Data for the paper on “Does early growth failure influence later school performance? A cohort study in Karonga district, northern Malawi”

Data Description

Anthropometric data collected as part of the baseline census of a demographic surveillance performed between 2002 and 2004 in the southern part of Karonga district in northern Malawi. Data was collected at birth, with follow-up after a year and with further rounds of anthropometric data for all children below the age of 10 between 2008-2011. Individual and household-level socio-demographic and schooling data were also combined to account for socio-economic and schooling histories of participants from 2007-2015, allowing us to examine the relationship between stunting in early years and school outcomes at older ages.

Data Collection Methods

Staff were trained to collect anthropometric data at regular intervals. Recumbent length was used for children below the age of 2, while height for children above age 2 was measured using a Leicester height measure. Survey data was administered by field staff using paper questionnaires and in the local language, Chitumbuka. Data collection was supervised and managed on site by a team of scientists, field enumerators and data entry personnel. Data was then double-entered into MS Access by the data entry office for further analysis. Informed consent to participate in the study was sought from household heads. Further details on data collection methods can be found in the paper.

Data Analysis and Preparation

Raw anthropometric data was transformed into height-for-age z-scores (HAZ) based on the WHO growth references for children below and above age 5. Growth failure/stunting was defined as the HAZ <-2SD. Growth trajectories between early and late childhood were defined as being never stunted, improvers (stunted in early but not stunted in late childhood), decliners (not stunted in early but stunted in late childhood) or persistently stunted (in early and late childhood). For data on school outcomes, age at entry was determined by the cut-off of age 6 (<6 being underage, >6 being overage enrolment). Age-for-grade is the number of years a child is ahead/behind in class based on the official age-for-grade and provides a cumulative measure of school performance. The effects of stunting on grade repetition in Standard 1, which is has the highest proportion of repeaters in primary school, was also examined.

Geographic regions

Karonga district, northern Malawi

Key dates

See description

Quality Controls

Data collection was supervised and managed on site by a team of scientists, field enumerators and data entry personnel. Data was then double-entered into MS Access by the data entry office for further analysis.

Species:

Human population

Privacy:

The raw dataset enabled individual participants to be identified and linked within and across households over time. To protect their confidentiality, data was suppressed to avoid deductive disclosure, resulting in slight deviations of results from the original dataset.

• ID numbers have been scrambled using Nesstar Publisher software’
• Quasi-identifiers (name & location) and sensitive (personal information that may otherwise violate data confidentiality) variables were anonymised as needed to minimise the risk of re-identification using Statistical Disclosure Control methods available in the sdcMicro software.
• Some dates have been changed to years only
• some variables are grouped

The dataset has been checked with Statistical Disclosure Control (SDC) methods (Templ, M., Kowarik, A. and Meindl, B., 2015. Statistical disclosure control for micro-data using the R package sdcMicro. Journal of Statistical Software, 67(1), pp.1-36). Further details on the methods used for data capture and analyses have been elaborated in the methods section of the paper.

Ethics

Ethics approval for the demographic surveillance and anthropometric studies was obtained from the:

• National Health Sciences Research Committee in Malawi (#419)
• London School of Hygiene and Tropical Medicine Research Ethics Committee

For the demographic surveillance and anthropometric data collection, informed verbal consent was given by the head of household.

Keywords

Stunting; School progression; school performance; 20

Language of written material

Chitumbuka, English

Project title

Does early growth failure influence later school performance? A cohort study in Karonga district, northern Malawi

Funder/Sponsor

Wellcome Trust

UK Economic and Social Research Council

Grant Number

Wellcome Trust - grant numbers 079828/Z/06/C and 098610/Z/12/Z

UK Economic and Social Research Council - grant number ES/L013967/1

Data Creators

Associated Roles

File Description