Public Lab Research note


Reflections on Sensor Journalism

by AbbeyCollins | October 06, 2014 01:33 06 Oct 01:33 | #11231 | #11231

The chorus of sound omitted from a handful of hastily-built homemade sensors delighted a room of journalists with little experience in electrical wiring. Not only did the wiring succeed, the sensors presented a new way to tell a story. It was an experiment in sensor journalism, the emergence of which presents new challenges and opportunities for journalists across all mediums.

For Emerson College’s Data Visualization class, we built Coqui water conductivity probes with the assistance of Don Blair from Public Lab. We then used the sensors to test water from different sources throughout the city of Boston. Diving into this form of research was very exciting, given the evolution of journalism that is happening in the digital age. According to the Tow Report, Sensors and Journalism, “Sensors are a way of collecting information about the world. Journalists trade in acquiring information, analyzing it, organizing it, and distributing it. That alone suggests a natural fit,” (Pitt 18). As journalists we have this task to find information and use it to tell the public a story and sensors provide a whole new resource for obtaining that information.

The Coqui itself generates an audible frequency and with the addition of a conductivity probe generates different frequencies based on how conductive the water is. In a broad sense, the more “stuff” there is in a given water sample, the more conductive it will be and the higher the frequency generated. This exercise represents many of the opportunities and challenges of the use of sensors the collect data for journalistic purposes. 
The use of sensors to collect data presents an interesting chance for journalists to collect research first hand. That means no sorting through other people’s potentially disorganized spreadsheets and dealing with FOIA requests. It also presents the ability to report on information that might not already be researched or may be falsely reported by agencies. Sensors and Journalism notes, “One of the rationales for journalists producing their own data (using sensors or any other method) is that official sources cannot always be trusted; perhaps because they are incomplete, or they are wrong,” (Pitt 63). There are a few glaring complications here, the first being that journalists are generally not scientists. While there are certainly some journalists who are suited to conduct this sort of scientific research, not all fall into this category and there are scientific standards that this research must be held to. In the same realm, as scientific standards need to be upheld, as do journalistic standards. In any credible journalistic story multiple sources need to be taken into account, meaning perhaps the FOIA request needs to be filed alongside primary sensor research. Otherwise, objectivity and accuracy may be at stake. When Allison Young was investigating contaminated soil for USA Today, she was successful in her data collection and telling of the story and yet the industry pushed back.

“When USA Today published, according to Young, experts who worked in environmental agencies said that her process was not the same one they would use to plan and estimate the cost of a cleanup and suggested her story was therefore invalid, (Pitt 79-80).”

Additional documents and research can, in some circumstances prevent this pushback from squashing sensor journalism. In this case, a paper by Dr. William Eckel already brought the issue to light and Young did additional research using documents from the environmental protection agency.

Sensor journalism allows data collection to be done at a very low cost by a range of people without engineering experience. Many sensors range from smartphone apps to simple circuits built out of low-cost materials, like our Coqui. Not only does this allow research to be crowdsourced, it allows everyday people to participate. Matt Waite at Poynter writes, “We can build ideas that let people know data about their own location, and feed that data into a greater whole. We can build ideas that let people know data about their own location, and feed that data into a greater whole,” (Waite). This level of community engagement not only furthers scientific research that can then be used for journalistic purposes, it also creates a greater sense of awareness within communities about the world around them. With WNYC’s cicada tracking project, “people in WNYC’s community bought parts, assembled them through 29 steps, wrote code, took sensor readings, and manually contributed them via an online form. One person (operating within his own community who contributed help), went even further by refining WNYC’s designs and producing a newly improved version,”(Pitt 103).

One of the most obvious issues here is the amount of room for error. With crowdsourced research, you must account for not knowing exactly how each phase of research is done. This can likely be accounted for by disregarding outlying data. There is also room for error with the tools used to collect data. Low-cost production also means potentially lower quality sensing then with a factory made sensor. Sensors and Journalism looks at what Pubic Lab is doing, for example,

“In comparison to lab-grade equipment, Public Lab’s current hardware is extremely immature, and while it does produce data, it is not clear whether that data can be compared to known health or regulatory standards to the degree that would stand up in court or withstand a contest from polluters,” (Pitt 67).

In a broader sense, sensor journalism, still in its early stages of development, lacks significant regulation. Though it is easy to collect all sorts of data with basic sensors, things like privacy and safety must be taken into account. Our conductivity sensors were fairly harmless in the realm of privacy, but safety could have been an issue if, say, someone collecting a water sample fell in the body of water they were collecting from. Though it was not really an issue with our conductivity sensors, privacy is at stake in sensor journalism and needs to be highlighted. As this facet of journalism develops, it is essential to maintain the public’s trust. That means being transparent about data collection. Kathleen Bartzen Culver writes in Sensors and Journalism, “They must be made aware when data about them is being collected and stored. News organizations adhering to the common good cannot be opaque about their data collection,” (Bartzen Culver, 149).

In an editorial sense, sensor journalism, while an efficient way to gather data, does not tell a story in itself. More research must be done, more sources consulted, and more human elements added to create impactful journalism. For example, our conductivity sensors were able to tell us how conductive each water sample was relative to the others we tested. What the sensor could not tell us was why each sample tested the way it did. Did the conductivity seem higher because of salinity or because of large amounts of algae? Or was the water actually polluted? These are questions that would take additional research and monitoring of the same bodies of water to figure out. Sensors themselves do not necessarily tell the whole story and even if they appear to, the data must be verified. Then there are circumstances where, despite accurate research, the results of sensor journalism do not warrant the desired story. This is true with traditional reporting as well, but in the case of data-driven sensor journalism, the work can still drive a positive effect. For example, when the Houston Chronicle investigated air quality in Texas in 2002, the findings fell within the legal limits for air quality. Still, the investigation drew attention to the issue and resulted in action. “In the weeks following publication, the mayor of Houston, Bill White, funded a task force of experts to investigate Texas’ air pollution and its health effects,” (Pitt, 56). That is, despite the findings falling short of significance. In this example, the data was humanized, focusing on areas around homes. Humanizing the data is essential for participation and meaningful storytelling.

Sensor journalism is still in a state of limbo. The benefits of being able to collect your own data are huge, but there are also big considerations that need to be made in order to ensure both journalistic and scientific integrity. It is essential, at this point in the history of sensor journalism, to continue to facilitate discussion around the use of sensors and to increase awareness about what is at stake. There are huge opportunities here, but only if we, as journalists, do not take shortcuts and sacrifice journalistic standards.

_Pitt, Fergus. Sensors and Journalism. Rep. Tow Center for Digital Journalism, May 2014. Web. 05 Oct. 2014.

Waite, Matt. "How Sensor Journalism Can Help Us Create Data, Improve Our Storytelling." Poynter, 17 Apr. 2013. Web. 05 Oct. 2014._


0 Comments

Login to comment.