Background
Academic achievement of school age children can be affected by several factors such as nutritional status, demographics, and socio-economic factors [
1,
2]. No nation can afford to waste its greatest national resource: the intellectual power of its people. In many poor countries where malnutrition is widespread, it is considered a problem that negatively affects the ability of children to learn and causes them to perform at a lower level in school [
3]. Studies have shown that food insecurity is associated with academic performance of school age children [
4,
5].
An intervention study showed that free school breakfast programs resulted in significant improvements in attendance, mathematics grades, behavior and decreases in hunger [
6]. Another study showed that improvement in academic performance would be achieved once proper nutrition is achieved [
7]. Children who were malnourished in early life are more likely to have lower school attendance and to score poorly in cognitive tests during school age as compared to well-nourished children [
8]. Many studies done in different parts of the world showed that poor growth characteristics are associated with poor academic performance. A prospective cohort study from Canada revealed that though overweight does not increase risk for poor educational outcomes, being underweight increase risk for poorer cognitive outcomes [
9]. A study from Southeast Asia showed that all poor growth characteristics (HAZ, WAZ and BAZ) and non-verbal IQ scores were associated negatively [
10]. A study from Southern Ethiopia also found that better HAZ and WAZ were positively correlated with cognitive performance tests [
11].
Ethiopia is one of the sub-Saharan African countries largely affected by child malnutrition [
12]. Studies conducted in different parts of Ethiopia have shown that under nutrition is common among school age children. A study from northwest Ethiopia revealed that 32.3 % of school age children were undernourished (27.1 % underweight and 11.2 % stunted) [
13]. Another study from Gondar showed that the prevalence of stunting among school-aged children was 42.7 % in rural areas and 29.2 % in urban areas, while the corresponding figures for thinness were 21.6 and 20.8 % [
14]. A study from eastern Ethiopia showed prevalence of stunting among school-aged children was 8.9 % of which 2 % had severe stunting [
15]. A 2014 study in Addis Ababa, the nation’s capital found about 31 % of children were undernourished (19.6 % stunted, 15.9 % underweight) [
16]. This implies that child malnutrition is still a public health problem in both urban and rural areas.
The future of a nation entirely depends on the welfare of its younger generation. Ethiopia is currently transitioning from the end of its first national growth and transformation plan to the beginning of its second growth and transformation plan, of which its success will largely depend on the academic performance of its school children and their future contributions. In 2012/2013 the net enrolment rate in Primary Level (1–8) was 85.9 %. Similarly, the net enrollment rate at Primary Level in the nation’s largest and most populous region, Oromiya region, was 83.9 %. The national repetition rate among primary school students was found to be 7.9 % while the repetition rate of Oromiya region was 9 %. The dropout rate for Oromiya was 18.8 % while the national was 15.7 % [
17]. Therefore proper attention should be given to this group of children because their performance will impact the socioeconomic development of the entire country.
Though evidence about the magnitude of malnutrition is well established in Ethiopia, there is a paucity of evidence regarding how nutritional status is associated with academic performance among the nation’s school age children. Hence, this study is aimed to determine the association of nutritional status and cognitive performance with academic performance of school children in Goba town, South East Ethiopia.
Results
Socio-demographic characteristics
The majority of the study participants were Orthodox (97,74.8 %) by religion and Oromo by ethic group (66, 51.1 %). Ninety five (74.8 %) and 122 (93.1 %) of the school age children were from families who were currently on marriage and residing in urban areas, respectively. Nearly one third of (42,32.8 %) the children were from households headed by husbands. Fifty four (47.0 %) of the school age children had a father who attended formal education up to the level of high school. Nearly two thirds (84, 64.9 %) of the school age children were living with both their fathers and mothers. Nineteen (14.5 %) of school age children were living with their grandparents. Eighty four (63.4 %) of the school age children were male by sex. The mean (±SD) age of the school age children was 10.02 (±0.86) years (Table
1).
Table 1
Socio-demographic and economic characteristics of respondents in Goba Town, South east Ethiopia, 2014
Religion | Orthodox | 97 | 74.8 |
Muslim | 27 | 20.6 |
Protestant | 6 | 4.6 |
Ethnic group | Oromo | 66 | 51.1 |
Amhara | 60 | 45.8 |
Othersb
| 4 | 3.1 |
Marital status | Married | 95 | 74.8 |
Divorced | 6 | 4.6 |
Separated | 14 | 10.7 |
Widowed | 13 | 9.9 |
Head of the household | Husband | 42 | 32.8 |
Wife | 17 | 13.0 |
Both | 54 | 41.2 |
Othersc
| 17 | 13 |
Place of residence | Urban | 122 | 93.1 |
Rural | 9 | 6.9 |
Educational status of the father | Illiterate | 12 | 10.3 |
Primary (1–8) | 24 | 20.5 |
Secondary school (9–12) | 54 | 47.0 |
Tertiary (>12) | 26 | 22.2 |
Educational status of the mother | Illiterate | 13 | 9.9 |
Primary (1–8) | 43 | 32.8 |
Secondary (9–12) | 51 | 38.9 |
Tertiary (>12) | 17 | 13.0 |
Wealth index | Poor | 46 | 36.8 |
Medium | 35 | 28.0 |
High | 44 | 35.2 |
With whom the child lives | Father only | 4 | 3.1 |
Mother only | 19 | 14.5 |
With both | 84 | 64.9 |
Grand parents | 19 | 14.5 |
Othersa
| 4 | 3.1 |
Sex | Male | 83 | 63.4 |
Female | 48 | 36.6 |
School type | Private schools | 67 | 51.1 |
Governmental schools | 64 | 48.9 |
Family size | ≤3 | 32 | 24.6 |
4–5 | 61 | 46.9 |
≥6 | 37 | 28.5 |
There was a statistically significant positive association between all cognitive test scores and average academic performance except for number recall (
p = 0.12) and hand movements (
p = 0.08). This study found that there was a statistically significant positive correlation between all cognitive test scores and mathematics score (
p < 0.05) (Table
2).
Table 2
Correlation between cognitive function test and academic performance among school aged children in Goba Town, South east Ethiopia, May 2014
Number Recall score | r | 0.14 | 0.19* |
p-value | 0.12 | 0.03 |
N | 131 | 130 |
Rovers score | r | 0.22* | 0.22* |
P-value | .013 | 0.01 |
N | 131 | 130 |
Hand Movement score | r | 0.16 | 0.20* |
P-value | 0.08 | 0.03 |
N | 131 | 130 |
Pattern score | r | 0.24** | 0.27** |
P-value | 0.005 | 0.002 |
N | 131 | 130 |
Word Order score | r | 0.23** | 0.19* |
p-value | 0.008 | 0.028 |
N | 131 | 130 |
Triangles test score | r | 0.33** | 0.29** |
p-value | 0.001 | 0.001 |
N | 131 | 130 |
Ravens CPM test score | r | 0.38** | 0.38** |
p-value | 0.001 | <0.001 |
N | 129 | 128 |
As showed in Table
3, there was a statistically significant positive correlation between height for age Z score (HAZ) and mathematics score among school aged children (
p = 0.026). However, both weight for age Z score (WAZ),and body mass index for age Z score (BAZ) had no statistically significant association with academic performance (average semester result and mathematics score) (
p > 0.05).
Table 3
Correlation between anthropometric Z scores and academic Achievement among school aged children in Goba Town, South east Ethiopia, May 2014
HAZa
| r | 0.103 | 0.197a
|
p-value | 0.249 | 0.026 |
N | 128 | 127 |
WAZb
| r | 0.061 | 0.099 |
P-value | 0.497 | 0.270 |
N | 127 | 126 |
BAZc
| r | −0.029 | −0.061 |
p-value | 0.743 | 0.496 |
N | 127 | 126 |
Variables including residence, maternal education, paternal education, diet diversity, meal frequency, breakfast habit, iodized salt consumption, sex of the child, occupation, attendance of preschool program and family size were not significantly associated with academic performance (P > 0.25). Hence those variables were excluded from the multivariable linear regression model.
In the bivariate analysis only height for age Z score and wealth index were significantly associated with mathematics score, qualifying as candidate variables for the multivariable linear regression model. Age and wealth index were positively associated with average score of all academic subjects (courses) of the preceding semester in the multivariable model.
In the multivariable linear regression model better wealth index was significantly associated with better mathematics score (ß = 0.63; 95 % CI: 0.12–0.74). A single unit change of wealth index resulted in 0.63 unit change in mathematics score. Height for age Z score was found positively associated with mathematics score in the multivariable model. A single unit change in height for age Z score resulted in 2.11 unit change in mathematics score (ß = 2.11; 95 % CI: 0.002–4.21). A single unit change in wealth index resulted in 0.53 unit change in average score of all academic subjects among school age children (ß = 0.53; 95 % CI: 0.11–0.95). Older children scored significantly higher in average semester result. A single unit change of age results 3.23 unit change in average score of all academic subjects among school age children (ß = 3.23; 95 % CI: 1.20–5.27).
Discussion
This study found that WAZ and BAZ were not significantly associated with average semester results or mathematics score among school age children, age 8–11 years. Similarly a study from Malaysia found no association between poor nutritional status and school performance among a sample of 7 to 8 years old primary school children [
25]. However, in contrast to the finding of this study, studies have shown that higher WAZ is positively associated with academic performance [
26,
27]. High WAZ score is associated with better cognitive performance [
11]. The small sample size of this study possibly a reason for absence of association between WAZ and academic performance. Additionally, this study was conducted among school age children attending urban schools, where prevalence of underweight is tends to be lower than in rural schools. Therefore the effect of underweight on academic performance in a study population where malnutrition prevalence is low might not reflect the effect at a population level. The absence of association between BAZ and academic performance might be explained by the fact that BAZ indicates acute nutritional status and do not interfere with cognitive functioning.
This study found that good HAZ score was significantly associated with higher mathematics scores. Similar finding came from a study conducted in Sri Lanka which showed that higher HAZ score is associated with better academic score [
27]. Height-for-age reflects the accumulation of nutritional deprivation throughout the years, which may consequently affect educational achievement of children [
28]. A study from Uganda revealed that HAZ had statistically significant positive associations with learning achievement in English (language) and mathematics among grade 4 children [
29]. A study among grade 4 children in Sri Lanka showed a significant positive association between height-for-age and examination scores [
30]. Stunted children have shown more anxiety, depression and lower self-esteem than their non stunted counter parts [
31]. Similarly, protein malnutrition (and low energy availability in general) may have negative effects on brain development. Chronic protein energy malnutrition (stunting) affects the ongoing development of higher cognitive processes during childhood years [
32,
33]. Stunted children have impaired behavioral development in early life [
34] and have poorer cognitive ability [
35] than non-stunted children. Brain functions such as cognition, memory and locomotors skills are affected by under nutrition [
36,
37]. However, the associations between poor linear growth and impaired neurodevelopment are not well understood [
36]. Cognitive functioning might mediate the association between stunting and academic performance. This study found that all the cognitive test scores were significantly associated with the mathematics scores. Except number recall and hand movement, all other cognitive tests were correlated with average semester results.
This study found that higher wealth index is associated with better mathematics score of school age children. Many consistent findings were reported from different part of the world. Food security, which is correlated with wealth index has predicted impaired academic performance in mathematics for girls and boys [
4]. A meta analysis study showed that there is a strong correlation between socio economic status and academic performance [
38]. Poor socioeconomic condition of the family is one of the determinants for poor school performance among children [
39]. Primary school children in Malaysia from the lower income group had significantly poorer academic performance [
2]. The association between socio economic status and academic achievement widens with increasing age. Socio‐economic gaps in the early school years has long lasting consequences. Particularly, as children in low socio economic status get older, their situation tends to worsen [
40]. The association between socioeconomic status and academic achievement is complex and included several factors such as nutrition, school environment, home living environment, material support etc. Socio economic status might affect educational performance due to its influence on affordability of quality of residence, access and affordability of Information Communication facilities and services, and library materials which deal with academic matters [
41]. Low socioeconomic status is associated with chronic stress which has long term negative consequences on brain development [
42]. Furthermore, low socioeconomic individuals have high levels of stress hormones such as cortisol and catecholamines [
43].
This study implies that academic performance is affected by chronic malnutrition and wealth. The education sector, which largely determines the welfare of a country, can be improved by investing in nutrition, particularly on chronic malnutrition. Stunting begins in utero and continues for at least the first 2 years of postnatal life; the period from conception to a child’s second birthday (the first one thousand days) has therefore been identified as the most critical window of opportunity for interventions. There are proven interventions to prevent stunting such as exclusive breastfeeding, complementary feeding, infection prevention, sanitation and hygiene. Nutrition interventions should also be strengthened in schools to improve children’s academic performance. Improving the socioeconomic status of a community has profitable gains in terms of students’ academic performance and development of the nation.
This study had some limitations. The first limitation was its small sample size and biased inclusion of children mostly from urban residency. This study used only anthropomorphic measurements and did not assess the micronutrients status of study participants. Furthermore, there may have been differences in the evaluation system for students’ academic performance among the public and private schools. Finally, physical performance capacity and motor skills were not measured in this study.
Competing interest
The authors declare that they have no competing interests.
Authors’ contribution
DH, KG, DN and HD conceived the study. DH performed the data management and analysis, contributed in the interpretation of data and drafting the manuscript and critically reviewed the manuscript. KG performed the data collection, contributed in the interpretation of data, drafted the manuscript and critically reviewed the manuscript. DN performed the data collection, contributed in the interpretation of data, and critically reviewed the manuscript. HD performed the data collection, contributed in the interpretation of data, and critically reviewed the manuscript. All authors have contributed equally to this manuscript. All authors read and approved the final manuscript.