Republic of Zambia Overview Republic of Ghana - Global Reading ...
Assessing the Impact of Language on Pupil Learning Utilizing GIS data to map ethnologue with school language of instruction to further investigate the impact of language on reading skills acquisition Jennifer Pressley, Sarrynna Sou, and Susan Edwards • RTI International, Research Triangle Park, NC Overview Languages Spoken
Having a strong reading background in one’s native language is associated with being able to develop strong vocabulary knowledge, listening comprehension, and syntactic skills in a second language (August & Shanahan, 2008). In other words, pupils who do not know how to read in their native tongue may not be able to learn how to read in a second language.
Languages of Instruction
Findings Pupils who had a reading textbook in the LOI read 4.7 words per minute more (+4.7 wpm).
Mambwe-Lungu Nyamwanga Lambya
Pupils who were taught in their native tongue read +1.7 wpm than those pupils who were not taught in their native tongue, when controlling for LOI.
8.4 wpm
Pupils not taught in their native tongue were 1.7 times more likely to not be able to read any word in a connected passage, compared with students taught in their native tongue.
Bemba
Lunda
Pupils who were taught in their native tongue read +0.73 wpm than those pupils who were not taught in their native tongue.
2.7 wpm Nyanja
Kaonde
Luvale Lozi
Nkoya
27.3% of pupils had read books on their own during school on the 65.02% of pupils who were not taught in their native tongue
Yauma
scored 0 for ORF.
Shona
L usak a
4 wpm Bemba
Kunda
Kaonde
8.9 wpm
Shona
Lozi
4.5 wpm 7.9 wpm
Nsenga Kaonde
previous school day.
In exploring the concept of language of instruction (LOI) and its impact on a pupil’s ability to read, we examine key differences in pupils who learn to read in their native tongue as compared with pupils who do not learn to read in their native tongue. Ghana and Zambia are two examples of countries whose schools use several LOIs. Because their LOI policies differ substantially, these countries’ data lend themselves to a broader examination of the topic. Zambia—a country with 7 officially recognized languages, but an official count of 46 different languages spoken by the population (Lewis, Simons, & Fennig [Eds.], 2016)—declares a school’s LOI based on province. Ghana boasts 81 different living languages within its ethnologue (Lewis, Simons, & Fennig [Eds.], 2016), but only 11 are recognized as national languages. However, Ghana’s policy allows LOI to be chosen by the head teacher on a school-by-school basis. This policy, in theory, allows head teachers to better cater to their specific population of pupils. These vastly different LOI policies provide a rich setting to understand the impact of LOI and how it relates to reading skills acquisition among Grade 2 pupils.
Nyiha
Tumbuka
Tonga
Lozi Lunda
2 wpm
Luvale Nyanja
Kuhane Tonga
Tonga
Lozi
58.27% of pupils who were taught in their native tongue scored 0 for ORF.
14.7% of pupils had reading textbooks.
Zambian Language Oral Reading Fluency (ORF)
40
7.9 wpm
30 20
LOI-Native Tongue No Match
10 93
94
95 Percent
96
Sisaala, Tumulung
Dagaare, Southern Paasaal
97
98
99
Buli
Paasaal
Konni
Tampulma
Chakali
Birifor, Southern
Konkomba
Deg
Gonja
Deg
Chala
Chumburung Kplang
4.1wpm
4.3 wpm
Population Density
Pupils who read aloud to someone at home every day read +5.7 wpm (compared with pupils who did not read aloud at home).
Adele Chala
Krache Chumburung
Dwang
Delo
Lelemi Siwu
Nkonya Selee
Nyangbo
Akan
Akan
Wasa Awutu
Logba
Kumasi
Ewe
Adangbe
8.6 wpm
Dangme
36.9% of pupils had reading textbooks.
4.8 wpm
32.9% of pupils had books at home.
5.5 wpm
Ga
Accra
Awutu
Nzema
Pupils who were read aloud to by someone at home every day read +1.7 wpm (compared with pupils who were not read aloud to at home).
Ho
Tafi Gua
Cherepon Larteh
3.4 wpm
Sekpele
Avatime
Dangme
Sehwi
Pupils who read aloud to someone at home only sometimes read +2.4 wpm (compared with pupils who did not read aloud at home).
Akposo
Akan
JwiraPepesa
2.5 wpm
51.3% of pupils spoke the LOI as their primary home language.
Cape Coast
29.5% of pupils were NEVER read aloud to at home.
Ahanta
41.5% of pupils had NEVER read aloud to someone at home.
Language Spoken Matches Language of Instruction
Percentile Plot
1.9 wpm Akan Dagaare, Southern
2.3 wpm
Dagbani Dangme Ewe
1.9 wpm
Ga
Tamale
Gonja Kasem
Ghana Education Service, National Education Assessment Unit and RTI International. (2014, draft). Ghana 2013—Early Grade Reading Assessment and Early Grade Mathematics Assessment: Report of Findings. Prepared under the USAID Education Data for Decision Making (EdData II) project, USAID/Ghana Partnership for Education: Testing, Task Order No. AID-641-BC-13-00001 (RTI Task 21). Research Triangle Park, NC: RTI International.
3.5 wpm
Other
Gikyode
Chumburung
August, D., & Shanahan, T. (Eds.). (2008). Developing reading and writing in second-language learners: Lessons from the report of the National Literacy Panel on Language-Minority Children and Youth. New York, NY: Taylor & Francis.
8 wpm
100
Nzema
Nawuri
Gonja
Abron
Pupils who had a reading exercise book in the LOI read +2.5 wpm.
Kasem
4.7 wpm
Tuwuli
Anyin
Gonja
Konkomba
Gonja
Kamara
Pupils who had a reading textbook in the LOI read +3.3 wpm.
Ga
Ntcham
Gonja Gonja
Ewe
Tamale
Kamara
Pupils who reported having books at home read +3.1 wpm.
Dangme
1.8 wpm
Dagbani
Hanga
Vagla
Kulango, Bondoukou
Dagbani
Kantosi
Wali
Safaliba
Dagaare, Southern
1.7 wpm
Anufo
Mampruli
Pupils who spoke the LOI as their primary home language read +1.1 wpm.
Akan
Bimoba
Tampulma
Kantosi
Kantosi
Findings
1.5 wpm
Kusaal
Farefare
Kantosi
References
LOI-Native Tongue Match
92
The data used for Zambia come from the 2014 National Assessment Survey of Learning Achievement at Grade 2, which was intended to provide a measure of pupils’ reading ability. The sample design used was probability proportional to size, with oversampling for subpopulations that constitute a very small proportion of the population.
Overall, pupils who are taught to read in their primary home language have a higher oral reading fluency (ORF) score than those who are not, but the differences are not substantial at this time. It is believed that numerous interfering factors, such as low distribution of textbooks and low parental involvement, have sufficiently confounded results. When these issues are addressed, it may be possible to show that reading skills acquisition occurs more easily and quickly for pupils who are taught in their primary home language. We believe that using GIS data to compare ethnologue maps and LOI is a meaningful way to begin a data-rich discussion around LOI policy and pupil performance.
ORF Score Based on 90–99th Percentile
91
Kasem
Conclusions
50
90
For this study, Early Grade Reading Assessment data were analyzed and combined with geographic information system (GIS) data detailing the languages spoken throughout the country and the LOI employed in schools. The combination adds background and provides a rich visual component for the discussion around LOI policy.
Language Spoken Matches Language of Instruction
60
0
Methodology
Ghana. The models used to estimate the effects of different variables were based on socioeconomic status, location of school in an urban area, and pupil’s status as overage for grade level. All results were weighted.
native tongue.
Sisaala, Western
Languages of Instruction
Bissa
Ligbi
Zambia. The models used to estimate the effects of different variables were based on socioeconomic status, school type, and province. The effects of gender, age, and location of school in an urban area were not found to be significant in this case; therefore, these variables were dropped. All results were weighted.
14.2% of pupils were taught in a language that was not their
Languages Spoken
Nafaanra
The data used for Ghana come from the 2013 Ghana Early Grade Reading Assessment (EGRA) study. The sample was selected through a two-stage sampling process, with proportional selection by region and gender.
The proportion of pupils taught in their native tongue and able to read at least 1 wpm was significantly larger at the 0.01 level, compared with pupils not taught in their native tongue.
70
Republic of Ghana
Ghanaian Language Oral Reading Fluency (ORF)
Republic of Zambia
Population Density
Nzema
4.4 wpm
Population per Square Kilometer 0
RTI International. (2015). National Assessment Survey of Learning Achievement at Grade 2: Results for Early Grade Reading and Mathematics in Zambia. Report prepared under the USAID EdData II project, Data Collection Services for the USAID/ Zambia Education Project, Task Order No. AID-611-M-14-00002 (RTI Task 28). Research Triangle Park, NC: RTI International.
10 - 40
1-2 3-9 41 - 150
!(
151 - 600
Tamale
601 - 2,500 2,501 - 10,500 10,501 - 45,937
Acknowledgment Bemba Kaonde
8.4 wpm
Lozi Lunda
1.9 wpm Lusaka
1-2 3-9 10 - 40 41 - 150 151 - 600 601 - 2,500
The Examinations Council of Zambia of the Ministry of Education, Science, Vocational Training and Early Education
3.6 wpm Kumasi
Mismatched Areas
Nyanja Tonga
Sarrynna Sou +1.919.485.2722 [email protected]
RTI International 3040 E. Cornwallis Road Research Triangle Park, NC 27709
!(
!(
Ho
7.4 wpm
Presented at: CIES 2016 Conference Vancouver, BC, Canada
Accra
2.8 wpm
2,501 - 10,500 10,501 - 43,059
www.rti.org
Kumasi
6.5 wpm
More Information Jennifer Pressley +1.919.541.6898 [email protected]
Ho
8.7 wpm
Luvale
Population per Square Kilometer 0
The National Education Assessment Unit of the Ghana Education Service
RTI International is a registered trademark and a trade name of Research Triangle Institute.
Accra
Cape Coast
60
Primary Home Language Not Primary Home Language
40
20
0 0
Lewis, M. P., Simons, G. F., & Fennig, C. D. (Eds.). (2016). Ethnologue: Languages of the World, Nineteenth Edition. Dallas, TX: SIL International. Available at: http://www.ethnologue.com
This poster is made possible by the generous support of the American people through USAID, under the EdData II funding mechanism: the USAID/ Ghana Partnership for Education: Testing, Task Order No. AID-641-BC-13-00001, and the Data Collection Services for USAID/Zambia Education Project, Task Order No. AID-611-M-14-00002. The views expressed by the authors at RTI International do not necessarily reflect the views of USAID, the United States Government, or the governments of the Republic of Zambia and the Republic of Ghana.
ORF Performance Percentiles by Language of Instruction as Primary Home Language
Cape Coast
10
20
30
40
50 60 Percentiles
70
80
90
100
Having a strong reading background in one’s native language is associated with being able to develop strong vocabulary knowledge, listening comprehension, and syntactic skills in a second language (August & Shanahan, 2008). In other words, pupils who do not know how to read in their native tongue may not be able to learn how to read in a second language.
Languages of Instruction
Findings Pupils who had a reading textbook in the LOI read 4.7 words per minute more (+4.7 wpm).
Mambwe-Lungu Nyamwanga Lambya
Pupils who were taught in their native tongue read +1.7 wpm than those pupils who were not taught in their native tongue, when controlling for LOI.
8.4 wpm
Pupils not taught in their native tongue were 1.7 times more likely to not be able to read any word in a connected passage, compared with students taught in their native tongue.
Bemba
Lunda
Pupils who were taught in their native tongue read +0.73 wpm than those pupils who were not taught in their native tongue.
2.7 wpm Nyanja
Kaonde
Luvale Lozi
Nkoya
27.3% of pupils had read books on their own during school on the 65.02% of pupils who were not taught in their native tongue
Yauma
scored 0 for ORF.
Shona
L usak a
4 wpm Bemba
Kunda
Kaonde
8.9 wpm
Shona
Lozi
4.5 wpm 7.9 wpm
Nsenga Kaonde
previous school day.
In exploring the concept of language of instruction (LOI) and its impact on a pupil’s ability to read, we examine key differences in pupils who learn to read in their native tongue as compared with pupils who do not learn to read in their native tongue. Ghana and Zambia are two examples of countries whose schools use several LOIs. Because their LOI policies differ substantially, these countries’ data lend themselves to a broader examination of the topic. Zambia—a country with 7 officially recognized languages, but an official count of 46 different languages spoken by the population (Lewis, Simons, & Fennig [Eds.], 2016)—declares a school’s LOI based on province. Ghana boasts 81 different living languages within its ethnologue (Lewis, Simons, & Fennig [Eds.], 2016), but only 11 are recognized as national languages. However, Ghana’s policy allows LOI to be chosen by the head teacher on a school-by-school basis. This policy, in theory, allows head teachers to better cater to their specific population of pupils. These vastly different LOI policies provide a rich setting to understand the impact of LOI and how it relates to reading skills acquisition among Grade 2 pupils.
Nyiha
Tumbuka
Tonga
Lozi Lunda
2 wpm
Luvale Nyanja
Kuhane Tonga
Tonga
Lozi
58.27% of pupils who were taught in their native tongue scored 0 for ORF.
14.7% of pupils had reading textbooks.
Zambian Language Oral Reading Fluency (ORF)
40
7.9 wpm
30 20
LOI-Native Tongue No Match
10 93
94
95 Percent
96
Sisaala, Tumulung
Dagaare, Southern Paasaal
97
98
99
Buli
Paasaal
Konni
Tampulma
Chakali
Birifor, Southern
Konkomba
Deg
Gonja
Deg
Chala
Chumburung Kplang
4.1wpm
4.3 wpm
Population Density
Pupils who read aloud to someone at home every day read +5.7 wpm (compared with pupils who did not read aloud at home).
Adele Chala
Krache Chumburung
Dwang
Delo
Lelemi Siwu
Nkonya Selee
Nyangbo
Akan
Akan
Wasa Awutu
Logba
Kumasi
Ewe
Adangbe
8.6 wpm
Dangme
36.9% of pupils had reading textbooks.
4.8 wpm
32.9% of pupils had books at home.
5.5 wpm
Ga
Accra
Awutu
Nzema
Pupils who were read aloud to by someone at home every day read +1.7 wpm (compared with pupils who were not read aloud to at home).
Ho
Tafi Gua
Cherepon Larteh
3.4 wpm
Sekpele
Avatime
Dangme
Sehwi
Pupils who read aloud to someone at home only sometimes read +2.4 wpm (compared with pupils who did not read aloud at home).
Akposo
Akan
JwiraPepesa
2.5 wpm
51.3% of pupils spoke the LOI as their primary home language.
Cape Coast
29.5% of pupils were NEVER read aloud to at home.
Ahanta
41.5% of pupils had NEVER read aloud to someone at home.
Language Spoken Matches Language of Instruction
Percentile Plot
1.9 wpm Akan Dagaare, Southern
2.3 wpm
Dagbani Dangme Ewe
1.9 wpm
Ga
Tamale
Gonja Kasem
Ghana Education Service, National Education Assessment Unit and RTI International. (2014, draft). Ghana 2013—Early Grade Reading Assessment and Early Grade Mathematics Assessment: Report of Findings. Prepared under the USAID Education Data for Decision Making (EdData II) project, USAID/Ghana Partnership for Education: Testing, Task Order No. AID-641-BC-13-00001 (RTI Task 21). Research Triangle Park, NC: RTI International.
3.5 wpm
Other
Gikyode
Chumburung
August, D., & Shanahan, T. (Eds.). (2008). Developing reading and writing in second-language learners: Lessons from the report of the National Literacy Panel on Language-Minority Children and Youth. New York, NY: Taylor & Francis.
8 wpm
100
Nzema
Nawuri
Gonja
Abron
Pupils who had a reading exercise book in the LOI read +2.5 wpm.
Kasem
4.7 wpm
Tuwuli
Anyin
Gonja
Konkomba
Gonja
Kamara
Pupils who had a reading textbook in the LOI read +3.3 wpm.
Ga
Ntcham
Gonja Gonja
Ewe
Tamale
Kamara
Pupils who reported having books at home read +3.1 wpm.
Dangme
1.8 wpm
Dagbani
Hanga
Vagla
Kulango, Bondoukou
Dagbani
Kantosi
Wali
Safaliba
Dagaare, Southern
1.7 wpm
Anufo
Mampruli
Pupils who spoke the LOI as their primary home language read +1.1 wpm.
Akan
Bimoba
Tampulma
Kantosi
Kantosi
Findings
1.5 wpm
Kusaal
Farefare
Kantosi
References
LOI-Native Tongue Match
92
The data used for Zambia come from the 2014 National Assessment Survey of Learning Achievement at Grade 2, which was intended to provide a measure of pupils’ reading ability. The sample design used was probability proportional to size, with oversampling for subpopulations that constitute a very small proportion of the population.
Overall, pupils who are taught to read in their primary home language have a higher oral reading fluency (ORF) score than those who are not, but the differences are not substantial at this time. It is believed that numerous interfering factors, such as low distribution of textbooks and low parental involvement, have sufficiently confounded results. When these issues are addressed, it may be possible to show that reading skills acquisition occurs more easily and quickly for pupils who are taught in their primary home language. We believe that using GIS data to compare ethnologue maps and LOI is a meaningful way to begin a data-rich discussion around LOI policy and pupil performance.
ORF Score Based on 90–99th Percentile
91
Kasem
Conclusions
50
90
For this study, Early Grade Reading Assessment data were analyzed and combined with geographic information system (GIS) data detailing the languages spoken throughout the country and the LOI employed in schools. The combination adds background and provides a rich visual component for the discussion around LOI policy.
Language Spoken Matches Language of Instruction
60
0
Methodology
Ghana. The models used to estimate the effects of different variables were based on socioeconomic status, location of school in an urban area, and pupil’s status as overage for grade level. All results were weighted.
native tongue.
Sisaala, Western
Languages of Instruction
Bissa
Ligbi
Zambia. The models used to estimate the effects of different variables were based on socioeconomic status, school type, and province. The effects of gender, age, and location of school in an urban area were not found to be significant in this case; therefore, these variables were dropped. All results were weighted.
14.2% of pupils were taught in a language that was not their
Languages Spoken
Nafaanra
The data used for Ghana come from the 2013 Ghana Early Grade Reading Assessment (EGRA) study. The sample was selected through a two-stage sampling process, with proportional selection by region and gender.
The proportion of pupils taught in their native tongue and able to read at least 1 wpm was significantly larger at the 0.01 level, compared with pupils not taught in their native tongue.
70
Republic of Ghana
Ghanaian Language Oral Reading Fluency (ORF)
Republic of Zambia
Population Density
Nzema
4.4 wpm
Population per Square Kilometer 0
RTI International. (2015). National Assessment Survey of Learning Achievement at Grade 2: Results for Early Grade Reading and Mathematics in Zambia. Report prepared under the USAID EdData II project, Data Collection Services for the USAID/ Zambia Education Project, Task Order No. AID-611-M-14-00002 (RTI Task 28). Research Triangle Park, NC: RTI International.
10 - 40
1-2 3-9 41 - 150
!(
151 - 600
Tamale
601 - 2,500 2,501 - 10,500 10,501 - 45,937
Acknowledgment Bemba Kaonde
8.4 wpm
Lozi Lunda
1.9 wpm Lusaka
1-2 3-9 10 - 40 41 - 150 151 - 600 601 - 2,500
The Examinations Council of Zambia of the Ministry of Education, Science, Vocational Training and Early Education
3.6 wpm Kumasi
Mismatched Areas
Nyanja Tonga
Sarrynna Sou +1.919.485.2722 [email protected]
RTI International 3040 E. Cornwallis Road Research Triangle Park, NC 27709
!(
!(
Ho
7.4 wpm
Presented at: CIES 2016 Conference Vancouver, BC, Canada
Accra
2.8 wpm
2,501 - 10,500 10,501 - 43,059
www.rti.org
Kumasi
6.5 wpm
More Information Jennifer Pressley +1.919.541.6898 [email protected]
Ho
8.7 wpm
Luvale
Population per Square Kilometer 0
The National Education Assessment Unit of the Ghana Education Service
RTI International is a registered trademark and a trade name of Research Triangle Institute.
Accra
Cape Coast
60
Primary Home Language Not Primary Home Language
40
20
0 0
Lewis, M. P., Simons, G. F., & Fennig, C. D. (Eds.). (2016). Ethnologue: Languages of the World, Nineteenth Edition. Dallas, TX: SIL International. Available at: http://www.ethnologue.com
This poster is made possible by the generous support of the American people through USAID, under the EdData II funding mechanism: the USAID/ Ghana Partnership for Education: Testing, Task Order No. AID-641-BC-13-00001, and the Data Collection Services for USAID/Zambia Education Project, Task Order No. AID-611-M-14-00002. The views expressed by the authors at RTI International do not necessarily reflect the views of USAID, the United States Government, or the governments of the Republic of Zambia and the Republic of Ghana.
ORF Performance Percentiles by Language of Instruction as Primary Home Language
Cape Coast
10
20
30
40
50 60 Percentiles
70
80
90
100
