Example Field Notes

SDS 237 Fall ’24: Sarah Thieler

My background knowledge for this writing is a combination of concepts taught and discussed in AMS 215ir - Indigenous Climate Resiliency, as well as my experience growing up on Wampanoag land and interacting with the tribal community in Falmouth, Massachusetts.

On September 29, 2024, I participated in a field survey in Rockport, Massachusetts, collecting data about and specimens of the invasive European green crab (Carcinus maenas) and Asian shore crab (Hemigrapsus sanguineus) along the intertidal zone. The rocky shoreline, shaped over millennia by the waves of the Atlantic, now bears the ecological imprint of human interference and intervention - an ecosystem forever altered by the global movement and migration of species. As we waited for low tide, my team of four donned our waterproof gear and prepared our collection buckets with Ascophyllum, our boots filling with water as we navigated the slippery, algae-covered rocks. Once the water level was right, we turned over stones and scavenged through seaweed to collect our specimens.

The intertidal zone where we conducted our work is not simply a site for our own ecological interest - it is a landscape with a deep historical and cultural meaning. Before the focus of the coastline became invasive species and climate-change induced erosion, these lands and waters were home to the Pawtucket and Wampanoag people, who for generations practiced environmental stewardship including an understanding of the cyclical nature of seasons, species availability, and the balance of ecosystems. Yet, the data collection we prioritized on our trip - counting population numbers, measuring carapace widths, and later logging these numbers into digital spreadsheets - are part of our Western knowledge system that emphasize that data must be quantifiable and measurable to be valid and valuable. This approach greatly contrasts with Traditional Ecological Knowledge (TEK), that I had learned about previously in classes at Smith and through my own experience growing up on Wampanoag land.

We continued to categorize and collect these crabs into our buckets. I later reflected on the history of species invasion - how settler colonialism and species migration mirrored each other in this particular ecosystem. The arrival of European settlers to this very area, like the arrival of these invasive crabs, disrupted Indigenous life and communities as they were known. This thought has stayed with me; how were the methods we used (counting, sexing, categorizing) a reflection of colonial frameworks? How are we imposing a certain kind of order and boundary to an ecosystem that can’t respond to our categorization?

There has also been ongoing discussion about the terms we use to define species. “Invasive” species are the enemies, and we must fight to protect our land and resources from them. Ecosystems are war-torn and fought over to be either won or lost. This framing and metaphor suggests that environmental management as a whole is about defeat and submission rather than creating balance, hiding the possibility of coadaptation and coexistence. These metaphors carry heavy assumptions - that numerical, Western data holds the solution to fix the environment.

This data environment is clearly underpinned by a master narrative that supports Western scientific intervention as the most important tool for restoring ecosystem health. This narrative is supported by rewards for data collection and policy proposals through grants and continued funded research. This narrative obscures the more broad historical and cultural aspects of changing environments, putting the weight on “proper” ecosystem management systems supported by numbers. The labor of those who came before us, who maintained ecosystems without highly advanced technology and grants from the NSF, are effectively rendered invisible.

When I think about my role in this process, I have to think about the tension and difficulty between contributing to the systems of data collection that prioritize Western approaches and overlook the more historical methods. While I think the research I am conducting is valuable, I am also aware that my work is part of a long legacy of colonial scientific frameworks. As researchers, we are complicit in this system. Yet we must also question it.

SDS 237 Fall ’23: Vivian Wei

In my economics of crime seminar, I’m conducting a semester-long research project exploring the impact of recreational marijuana legalization on intimate partner violence. This week, our task is to compile a summary statistics table. However, I ran into a significant challenge because I couldn’t find readily-made, reliable, and comprehensive data on the specific dates of marijuana legalization in different states.

After extensive searching, I found procon.org, a nonprofit organization that compiles information on controversial issues. They provide a list of marijuana legalization dates sourced from official state websites. However, given the structure of the webpage, I need to convert the legalization date information into a binary variable so that I can integrate it with the National Incident-Based Reporting System (NIBRS) dataset.

This seemingly simple task presents multiple classification challenges. For instance, categorizing the year a state passed the law and the following years as ‘1,’ with any year prior as ‘0,’ oversimplifies the situation. It assumes all laws take effect on the first day of the year, disregarding the fact that some states pass laws in various months.

Alternatively, categorizing only the year following the effective date as ‘1’ may under-represent the time span of legalization. Judgments are also needed to determine which year should count as the legalization date, especially when states have different effective dates for marijuana possession and resale. Different states having varying possession caps further complicates the issue. Reducing complex legalization conditions to a binary variable overlooks critical details, forcing me to make assumptions that don’t capture the real-world complexity.

This situation underscores the importance of infrastructure. To conduct meaningful statistical analysis, I must create some kind of binary that is compatible with the structure of the more complicated datasets to compare and contrast the state that legalized marijuana from those that didn’t. This involves relying on research staff and website editors from procon.org to establish the necessary infrastructure.

This experience also closely ties to the concept of translation. Converting intricate marijuana legalization conditions into a simplified binary format requires me to make judgment calls and assumptions, simplifications that are necessary to construct the research infrastructure but may not fully represent the real-world complexity. Given the current political landscape surrounding marijuana legalization, the dataset’s construction choices, if not explicitly detailed, may be misleading for varying interpretations and unintended usage.

SDS 237 Spring ’23: Casey MacGibbon

This entry documents a data environment Casey MacGibbon observed on 2023-02-10 in The Human Performance Labratory, Scott Gym, Smith College. The observations were written up on 2023-02-19.

My stopwatch hits 20:00:00.

“Alright, it’s been five minutes since I last asked, how rigorous do you feel your exercise is on a scale of 6-20?” I say, holding up a reference board for the participant. She points to a 12, which is labeled as “Somewhat hard.” She is on minute twenty of her acute bout of exercise, where she is walking at a moderate pace on the treadmill, and we are monitoring her heart rate from an electronic wristband.

Recently, I have begun research work and analysis for the Witkowski Vascular Function Lab, and this was the first time I went in to observe a visit with a participant. I am new to the lab, and my perspective still feels a bit like an outsiders’. After the visit, I asked Dr. Witkowski explained that each number on the reference board corresponded to a heart rate by multiplying the number by 10, essentially acting as a participant’s heart rate experience. She explained to me how there are two different “data streams” throughout the lab, one being objective and the other subjective.

The idea of objective/subjective data is an example of a binary opposition: a pair of terms with opposite meanings. I noticed how much more “objective” data collection was valued, while a lot of the data collected from the participant was thought of as “subjective,” and thought of to be more subject to inaccuracy. Data is categorized as one or the other.

No data can be subjective, because the mere decision to begin collecting data means that there are expectations and biases put into the collection of data. Though the target heart rate is an “objective” part of the study because it is numerical, does not mean it is without bias. Heart rate calculations have historically ignored weight, height, average physical activity, and many other factors. This is just one example of how all of the data is biased, and cannot be placed into these categorizations. Additionally, the “data streams” figure of speech is a part of popular data discourse, contributing to the idea that data is a natural resource waiting to be found, untainted and untouched by human interference.

Discourse is the way that people talk about and engage with a certain thing/person, and dominant discourse can often characterize how we think about this topic as a society. The imagery of a data stream indicates that data is natural and is something we can “drink” from at any time. In reality, all data is collected from humans, who are limited in their standpoints and views.

SDS 237 Spring ’23: Elm Markert

This entry documents a data environment Elm Markert observed on 2023-03-24 in CMB and email inbox. The observations were written up on 2023-03-29.

Please note that certain details in this entry were anonymized for the purposes of sharing. Names and titles have been changed.

The Center [redacted] is on the first floor of [redacted] down a long hall of labs. I go there regularly for my research project and am always greeted with a friendly face – Morgan. They’re the technical director of the Center and help students with research. This includes teaching students to use instruments that some senior researchers don’t even get their hands on. They’re a patient teacher and helpful during experiments. They’re also a core part of a lot of the data labor (work that goes into creating, supporting, processing, and analyzing information) that happens in my research field.

While Morgan is highly valued on a personal amongst colleagues, on an institutional level, they disappear a little bit. While most people outside the department recognize the research professors do, Morgan fades into the background a little more. They also don’t get to do their own research, and the Center’s budget is generally devoted to making sure professors get what they need for their research. As a result, despite the amount that they help with student research, Morgan rarely gets put on papers as an author.

However, Morgan does love their job. They’re excellent at training people on equipment and prefer being technical support to lecturing. They know the ins and outs of the instruments and do more hands-on work than most professors. Data produced in the Center is as accurate and uncontaminated as possible because of the care that they take when working. The data provides an excellent basis for publications. However, the labor that went into creating that data is hidden in the final publication.

This obscuration of labor says a lot about what kind of labor academia values. Spoiler alert: it’s not “grunt” work. Instead, it values high level thinking. This ignores that the “grunt” work also requires expertise and is vital to the research that professors do. As an undergraduate researcher who gets to be first author but also is doing the “grunt” work, it is easy to see this discrepancy. I do actively benefit from this hierarchy in many ways because of my educational and career trajectory, but it appears that the more one benefits, the harder it is to notice. This is partially because the people who benefit most often don’t do much lab work.

…I will add more as I get permission from former students.