Discussion for Leveraging Authentic Data Across STEM Curricula

The Discussion related to Leveraging Authentic Data Across STEM Curricula is now live! Introduce yourself and let us know your interest in exploring this theme. This discussion will continue through October.
Public Discussion
  • Icon for: Traci Higgins

    Traci Higgins

    Senior Research Associate
    September 2, 2020 | 12:31 p.m.

    Hi all. This is Traci Higgins, affiliated with TERC (in Cambridge, MA), where I have been working with Andee Rubin and team on the Data Clubs project. I'm particularly interested in the question of how students approach and use data especially when interacting with big messy data or multivariate datasets that have been collected and made available by others. What are students initially drawn to in these data sets, what sorts of questions do they think can be answered by data, how do their ideas change as they gain experience using technology to explore such data?

     
    Mark this discussion post as helpful
  • Icon for: Josephine Louie

    Josephine Louie

    Senior Research Scientist
    September 16, 2020 | 12:28 p.m.

    Hi all -- This is an exciting topic and theme for the month. I'm Jo Louie, at EDC, leading two projects to promote students' statistical reasoning and data practices. In one project, Strengthening Data Literacy across the Curriculum (SDLC), we're targeting students in non-AP math/statistics classes with data investigations that explore social justice topics using American Community Survey and U.S. Census data. We have an explicit goal to promote multivariable thinking in this project, which we define as the ability to attend to interaction effects (and awareness of Simpson's paradox) when dealing with more than two variables. We've found that when high school students are thinking about data with a meaningful and relatable context, multivariable thinking seems to come up quite naturally. For example, one of our topics is income inequality and the male-female wage gap. They naturally generate ideas of other variables that might explain the gap (e.g., part-time vs. full-time work, marital status, type of occupations entered). Tools that Bill Finzer has created in CODAP (such as a third variable "splitting function") allow students to explore interaction effects visually, which really helps.

    In our other project, WeatherX, we are working with rural middle school students and examining extreme weather events using NOAA data and data collected by the Mount Washington Observatory in NH. Storms and extreme weather events are highly multivariable, and the relationships among variables are very complex. We are in the early stages of this project, and working with a slightly younger group of students, so it remains to be seen how far we can go in having students make meaning from associations of data with complex interrelationships, and with variables that may be a little more abstract (e.g., air pressure, wind speed).

    + Reply
     
    Mark this discussion post as helpful
  • Icon for: CADRE Team

    CADRE Team

    Researcher
    September 2, 2020 | 05:22 p.m.

    Thanks Traci. Those are great questions. We at CADRE, a network for a PreK-12 research funding program at National Science Foundation, were excited to organize September’s Theme of the Month around Leveraging Authentic Data Across STEM Curricula. We have seen a growing number of NSF-funded projects focused on developing students’ data skills across disciplines and think this is a timely topic. We have a few additional questions for our different audiences. Educators: How are you incorporating data into your classrooms? Researchers: What are some of the challenges you’ve faced in integrating data into curriculum and professional development? Where have you seen the most exciting successes? All: What questions do you have for our webinar’s expert panel?

    + Reply
     
    Mark this discussion post as helpful
  • Icon for: Jacob Sagrans

    Jacob Sagrans

    Researcher
    September 18, 2020 | 05:52 p.m.

    I just watched the recording of this great panel from yesterday. Thank you!

    Since I wasn't able to attend live yesterday to ask questions in the live chat, I'll post here.

    I work with Traci and Andee on the Data Clubs project. We've recently begun having middle school youth explore COVID-19 data and have incorporated discussion about how to do this into our teacher professional development work. We've used CODAP a lot in Data Clubs. I think is a really great tool for getting youth exploring data, because it's intuitive, accessible, customizable, and interactive.

    One thing I'm thinking about now, though, that I hadn't been thinking about as much before the pandemic is how to explore data on maps. The spread of a disease like COVID-19 has a geographical component. I would love to get youth and teachers exploring COVID-19 data on maps, ideally in a way that also is highly accessible and interactive, like CODAP is when it comes to making tables and graphs (a lot of the COVID maps I've seen so far have been static images, or not very accessible).

    I'm curious if any of the panelists or other people who attended or are watching the recording now have experience with guiding youth in exploring data on maps. What have you done? What worked well? What tools/maps/data sets have you used? Have any of you done this with COVID-19 data?

    Thanks!

     
    Mark this discussion post as helpful
  • Icon for: Jessica Sickler

    Jessica Sickler

    Researcher
    September 24, 2020 | 11:32 a.m.

    The Ocean Tracks project was all about engaging students (HS and college) in geospatial data, including tracking data from several marine species and oceanographic data (sea surface temp, net productivity, etc.).  The team created an online interface for interacting with the data and instructional modules to scaffold learning, so you can get a sense of how they approached the task.  https://oceantracks.org/  (And a shout out to Randy Kochevar and Jo Louie, both of whom were part of Ocean Tracks, and may be able to add more.)

     

    In our ongoing Streams of Data project, we’ve been doing some baseline research with students in grades 3-5 to see what skills they naturally bring to interpreting data representations – one of which was a map of 7-day rainfall in the US.  One of my takeaways is that maps offer a very familiar starting point, but that comfort is deceptive, because map-based data often contains really complex information and nuance.  We had a number of young students confidently assert that the map was easy to understand, even when they had clearly not understood what it was showing.  And others who realized their struggles, essentially noting how information-dense it was: “Every single piece of the country is a different color… I think we could get rid of some of the colors.”  Which I think pointed to the importance of supports to scaffold a process of how you look at and reason through map-based data (something Ocean Tracks does quite well, I think).

     
    1
    Mark this discussion post as helpful

    Jacob Sagrans
  • Icon for: Susan Meabh Kelly

    Susan Meabh Kelly

    K-12 Teacher
    September 27, 2020 | 10:45 p.m.

    As additional response to the query of Jacob Sagrans, there is a history of NASA- and NSF-funded efforts to support K-12 students' and teachers' use of geospatial data. Examples that foremost come to mind include NASA-funded Eyes in the Sky II (2010) in which Randy Kochevar, Oceans of Data, had contributed (https://serc.carleton.edu/eyesinthesky2/index.html) and Alec Bodzin's long-standing commitment to supporting classroom integration of geospatial data (https://cadrek12.org/users/alec-bodzin: https://ed.lehigh.edu/faculty/directory/abodzin).                            

    Last week I had a virtual meeting with two of the high school teachers Alec Bodzin has been collaborating with (Shannon Salter (Allentown, PA) and Ian Hanson (Wilmington, DE)) and learned that they had facilitated student COVID-19 data investigations (number of cases/allocation of funds) "on the fly" as a result of their past experience in Alec Bodzin's environmental GIS projects. There is much to be learned from teachers as to their use and foreseen potential for geospatial models, as well as principal investigators with long-standing commitment to facilitating K-12 construction and meaningful use of locally relevant geospatial models.

     
    1
    Mark this discussion post as helpful

    Jacob Sagrans
  • Icon for: Jacob Sagrans

    Jacob Sagrans

    Researcher
    September 28, 2020 | 09:51 a.m.

    Thank you for these very helpful responses! I will check out these resources.

    + Reply
     
    Mark this discussion post as helpful
  • Icon for: BEVERLY SIMMONS

    BEVERLY SIMMONS

    Informal Educator
    September 22, 2020 | 06:35 p.m.

    HI, the InterNational STEM League Competitions and our Ten80 classes/camps have been working with DDD MODS (Data-Driven Design and Data-Driven Decisions Modules) for the past decade.  Some of our students have gone from middle grades all the way to graduation. We have collected data on both rural and inner-city youth that are traditionally underrepresented in STEM courses and careers.  We were one of the four Change the Equation endorsed programs as Exemplary and Scalable.  The most difficult challenge we faced was getting students to see data as predictive rather than explanatory.  In their formal education, they are required to interpret what happened by reading a graph or creating a graph from some data set.  That is very different from what they will need to do as an engineer visiting a limestone kiln in rural AZ in which limestone rocks are put in the 14 story kiln, but the lime that comes out the other end has a chemical composition that renders it useless to the customer.  What skills will be needed by the engineer who is called into a plant creating building blocks that are "suddenly" exhibiting small cracks?  When our engineering team faced these kinds of issues, we were usually called in when on-site teams had not solved the problem so there was often a critical time constraint.  The one thing I heard about our team was that "they can tell more stories from data than anybody we have ever seen."  It is a subtle difference but an important one.  Storytelling from data also means you can see the data that is missing and find a way to collect it to complete the story.  Starting with problems that can be seen, touched, and affected is important as the first step toward DDD.  Students must build a mindset that lets them trust their work with data, trust its predictive power, and trust the story they will weave with more complex issues such as climate change.

    + Reply
     
    Mark this discussion post as helpful
  • Post to the Discussion

    If you have an account, please login before contributing. New visitors may post with email verification.


    For visitors, we require email verification before we will post your comment. When you receive the email click on the verification link and your message will be made visible.



    Name:

    Email:

    Role:
    NOTE: Your email will be kept private and will not be shared with any 3rd parties