Information Technology, Adult Education and Welfare
Last registered on August 07, 2014

Pre-Trial

Trial Information
General Information
Title
Information Technology, Adult Education and Welfare
RCT ID
AEARCTR-0000476
Initial registration date
August 07, 2014
Last updated
August 07, 2014 5:29 PM EDT
Location(s)
Primary Investigator
Affiliation
Tufts University
Other Primary Investigator(s)
PI Affiliation
London School of Economics
PI Affiliation
University of Ottawa
Additional Trial Information
Status
On going
Start date
2014-02-15
End date
2016-06-30
Secondary IDs
Abstract
Education is essential for economic and social development. Despite massive improvements in enrollment rates over the past 20 years, over 775 million adults worldwide—or 18 percent of the population—are still unable to read and write in any language (UNESCO 2012. These indicators are particularly low in the landlocked countries of sub-Saharan Africa, where literacy rates typically range between 30 and 50 percent. In Niger, the subject of our study, fewer than 30 percent of the population is literate, with large discrepancies between men and women.

While there is a substantial amount of research on how to increase school participation, especially for school-aged children, there is still considerable debate about how to improve learning in a cost-effective way. A number of randomized evaluations have shown that spending on particular inputs has not increased test scores, thereby leading to general skepticism about the value of such inputs (Hanushek 1995, Glewwe, Kremer and Moulin 2009). Yet a key question emerging from many of these studies is whether such inputs are relevant to the local context.

Despite the immediate private and social returns to adult education, adult education programs are a highly neglected entry point for development interventions. This is often attributed to low enrollment, high drop-out and rapid skills depreciation (Romain and Armstrong 1987, Abadzi 1994, Oxenham et al 2002, Ortega and Rodriguez 2008), possibly due to the limited relevance of such skills in daily life or limited opportunities to practice in one’s native language.

The widespread growth of mobile phone coverage in many developing countries, including Niger, has the potential to increase the incentives for and facilitate the acquisition of literacy and numeracy skills by illiterate adults (Aker et al 2012). By teaching students how to use mobile phones, adult learners may be able to practice their literacy skills outside of class by sending and receiving short message services (SMS), making phone calls and using mobile money (m-money) applications, all of which require basic fluency with the numbers, symbols and letters on mobile phone keypads. Mobile phone technology could also affect returns to education by allowing households to use the technology for other purposes, such as obtaining price and labor market information and facilitating informal private transfers (Aker and Mbiti 2010, Jack and Suri forthcoming).

Our research team ran a successful randomized evaluation in Niger in 2009-2011 (Project ABC) and showed that a mobile phone-enhanced adult education program improved literacy and math skills of adult education participants by 19-25 standard deviations, equivalent to an additional year of education (Aker et al 2012). This difference remains significant over time, even as learning depreciates. The results were not only cost effective as compared with the regular adult education intervention, but also as compared with other education interventions also cost effective, costing less than $US 2 for every additional .1 s.d. increase in test scores (Evans and Ghosh 2008).

This research will build upon the existence research in five ways. First, this new research will assess the impact of the traditional adult education program (without mobile phones), which was not possible in previous work. Second, the mobile phone-based literacy program will be implemented in more villages and more literacy participants, allowing us to determine whether the results remain once scaled. Third, the program will be implemented without the distribution of shared mobile phones, thereby overcoming an important cost constraint to implementation by the Ministry of Non-Formal Education (MNEF). Fourth, the program will assess the potential role of mobile phone technology in distance learning by providing educational content on agriculture and health-related themes over the mobile phone. Finally, the research will assess the impact of both the adult education and ABC program on other economic outcomes, self-esteem and self-efficacy and educational attainment of children, key issues justifying the investment in adult education programs (Blunch and Portner 2011, Banerji et al 2013). And finally, the program will assess the way in which education and technology affect intra-household and inter-village dynamics.

This study will use a randomized evaluation methodology, whereby villages within two regions of Niger will be randomly assigned to either a treatment (an adult education program with or without a mobile learning component) or control (no adult education program). We will also cross-cut these treatments by assigning certain villages to the SMS-based adult education program. In addition, a subset of villages in each group will be assigned to mobile phone monitoring, in an attempt to understand whether information technology can reduce the moral hazard problem.
External Link(s)
Registration Citation
Citation
Aker, Jenny, Christopher Ksoll and Frank-Borge Wietzke. 2014. "Information Technology, Adult Education and Welfare." AEA RCT Registry. August 07. https://www.socialscienceregistry.org/trials/476/history/2315
Sponsors & Partners

There are documents in this trial unavailable to the public. Use the button below to request access to this information.

Request Information
Experimental Details
Interventions
Intervention(s)
The research design consists of three main interventions:

1. In the basic intervention (literacy), adults will participate in an education course for ten months over two years, during which time they will learn reading, writing, and math skills in their primary indigenous language (Hausa). The course will follow the Ministry of Non-Formal Education’s (MNFE’s) adult education curriculum, which first provides basic literacy and numeracy training, and then thematic training in agriculture and health. Each village will have two classes separated by gender which will be taught by a local literacy teacher. Literacy participants will be randomly chosen from among all eligible (i.e. illiterate) individuals in the village.
2. The second intervention (ABC) will follow the same curriculum as the basic intervention, but literacy participants will learn simple operations on the mobile phone. This innovation builds on a tested and successful pilot intervention in Niger (Aker et al., 2012). However, unlike the earlier intervention, literacy participants will not be provided with a shared mobile phone (mobile phone ownership is around 50 percent). This decision is partly a response to the pilot phase (when few of the shared mobile phones were used) and partly a strategy to increase the cost-effectiveness and scale of the program for the MNFE, which would not be able to distribute mobile phones in this context.
3. The third intervention (ABC2) will provide adult learners with educational content via the mobile phone, in particular sending SMS that reinforce the concepts learned in class. The SMS will vary in terms of frequency, type of information provided (i.e., a reminder or detailed technical information) and intensity (i.e., number of students).

Finally, a subset of villages above will be randomly assigned to a mobile phone monitoring intervention, to better understand how monitoring affects moral hazard of teachers' performance.
Intervention Start Date
2014-03-01
Intervention End Date
2015-08-31
Primary Outcomes
Primary Outcomes (end points)
The key quantitative outcomes of interest in this research are as follows: 1) Individual level (students’ test scores, including literacy tests, math tests, self-esteem and self-efficacy); 2) Household level (farm and agricultural marketing outcomes, investments in children’s education, health knowledge and intra-household bargaining); and 3) Child level (test scores and cognitive tests).

In addition, for the monitoring research, we will also collect attendance data at the student and teacher level, as well as village-level infrastructure data.
Primary Outcomes (explanation)
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
Our research design will use a randomized control trial with randomized assignment to different treatment variations. Out of the 500 villages in the project, we will randomly assign 175 villages to one of four interventions: 1) a traditional adult education program; 2) a mobile phone-enhanced education program (ABC); 3) a mobile phone content program (ABC2); or 4) no program. Within the third treatment, randomization of the message type, frequency and intensity will occur at the individual level, thereby providing variation in the extent to which other learners in the network of a particular individual are treated. These messages will include simple reminders, technical and non-technical information, which will allow us to disentangle the impact of the information provided from the impact of receiving a message. Since the randomization of messages will be at the individual level, we will be able to provide evidence on the cost-effectiveness of sending messages to all individuals, as well as spillover impacts between recipients and non-recipients. The identification of the impact of both the adult education and ABC programs is based upon the assumption that random assignment will balance observable and unobservable characteristics prior to the program to ensure that the groups are comparable. The unit of randomization will be at the village level. Villages will be stratified by region and sub-region before being randomly assigned to one of three interventions.
Experimental Design Details
Randomization Method
The randomization will be conducted in an office by a computer program (STATA), and shared with the NGO.
Randomization Unit
Village and individual level.
Was the treatment clustered?
Yes
Experiment Characteristics
Sample size: planned number of clusters
The planned number of clusters is 175.
Sample size: planned number of observations
We plan to have approximately 2625 individuals
Sample size (or number of clusters) by treatment arms
There are 75 schools in the adult education program, 75 in the ABC program and 25 in the comparison group. The messages will be cross-cutting in the adult education and ABC schools.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Name
Tufts University Medford Campus
IRB Approval Date
2014-01-28
IRB Approval Number
1312027
Post-Trial
Post Trial Information
Study Withdrawal
Intervention
Is the intervention completed?
No
Is data collection complete?
Data Publication
Data Publication
Is public data available?
No
Program Files
Program Files
Reports and Papers
Preliminary Reports
Relevant Papers