Data Equity
The data we are using is a recorded history of lethally shot suspects during police enforcement operations. These records are fairly sensitive since the victims’ names are collected in the record and our project should not include this information. It is not necessary nor contributive.
According to the Washington Post, they are “not tracking deaths of people in police custody, fatal shootings by off-duty officers or non-shooting deaths”. The data is specifically about information regarding the victims of shootings by on-duty officers. The information and conclusions drawn can not be applied to different types of victims or suspects in the US.
The principles related to privacy are harder to judge since this information is collected from local news, social media posts, interviews, and emails sent to the Washington Post. Further details on how the data is gathered and verified are not disclosed on the Washington Post Github page. Hence, we should, again, report our findings that are combined with other data on the macro level. And since the process is not transparent enough, we may need to critically think about the results generated.
Our dataset has limits and we would like to point this transparent as the equity article mentions. Among all of the variables we have, we can not decide why that shooting happened, and we should not simply conclude from the data of races that because white people were involved in those cases the most. Therefore we should pay extra attention to them without any other supporting data. As mentioned in the “Beneficence” principles for advancing equitable data practice, We need to be aware of such sensitive topics before we draw conclusions based on our analysis of the dataset. The most important problem that we have in the perspective of transparency is we do not have the reasons why police have to shoot under which kind of scene.
Many datasets like ours have the problem of getting consent. However, data released by some companies and government departments are open to use. If researchers make significant findings, the community can be benefited. Moreover, as mentioned in the principle of justice during the dissemination stage in the equity article, we need to ensure that our results and the way we present our results can be utilized and understood by the communities. For example, we can adjust the graphs or plots into a clearer and more readable format or adjust the coordinate axis or colors of plots to emphasize the possible pattern.
Another important aspect of transparency is that we should clarify who would have access to the data, what are their purposes, and what to be done with the data to destroy it after the project. Being transparent on this aspect is for the sake of respecting and protecting data providers. Although the data set we are currently working on is open to anyone, it is still possible in the future that we need other data sets that are not open and require us to ask for the providers’ permission, and having such clarification would help the providers to decide whether to provide the data set or not. Therefore, it is always worth it and necessary to keep such clarification.