Assessing the Impact of the National Apprenticeship Programme (NAP) in Ghana
Last registered on September 08, 2017

Pre-Trial

Trial Information
General Information
Title
Assessing the Impact of the National Apprenticeship Programme (NAP) in Ghana
RCT ID
AEARCTR-0001948
Initial registration date
September 07, 2017
Last updated
September 08, 2017 2:20 PM EDT
Location(s)

This section is unavailable to the public. Use the button below to request access to this information.

Request Information
Primary Investigator
Affiliation
University of Virginia
Other Primary Investigator(s)
PI Affiliation
NYU Abu Dhabi
PI Affiliation
University of British Columbia
Additional Trial Information
Status
On going
Start date
2012-08-30
End date
2018-06-01
Secondary IDs
Abstract
This study seeks to estimate the effects of completing an apprenticeship on labor market and other outcomes in Ghana. It includes three nested randomizations, the first of which randomly offered applicants entry into an apprenticeship coordinated by the National Apprenticeship Program (NAP). The second randomization placed treatment applicants from the first randomization with interested firms, conditional on applicant interest and geographic feasibility. This design allows us to experimentally explore how outcomes vary with firm characteristics. The third assigned a random half of training NAP apprentices from the second randomization to a program which provided cash incentives to firm owners conditional on apprentice performance on a skills test.

External Link(s)
Registration Citation
Citation
Hardy, Morgan, Isaac Mbiti and Jamie McCassland. 2017. "Assessing the Impact of the National Apprenticeship Programme (NAP) in Ghana." AEA RCT Registry. September 08. https://www.socialscienceregistry.org/trials/1948/history/21293
Experimental Details
Interventions
Intervention(s)
Apprenticeship training in garments, cosmetology, carpentry, welding, or blocklaying.
Intervention Start Date
2013-11-01
Intervention End Date
2015-12-31
Outcomes
Outcomes (end points)
The main outcomes of interest are labor market related: wage or self employment, earnings, labor supply, migration, and characteristics of wage or self employment. Other outcomes include trade skills, marriage and fertility, and mental health and wellness.
Outcomes (explanation)
Experimental Design
Experimental Design
The study contains three nested randomizations, the first of which randomly offered applicants entry into an apprenticeship coordinated by the National Apprenticeship Program (NAP). The second randomization placed treatment applicants from the first randomization with interested firms, conditional on apprentice preferences and geographic proximity. The third assigned a random half of training NAP apprentices from the second randomization to a program which provided cash incentives to firm owners conditional on apprentice performance on a skills test. A set of about 300 additional "priority" applicants to the first randomization, who were hand-selected by district officials and guaranteed a place in the program, were also included in the second and third randomizations and will be part of the sample targeted by the endline survey. Sample numbers and further detail in this trial registration refer only to the first/primary randomization, and exclude "priority" applicants.
Experimental Design Details
Not available
Randomization Method
We created an electronic randomization tool which contained pre-populated information regarding the breakdown of treatment slots available per district and trade, and then divided applicants into treatment and control groups. Pre-populated digitized NAP application form data was included, and the actual randomizations were then done electronically in each district with district officials in attendance.
Randomization Unit
Individual, stratified by district and trade
Was the treatment clustered?
No
Experiment Characteristics
Sample size: planned number of clusters
The treatment was not clustered.
Sample size: planned number of observations
Total endline target sample: 3,947; Of these experimentally included in the first/primary randomization: 3,599
Sample size (or number of clusters) by treatment arms
Treatment: 2,031 individuals Control: 1,568 individuals
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Name
IPA IRB
IRB Approval Date
2012-08-30
IRB Approval Number
563.12August-004
IRB Name
UBC IRB
IRB Approval Date
2015-11-13
IRB Approval Number
H15­02708
IRB Name
UVA IRB
IRB Approval Date
2016-01-28
IRB Approval Number
2015000700