language
• not for tasks involving written language
• might be due to the fact that written Wolof is
  not as developed as written French.
• Thus, two policy options are possible
• either to develop the writing culture in Wolof 
• Or to learn to read in a language that is not the
  mother tongue of the children
• but for which written materials are numerous,
  diverse, and accessible.

15
E. Implications for future EGRA applications

• see for some suggestions, the reports for the
  World Bank on
• Senegalese results (Report in English and in
  French)
• Sprenger-Charolles, L. (2008). Results from
  Senegalese Primary School Students Learning to
  Read in French and in WolofReport for the World
  Bank. Web site www.eddataglobal.org
• Sprenger-Charolles, L. (2008). Résultats délèves
  sénégalais des trois premiers grades ayant appris
  à lire en français et en wolofrapport pour la
  Banque mondiale. Web site www.eddataglobal.org
• Gambian results
• Sprenger-Charolles, L. (2008). Early grade
  reading assessments (EGRA) Results of 1200
  Gambian children learning to read in English
  (Report for the World Bank). Web site
  www.eddataglobal.org