Numeracy in Kenya: Examining the Effect of Household Assets using the Uwezo Dataset

Casey Lynn Haney; Brenden Christopher Drinkard-McFarland; Jennifer Deboer

Download Paper | Permalink

Conference: 2021 Illinois-Indiana Regional Conference
Location: Virtual
Publication Date: April 16, 2021
Start Date: April 16, 2021
End Date: April 17, 2021
Conference Session: Diversity, Inclusion, and Access
Tagged Topics: Diversity and Diversity, Inclusion, and Access
Page Count: 10
DOI: 10.18260/1-2--38272
Permanent URL: https://peer.asee.org/38272
Download Count: 390

Request a correction

Paper Authors

biography

Casey Lynn Haney Purdue University at West Lafayette (COE) orcid.org/0000-0002-5391-6157

visit author page

Casey Lynn Haney is a Ph.D. student in engineering education at Purdue University and a research assistant in Dr. Jennifer DeBoer's Lab. She completed her undergraduate degree in Engineering/Technology Teacher Education. Her interests include STEM self-efficacy for diverse students and data analysis within educational datasets.

visit author page

biography

Brenden Christopher Drinkard-McFarland Purdue University at West Lafayette

visit author page

Brenden Drinkard-McFarland is a current 1st year PhD student at the Ohio State University studying engineering education. Working under Dr. David Delaine in the inclusive Community Based Lab (iCBL), he looks to broaden his horizons as he continues down the path of engineering education.

visit author page

biography

Jennifer Deboer Purdue University at West Lafayette (COE)

visit author page

Jennifer DeBoer is currently Associate Professor of Engineering Education at Purdue University. Her research focuses on international education systems, individual and social development, technology use and STEM learning, and educational environments for diverse learners.

visit author page

Download Paper | Permalink

Abstract

The 2015 Uwezo dataset, an East African assessment of child literacy and numeracy, offers unique and critical insight into the effects of family resources on critical “gateway” skills like numeracy by collecting data at the household level. Research around what educational researchers call “the Heyneman-Loxley effect”, which is the observed relation of family versus school effects for lower income countries compared with higher income countries, has found that in lower income countries school effects are more important than family effects for educational achievement. Further research has shown that family effects have increased in all countries and are still significant in their contributions to educational achievement. Thus, to contribute to this conversation on the importance of family versus school effects, this paper examines two particular aspects of the family effect: educational assets available at home and socioeconomic status. Previous findings for other datasets have shown that the availability of home resources (such as educational resources and physical assets) and household wealth correlated with higher academic achievement. However, the Uwezo assessment specifically collects data from households, not schools, allowing for the sampling of all youth ages 6-16 including out of school youth and recording the assets available to children in their home. Utilizing nested models to allow for comparison between models, this study estimates four models analyzing the correlation of family education assets with student numeracy and then adding individual demographics, educational attainment, and socioeconomic status with each subsequent model. The first regression model showed that education assets were positively associated with increased likelihood of higher math scores. With the addition of other factors, the correlation of educational assets with math scores remained significant (p < 0.05). Individual demographic factors added in model 2 also remained significant throughout other models and supported previous research regarding the importance of these factors for predicting numeracy. Educational attainment ordinal variables were added in model 3 including school type and current enrollment status. Compared with students who were not attending school, those who attended public school were less likely to be numerate while those attending private school were more likely to be numerate. Students who completed their elementary education had a significantly higher likelihood of numeracy than students who had never enrolled in school. However, enrolled students were not significantly different from those students who had never enrolled. Model four added an ordinal variable for economic status with the likelihood of numeracy increasing for each subsequent level of SES when compared with the lowest value of SES. Overall, this analysis supports the significance of family effects (specifically family educational assets and household wealth) which remain significant with the addition of individual demographics and educational attainment. Household educational assets need to be examined further to understand what if any interventions could be implemented at the household level to benefit students. Further research should investigate how educational resources can be best utilized to strengthen student numeracy skills across the globe and examine how other types of family resources (nutritional, vocational, etc.) affect numeracy.

Citation
Format

Haney, C. L., & Drinkard-McFarland, B. C., & Deboer, J. (2021, April), Numeracy in Kenya: Examining the Effect of Household Assets using the Uwezo Dataset Paper presented at 2021 Illinois-Indiana Regional Conference, Virtual. 10.18260/1-2--38272

TY  - CPAPER
AB  - The 2015 Uwezo dataset, an East African assessment of child literacy and numeracy, offers unique and critical insight into the effects of family resources on critical “gateway” skills like numeracy by collecting data at the household level. Research around what educational researchers call “the Heyneman-Loxley effect”, which is the observed relation of family versus school effects for lower income countries compared with higher income countries,  has found that in lower income countries school effects are more important than family effects for educational achievement. Further research has shown that family effects have increased in all countries and are still significant in their contributions to educational achievement. Thus, to contribute to this conversation on the importance of family versus school effects, this paper examines two particular aspects of the family effect: educational assets available at home and socioeconomic status. Previous findings for other datasets have shown that the availability of home resources (such as educational resources and physical assets) and household wealth correlated with higher academic achievement. However, the Uwezo assessment specifically collects data from households, not schools, allowing for the sampling of all youth ages 6-16 including out of school youth and recording the assets available to children in their home. Utilizing nested models to allow for comparison between models, this study estimates four models analyzing the correlation of family education assets with student numeracy and then adding individual demographics, educational attainment, and socioeconomic status with each subsequent model. The first regression model showed that education assets were positively associated with increased likelihood of higher math scores. With the addition of other factors, the correlation of educational assets with math scores remained significant (p &amp;lt; 0.05). Individual demographic factors added in model 2 also remained significant throughout other models and supported previous research regarding the importance of these factors for predicting numeracy. Educational attainment ordinal variables were added in model 3 including school type and current enrollment status. Compared with students who were not attending school, those who attended public school were less likely to be numerate while those attending private school were more likely to be numerate. Students who completed their elementary education had a significantly higher likelihood of numeracy than students who had never enrolled in school. However, enrolled students were not significantly different from those students who had never enrolled. Model four added an ordinal variable for economic status with the likelihood of numeracy increasing for each subsequent level of SES when compared with the lowest value of SES. Overall, this analysis supports the significance of family effects (specifically family educational assets and household wealth) which remain significant with the addition of individual demographics and educational attainment. Household educational assets need to be examined further to understand what if any interventions could be implemented at the household level to benefit students. Further research should investigate how educational resources can be best utilized to strengthen student numeracy skills across the globe and examine how other types of family resources (nutritional, vocational, etc.) affect numeracy.
AU  - Casey Lynn Haney
AU  - Brenden Christopher Drinkard-McFarland
AU  - Jennifer Deboer
CY  - Virtual
DA  - 2021/04/16
PB  - ASEE Conferences
TI  - Numeracy in Kenya: Examining the Effect of Household Assets using the Uwezo Dataset
UR  - https://peer.asee.org/38272
DO  - 10.18260/1-2--38272
ER  -