A Look at the First CTTM Data and What We Can Do With It

Before the holidays, students working with Ed Lindsey and Chuck Neeley at Old Town High School collected water samples from taps in about 30 buildings around Old Town, Maine. They entered data about water chemistry, total dissolved solids, and visible evidence of water quality issues into the Anecdata repository. This post provides a peek at what they might find in when they look at the data in Tuva.

One goal of this post is to provide other CTTM teachers with a better idea of the kinds of things that students might do with CTTM data when using Tuva, which is the data analysis and visualization platform that we are using. Another goal is to stimulate a conversation among the pilot program teachers about how the data we are collecting and analyses we can do in Tuva support teaching and learning.

So, this post is out ahead of the OTHS students a bit. They are seeing their data in Tuva for the first time this week and next. They will need some time to make sense out of it in their own way. My intent here is to provide some examples of what is possible. Over time we will figure out how “possible” relates to “educationally useful.”


Teachers have told us that many students have only a weak conception of their community as a physical place with hills, rivers, quarries, and other features that shape how the community has developed. One of the reasons that we initiated the CTTM project is that it provides students with opportunities to think about their communities geographically and spatially. Seeing the data on a map is an important way to stimulate conversations that deepen students’ thinking about their place. The map at the top of this post shows how the different kinds of water stains that students observed on sinks and other fixtures are distributed around the community. Below is another map. This one records the number of water quality issues (taste, odor, color, stains, particles) recorded at the different sites that students sampled.

Fig. 2. Spatial distribution of water quality issues.

And, here is a map that shows how readings of total dissolved solids vary around the community.

map of TDS measurments around Old Town
Fig. 3. Spatial distribution of TDS readings.

Class conversations about maps such as these are a good starting point for creating lists of questions that students can go on to explore. Some of these explorations might involve digging more deeply into the chemistry behind the data. Such work will be easier if teachers and students can meet in person and use lab equipment. Other questions might be explored by digging more deeply into the data.


In addition to looking at their data spatially, students can explore their findings with different kinds of graphs. For example, here is a look at the distribution of TDS values that, when coupled with the map in Figure 3, might stimulate a deeper conversation about what the students found out about total dissolved solids.

Box plot of TDS values
Fig. 4. Boxplot showing variation in TDS values.

The students might recall that they collected information about whether the water coming from a tap is filtered. They could explore the impact of filtering water using a graph such as the one below.

Graph of TDS from filtered and unfiltered taps
Fig. 5. Comparison of TDS values in samples from filtered and unfiltered taps.

It would be interesting to see what students would do with information. One possibility would be that they could dig more deeply into the kinds of filters being used, how often they are changed, and so on.

Returning to the counts of water quality issues that students observed, presented spatially in Figure 2, here is a graphical presentation of how the counts vary across the sampled buildings.

Number of water quality issues observed across all buildings
Fig. 6. The number of water quality issues observed for each building.

Putting CTTM into a Broader Science Education Context

CTTM is designed to provide exploratory data. This makes it different from other scientific work that students do in school. Much of the work in science labs is aimed at answering a question — at closure. The data we collect in CTTM is intended to generate new questions.

It might be useful to engage students in thinking about the difference between the kinds of data they collect in CTTM and in other scientific work. What is different here? Students might notice that the cost of collecting and analyzing each sample is very low when compared to sending samples out to a laboratory for analysis. What are the good things that this low cost enables? What are the limitations that are introduced by this design? How should we use CTTM to get the most out of what is good about it and not get tripped up by what it cannot do?

Students probably have the idea that science depends on careful record-keeping and being able to say things with certainty. What parts of the work in CTTM are careful and what are the things we can say with reasonable certainty? Which parts are less certain? How do these pieces fit together, or not?

If time permits and the dynamics within the class will support it, it might be worthwhile to move the conversation beyond CTTM to look at the broader question of where exploratory investigations fit into the bigger picture of what science does. Can they think of important problems where exploratory work could be useful?

Next Steps

Understanding that CTTM is exploratory, what comes next?

This might be a useful question to explore with students once they have made some progress in making sense of the data they collected. This could be a particularly interesting question if their work turns up evidence of problems in some buildings or in some parts of the community. “What’s next?” could also build on questions that emerged from their analysis.

It is also a good question for us as educators. Maybe one that we can pursue together later this spring or next summer.

— Bill

2 thoughts on “A Look at the First CTTM Data and What We Can Do With It

  1. Molly Schauffler

    Hi Bill — It’s good to see such rich opportunity for true exploration, not so much to answer specific questions yet, and without need to worry about a ‘right answer’, but to raise questions and determine next steps as a class, as you pointed out. Students will have to make some decisions about how to proceed to check hunches, verify tenuous patterns, all to start building a case that zeros in on potential real problems.

    The dataset is also ripe for exploring many possible associations between categorical factors using a bunch of different two-way tables, which I don’t see students using often. Take a look at the association between the number of issues (N_Issues, which I though at first was Nitrogen Issues, but figured out that it is a summary statistic counting the number of water issues detected in a sample (?) by Year of Construction. If you make Year of Construction and N_Issues both numeric categories (Decades for Year of Construction and bin size 2 for N_Issues). What would students think about this plot? What conjectures might arise and how could they be further tested?

    Thanks, Bill — a neat project and I look forward to seeing the data roll in!

  2. Pingback: Using CTTM to Introduce Core Data Literacy Concepts | CSI-Maine

Leave a Reply