ParentText Optimisation and Evaluation in South Africa

About this project

copy of gpi brand maps south africa yellow
  South Africa
  April 2022 - December 2024
Principal Investigators Dr Hlengiwe Sacolo-Gwebu (University of Fort Hare) and Dr Jamie Lachman (University of Oxford & University of Cape Town)
Research Team Professor Frances Gardner, Moa Schafer and Dr Maria Ambrosio (University of Oxford) , Dr David Stern (IDEMS)
Partners Clowns Without Borders South Africa (CWBSA); South African Department of Social Development (DSD); UNICEF South Africa; mothers2mothers

Overview

ParentText is a chatbot-led parenting intervention, with the aim of increasing positive parenting and reducing child maltreatment. The current optimisation trial will rigorously investigate the most effective and cost-effective combination of intervention delivery on participants engagement.  
 

Context

Research has found that in-person playful parenting programmes delivered in low- and middle-income countries, such as the Parenting for Lifelong Health (PLH) programmes, may be effective at increasing positive parenting, child development, and child wellbeing while decreasing violence against children, intimate partner violence, and parent/child mental health problems. There is also emerging evidence that digital and remotely delivered parenting programmes may be effective at achieving similar effects with greater potential for scalability. However, this evidence is primarily limited to high-income countries where access to web-based programmes is much higher than families living in low-resource settings. This study builds on a pilot study funded by the LEGO Foundation of ParentText in South Africa.

Objectives

The main aim of the study is to identify the condition that significantly contribute to the primary engagement outcome of ParentText in terms of the total number of modules completed as well as to identify the components that significantly contribute to goals completed, skills obtained and the total number of interactions of participants with ParentText. The findings from this optimisation study will be used to inform a future randomised control trial set to take place in 2024.
 

Study Setting

The study will take place in Mpumalanga, South Africa between July and October 2023 in collaboration with UNICEF, the South African Government, and NGO partners. It will include 3,840 caregivers and adolescent girls between the ages of 10 and 17.  This is a 2X2 randomised cluster design with four conditions;   

  1. ParentText + onboarding session
  2. ParentText + onboarding session + additional in-person sessions
  3. ParentText + onboarding session + weekly WhatsApp support group sessions
  4. ParentText + onboarding session + three in-person sessions + weekly WhatsApp support group sessions

Study Significance and Impact

This study will be the first to rigorously optimise and evaluate the effectiveness, cost-effectiveness, and scalability of a parenting programme delivered via a chatbot in a low- and middle-income country. Findings will inform the wider dissemination and scale-up of Parenting for Lifelong Health human-digital programmes, and more specifically (and immediately) the ongoing and forthcoming delivery of ParentText in multiple countries, including Cambodia, Jamaica, Malaysia, Mexico, Philippines, Sri Lanka, Thailand, and Viet Nam.