18 differences in capabilities between townships, there are certain common challenges that prevent township department officials from routinely and systematically making evidence-based policy: z Absence of a culture or history of data analysis at the local level. Most township officials have had little exposure to data or how data can be used. Without this exposure, and a clear sense of how data can help with their jobs, they are unlikely to increase their use of data. Local officials are familiar with their role as data producers, but they have little awareness of what it means to also be data consumers. Officials tasked with collecting data are expected to pass that data on to higher administrative levels. They are effectively precluded from conducting any analysis or thinking critically about the data they collect. Thus, there is little demand for more data or more analysis. z Collected data insufficient for effective analysis. Data that is collected tends to be aggregated statistical data, which may not capture the nuances of a given situation. For example, to understand traffic accidents, the times, places, and individuals affected are critical considerations, not just the raw numbers of accidents or fatalities. Spatial data in particular can be of great use to decision-makers, but the practice of including spatial codes with geocoordinates alongside other data is rare. z Ill-suited formats. Data stored as paper, as Microsoft Word documents, as unstructured spreadsheet files, or as PDFs is difficult to analyze. Aside from what can be done by hand, even simple analysis requires that the data first be formatted. Creating structured data from unstructured data is time consuming and often is simply not considered. z Poor digital infrastructure. Effective analysis depends on having not only properly structured data but also adequate hardware, software, and, of course, electricity. Although townships vary considerably in this regard, inadequate infrastructure in all these categories is common (box B). z Data-analysis skills. Township staff who pursue data analysis have few resources to develop their skills. Familiar with producing totals, averages, counts, and perhaps basic charts and graphs, many staff would like to develop their competencies further. When data trainings are offered, staff are enthusiastic participants, even at weekend classes. There is a need to build basic numeracy and critical thinking along with data-management and analysis skills. Officials must be able to ask themselves what they expect to find in the data, to determine whether the data matches those expectations, and to assess the overall quality of the data. In the case of spatial data, for example, only a few departments have staff with the skills for effective analysis. As already noted, hardware and software to support more advanced skills are often lacking. There are examples of township department offices carrying out more complex data analysis and use. For example, the Taunggyi DAO has turned to technology and data to help it expand its water supply network, improve sustainability, and enable residents to analyze their own water use. A team of engineers used google maps to spatially locate critical elements of its water supply system and to devise a sequential plan to extend the network to new households. It has introduced water meters to encourage households to ration water use and to obtain information on how water demand varies across different parts of the city. It has begun piloting smart water meters that will provide realtime data on household use and help identify where leaks have occurred. The smart water meters can be synced with households’ phones through the MyoTaw mobile application, enabling households to assess their own water use through their phones. Planned features include the ability to set usage caps and receive automatic push-notifications as their usage nears these limits. In another example, the Mandalay City Development Committee has installed remotecontrol traffic lights, high-definition video cameras, road sensors, and loudspeakers at intersections throughout the city. Software in the traffic control room collects data and generates dynamic traffic-flow visualizations for a handful of trained officials overseeing the system. The new technology, the Sydney Coordinated Adaptive Traffic System (SCATS), uses artificial intelligence to optimize traffic flows across the city. Learning and improving over time, it can answer questions such as, “To maximize traffic flows, how long should the green light stay on during rush hour?” The system runs daily experiments and makes continual, small adjustments without the need for a human operator. SCATS has already increased traffic flows over one of Mandalay’s main bridges by a reported 50 percent.