11 departments to have access.” In cases where individual departments are developing their own software solutions, “government departments need to define and follow specific policy, open-integration standards, and API standards.” While the plan’s goals align with learning from international practice, it is not clear how much progress has been made in implementation, and the plan is now in its final year. Historical ways of working with data in government departments continue to shape officials’ perceptions of, and approaches to, data collection, storage, use, and analysis: z During the socialist period (1962–1988), central planning relied heavily on targets and quotas, and local officials felt pressured to report that targets had been met, regardless of local realities. A legacy of this system is that officials may continue to use data to produce “right” answers rather than accurate ones. Similarly, the public, as a legacy of authoritarian rule, may feel pressure to give government data collectors “safe” answers rather than the truth. z Myanmar’s long history of centralized decisionmaking, which limited the authority of township officials, has meant that local officials lack decisionmaking experience. The promotion of evidencebased decision-making may rely incorrectly on the assumption that local officials are experienced decision-makers. z Historically, departments and ministries have been siloed and have tended to rely on their own data, and they may still be reluctant to share data with other departments where data-sharing practices were not previously in place. z Likewise, government data in Myanmar has been viewed as privileged information, and there have been sensitivities around sharing data that may reflect poorly on government. As a consequence, data sharing within government, data transparency, and data publications have been uncommon. z Traditionally, much of the work of township departments has been paper based, and new digital solutions must overcome lack of experience and significant apprehension. For any digital solution to succeed, officials must understand how it will help them do their work better. When digital solutions create added work with little perceived benefit, officials may simply revert to paper. Despite the pressure to switch to digital systems, the fact that paper remains essential—for transactions requiring signatures, for example—means it will remain the default medium for much government information. The skill and experience of those working with data are a critical and often-overlooked dimension of the local data governance framework. High levels of data literacy are uncommon in Myanmar, and the NSDS notes that statistical activities are often conducted by staff with no background in statistics, under inadequate professional supervision.23 Collecting high-quality data and using it effectively require training and guidance. Any work to strengthen Myanmar’s local data ecosystem must consider the skill and experience—the data literacy—of those who will do the work. 2.2 WHAT DATA IS COLLECTED AT THE LOCAL LEVEL, AND HOW IS IT COLLECTED? The government of Myanmar has considerable experience in collecting data at the local level, and data collection is a routine responsibility of township offices. Data collection is perceived to be a routine and fundamental part of the role of township officials. Most departments collect data on a regular basis, often annually, on a broad range of subjects and using a variety of collection methods. Numerous government departments are involved in the collection of data at the local level. Many, if not all, of the roughly 40 departments present at the township level collect data. While this situation is not unique to Myanmar, the large number of government departments involved in data collection creates a complex patchwork of data, covering many sectors and areas and creating significant challenges in understanding the local data ecosystem. Some data is collected through explicit data-collection projects; other data is information that officials collect in the course of their routine administrative duties. Consider, for example, the difference between data collected for Myanmar’s Population and Housing Census24 and the household information collected by DAO officials when collecting property tax. Local data collection is carried out across a broad range of sectors and areas, and can be broadly grouped into the following categories: z Demographic data. Government departments collect a broad range of demographic and socioeconomic