With the outbreak of the corona pandemic, surveys were launched to measure the population's attitudes and behavior concerning visiting public places during the corona pandemic. However, such questions are likely biased due to social desirability. As survey estimates can be used to evaluate pandemic-preventing measures, their accuracy is of high importance.

In this application, we aim to analyze correlates in survey and sensor-based time- series of people's attitudes and behavior concerning visiting public places. We focus on periods in which new measures were introduced to limit the spread of the virus. This is demonstrated using data of four different web surveys measuring population's attitudes and behavior in Germany. The development of these estimates will be compared with time-series of sensor-based measures of pedestrian flows. Sensors are installed in 61 different cities at 130 highly-frequented public places and streets. These automated systems count the incidence of pedestrians with 99% accuracy. Such data is typically used for investors, retailers, or city planners but was so far rarely used to obtain insight into population behavior concerning compliance with pandemic measures. The sensor and survey observations can be linked on a daily-level using the timestamp. Preliminary results show observed trends being similar in both sources. However, since the pandemic and data collection continues, results will be continuously updated.

From this research, we conclude that linking survey and sensor data of macro- phenomena has the potential of being a promising tool to confirm trends in survey estimates. Further, sensors measuring pedestrian flows provide data on a population's behavior much faster and more frequently than surveys. Therefore, it could be considered as an additional data source in critical situations, such as a pandemic.