I admit that when I signed up for this Data Visualization class I wasn’t sure what I was getting into—yes, I’d get to make a few charts and learn how to implement data into my reporting, but beyond that, I didn’t know the full scope of data journalism. I still don’t; I’m still learning, though I’m closer than I was before. Through listening to talks by Lily Bui and Patrick Herron, reading the Tow Center for Digital Journalism’s report on sensor journalism, and completing a water conductivity workshop in class, I have come to see the benefits of sensor journalism as well as its limitations.

Sensor journalism is something that, in hindsight, seems obvious, but I had never heard of it before this class. Sensor journalism is, as the Tow Center for Digital Journalism puts it, an attempt “to extend the reach of our five natural senses.” It involves the use of sensors and technology to gather data so we can then analyze it to be used in stories. At first, this didn’t make sense to me—gathering data in this way seemed like a job for scientists, not writers. But there are many benefits of sensor journalism:

1. First, using precise, reliable data lends more credibility to stories. Having the best, freshest data to analyze, visualize, and share with the world is a skill that more and more journalists should acquire, especially as we as a society are becoming more and more comfortable with using computers and technology.

2. The data we need for certain stories may not be readily available. In this case, journalists must have the capacity to gather this data themselves. Journalists are used to gathering data by observation and talking to people—sensor journalism is just another way to gather certain types of data that we may then use in stories to report on trends, call attention to problems, and call for change, like in the case of Patrick Herron’s experience with gathering data on water purity in the Mystic River to show which areas are cleaner than others and examine why that might be.

3. Data gathered by outside sources like scientists and labs may not always be reliable. This is something Patrick Herron pointed out to us during his presentation; he noted one particular example of a lab recently coming under fire for having inaccurate results, which may have affected years of data journalism. By gathering data ourselves with instruments and methods we trust, we can ensure that our numbers are accurate and useful.

For these reasons, sensor journalism is a useful skill to learn, especially as we continue to embrace the use of technology and digital media for reporting. The more I learned about sensor journalism, the more I realized all the ways it is helpful. I was particularly interested in Patrick Herron’s experience with the Mystic River—the data, and the way he presented it to us, made me genuinely interested in the process of gathering the data, as well as the outcome and the way the data was used to bring awareness to the Mystic River’s water quality.

We were given an opportunity to actually use sensor journalism in class in a workshop on water conductivity. In the workshop, we put together a Coqui, which is a simple device that can measure things like water conductivity, light, and temperature and then converts the measurement to an audio tone that plays through a speaker on the device. The beauty of the Coqui is that it measures the same data that very expensive, heavy-duty instruments used by sensor journalists and scientists measure, but it’s open-source, much less expensive, and relatively easy to put together with instructions that can be found online.

Once we put together the Coqui, we used it to test the water conductivity of several different water sources that we had gathered: water from the Charles River, tap water from our apartments and different locations on campus, and a few different brands of bottled water. When we used the Coqui on the different water samples, it generated an audio tone for each one; the higher the pitch, the more conductive the water was—or, the more “not water” was in the water, as our professor put it. Unsurprisingly, the samples from the Charles River had particularly high pitches, compared to most of the tap water. Two of the bottled waters, Evian and Pellegrino (a type of sparkling water), also generated higher pitches. This discovery caused us to think about the different reasons certain types of water might be more conductive than others. Originally, we assumed the most conductive waters would be the dirtiest, but we then had to consider the fact that minerals in water—so, decidedly not dirty water—could also cause a higher conductivity.

The outcome of the workshop caused me to think about aspects of sensor journalism I hadn’t considered before, including some of the drawbacks and complications of this method of gathering data. For one thing, this workshop confirmed that sensor journalism can help drive the stories we want to write; I went into the workshop already assuming that we would conclude that water from the Charles River isn’t the cleanest water around, and that’s exactly what we were able to confirm. But the discovery that mineral water is perhaps just as conductive as the dirty river water pointed out the flaws in the experiment:

1. Water conductivity does not necessarily equate water cleanliness. When gathering data, it is important to know exactly what data we are gathering and what it is actually able to prove.

2. The pitch generated from the Coqui is not the most accurate or helpful way to quantify water conductivity—it is certainly interesting, and the audio output allowed me to have a more immediate reaction to the data, because comparing pitches is easier for someone inexperienced in studying water conductivity than comparing numbers would have been, but when it comes to reporting on the data, it is difficult to prove anything when all you have to rely on is different screeching noises coming from a small instrument.

After the experiment, we discussed the Coqui and how useful (or not!) it was for us as a class. We talked about upcoming developments to the instrument that could make it more useful—like outputting actual numbers as well as a pitch, which would make the instrument more useful for reporting purposes.

Another flaw with the Coqui, and with sensor journalism in general, is that it may not always be accurate. The Coqui is great because it is open-source and you can put it together yourself, but that opens it up to plenty of human error, which could skew data. This is why many sensor journalists, like Patrick Herron, use much more expensive and complicated instruments to gather data. But even with instruments like these, we cannot always 100 percent trust the data because it relies on the instrument being correctly assembled and calibrated.

Sensor journalism is a new frontier of journalism that we should all strive to learn, and I’m glad to have had the opportunity to explore it in class. However, we should also be skeptical and aware of all the potential flaws of sensor journalism. In my opinion, these flaws need to be addressed before we let ourselves rely too much on sensor journalism. That being said, I look forward to seeing the continued rise of sensor journalism and data visualization.