It is common practice for data scientists to probe multiple data sources to achieve higher analytics reaults. In contrary to the conventional closed-world assumption, we take the open-world data analytics problem where, even after spending much time and efforts in collecting data, there is always a chance that some data items are missing. To this end, we try to quantify the impact of unknown data items to aggregate query results using the techniques described in https://dl.acm.org/citation.cfm?id=2882909