Challenge: Since the introduction of mobile technology in data collection and management e.g. use of table, the idea of near-real-time monitoring was born through use of CATI (computer assisted telephone interviewing) methods such as mVAM and others, but never has it been real like now after COVID-19 created barriers to the traditional data collection methods. This opportunity created by the pandemic should be extended beyond data colleciton using CATI to developing scientifically accepted methods on the use of remotely collected data e.g. sampling, in order to have acceptable evidence that informs policy. One of the challenges that remotely collected data face (mainly from academia) is mainly around representativeness and sampling.
Use case: In countries such as Yemen where traditional data collection is often a challenge due to conflict, the opportunity created is an acceptance by majority of stakeholders to expand the use of CATI methods in key informant interviews (KII) e.g. market price monitoring, agriculture extension services, desert locust monitoring and agriculture input survey as well as household food security and livelihood surveys. Such interviews are providing higher frequency data used to monitor the impact of the pandemic and can dynamically be adopted to new stressors and shocks as they emerge. Therefore, the use of remote data collection has a dual advantage in conflict countries i.e. high frequency data collection (although sample may be smaller – challenge stated above) and agility to monitor new shocks as they emerge.
Design: Real-time monitoring systems should be designed in a simple and easy to use methods in order to be useful and easily adopted in the country. As mentioned above, the main challenge remains the extent to which analyses generated from such systems can be generalized to inform policy and other decision making. Unlike in development work where time is not a major factor, in humanitarian context time is of essence and quick turnaround from data collection to decision-making can mean life or death for large populations. Therefore, in such context, real-time monitoring should be designed in a way that allows, as much as possible, generalization that are solid and trust-worthy in as much as the statistical requirements of randomly selected sample are not fully met. In Yemen, due to lack of a sampling frame of telephone numbers, the current remote data collection utilizes information form past beneficiaries and using that list as a sampling frame, a small sample is selected for interviews.
Kamau Wanjohi
Challenge: Since the introduction of mobile technology in data collection and management e.g. use of table, the idea of near-real-time monitoring was born through use of CATI (computer assisted telephone interviewing) methods such as mVAM and others, but never has it been real like now after COVID-19 created barriers to the traditional data collection methods. This opportunity created by the pandemic should be extended beyond data colleciton using CATI to developing scientifically accepted methods on the use of remotely collected data e.g. sampling, in order to have acceptable evidence that informs policy. One of the challenges that remotely collected data face (mainly from academia) is mainly around representativeness and sampling.
Use case: In countries such as Yemen where traditional data collection is often a challenge due to conflict, the opportunity created is an acceptance by majority of stakeholders to expand the use of CATI methods in key informant interviews (KII) e.g. market price monitoring, agriculture extension services, desert locust monitoring and agriculture input survey as well as household food security and livelihood surveys. Such interviews are providing higher frequency data used to monitor the impact of the pandemic and can dynamically be adopted to new stressors and shocks as they emerge. Therefore, the use of remote data collection has a dual advantage in conflict countries i.e. high frequency data collection (although sample may be smaller – challenge stated above) and agility to monitor new shocks as they emerge.
Design: Real-time monitoring systems should be designed in a simple and easy to use methods in order to be useful and easily adopted in the country. As mentioned above, the main challenge remains the extent to which analyses generated from such systems can be generalized to inform policy and other decision making. Unlike in development work where time is not a major factor, in humanitarian context time is of essence and quick turnaround from data collection to decision-making can mean life or death for large populations. Therefore, in such context, real-time monitoring should be designed in a way that allows, as much as possible, generalization that are solid and trust-worthy in as much as the statistical requirements of randomly selected sample are not fully met. In Yemen, due to lack of a sampling frame of telephone numbers, the current remote data collection utilizes information form past beneficiaries and using that list as a sampling frame, a small sample is selected for interviews.