Musirah Farrukh, 2019 Summer Fellow

Prompted by donor agencies and iNGOs, there is a growing reliance on data analytics for evaluation in development. Statistical analysis of the plethora of data that is available to us has become an integral part of effective decision making. However, in the developing world, there often remains a lack of technical knowledge that impedes the utilization of available data for effective decision making. To this end, Causal Design and GIZ held a capacity building training in Phnom Penh, Cambodia from June 11th to June 14th, 2019. The workshop – delivered by Economist and Country Director in Cambodia, Nathan Ives, and Research Assistant Musirah Farrukh– aimed to increase the capacity of Cambodian health sector professionals to handle large secondary datasets on STATA.

Although the training session was for a small group of people, each person came from a different background and had varying level of interaction with STATA. To accommodate these differences in experience and ability among participants, the session was designed in a more open format, where participants were free to ask questions on a rolling basis, and with plenty of time set aside to practice and one-on-one discussions of technical issues with the trainers. This was especially useful for participants with limited know-how of STATA, who could then be guided through the exercises designed for the workshop by our trainers.

The literature on evaluation capacity building emphasizes the value of grounding training in the context of the participants. To this end, the focal point of our training was the Cambodian Socio-Economic Survey (CSES), which holds an abundance of information useful to policymakers. The goal of the training was to equip participants to analyze data presented in the CSES. However, the learning context should also be kept in mind for training.

Participants requested hard copies of the slides, and other reference material, which was already available online and shared via email. In low internet environments, and with limited access from home, participants may require hard copies of training materials for reference and study. Some participants reported in a survey taken after the training had occurred that they preferred trainings in their local language. Although this survey form did not ask why this may be the case; possibly this preference is due to including the ability to catch nuance or subtle lessons, or simply to relate to the instructor or ask more complex questions with comfort. Thus, the learning context of the participants should also be kept in mind when designing a training session, and trainers would ideally instruct in the local language.

Another noted area to improve the efficacy of training is on the background ability of participants. From in-class observation in this training, we found that the greatest benefit of the program seemed to be for those with greater experience in these concepts. Future programming should reflect this. We theorize this dynamic arises because the software makes it easy to execute a certain action but interpretation often requires a deeper knowledge of concepts. Some of the participants in the training session who struggled most had both limited knowledge of STATA, but also of statistical methods more generally while those with backgrounds in statistics were not observed to have the same difficulties. Although participants were either self-selected or nominated by their affiliate organization, trainers should be ready to deal with participants who may not have a statistical background, since analysis using STATA / R or another statistical software is relevant to a diverse body of disciplines. Including an introduction to relevant statistical concepts in the training, or providing some basic resources or references, might be a beneficial refresher for some, and a crucial introduction to others.

Lastly, a key component of an effective capacity building activity is follow-up assistance. Practice is key in learning any software, and STATA is no different. Although participants were given exercises to complete within the training session, they were also encouraged to explore STATA on their own. Our trainers also shared their contact details with the participants, in case they had further queries. I believe that this is a better approach, as it puts the onus of reaching out on those who seek to learn. However, it would also have been useful to utilize the presence of people with a varying ability on STATA and nominate a local person for the participants who could also be contacted in case of queries. This is based on the belief that participants will be more comfortable with reaching out to their colleagues, as opposed to the trainers. It also helps build a community of people based on knowledge sharing, which may be beneficial beyond the ambit of donor-led training.

In conclusion, skill transfer is an integral part of technology transfer where skill transfer is a more painstaking process and requires greater commitment. Paying heed to the context of the beneficiary and creating a community to share knowledge are perhaps the best ways in which knowledge sharing can take place. This not only ensures that beneficiaries have the best chance of gaining from capacity building, but also ensures the longevity of learnings. Thus ample attention should be given to the context and sustainability of training when planning any capacity-building effort.

Tagged on: