Thanks James! It's great to be doing this important work together and thanks for sharing it with others. I will do the same. I just watched your video too and it is very inspiring! Great to see those students in action to help organizations reduce their carbon footprint --and to show such leadership and passion. It looks like a fantastic project that is helping students build many skills.  Enjoy the showcase!

Hi Lexi, thanks for sharing!  It's great to see the many different ways we're learning about activating STEM interest and identity.  

Thank you for your question, Ashley.
Other than the learning curve challenge, we faced difficulties translating real storm sceneries into middle school level learning activities. The storms can be affected by many variables (temperature, wind speed, pressure, altitude, geographic positioning, etc.) and have unique circumstances. Iteratively working with scientists and teachers has been helpful to focus on days/times that will show trends with a couple/multiple variables. The other challenge was also navigating a hands-on science classroom culture (you can see more detailed Q & A on this below).

This is so interesting! I'm curious if you have noticed whether students have had more trouble reading and interpreting certain types of graphs? We are developing data science activities for preschool and so it'd be interesting to see if providing certain types of data displays or visualizations could scaffold their ability to decipher graphs as they get older and into middle school.

Thanks for your question, Jess! Students spend the most time in our units creating and examining time series graphs as well as dot plots of weather components such as hourly (or daily or monthly) temperatures, winds speeds, precipitation levels, air pressure levels, etc. During an earlier iteration of our materials, we tried introducing dot plots first, because students in theory should have seen these in elementary school (or in preschool, if they had used your apps). ;)  In the end we introduced time series graphs first, because students found them less abstract and more intuitive to understand. In our weather context, it was easier for students to make sense of the fluctuating patterns of hourly temperatures over the course of two days (for example) than to grapple with the distribution of the same temperature measurements in a dot plot. Most of our students did not have much prior experience with any type of graphing. It would be interesting to do more research on whether time series graphs are generally easier to understand than dot plots (especially when working with weather data, which has an inherent temporal component).  

Hi Anne,

Thanks for visiting our video, all your support, and your question. COVID delayed our testing and development schedule – it hit just as teachers were about to implement the first iteration of our first unit (in the spring of 2020). We had to wait until fall 2020 before teachers’ school lives had settled and they were willing to test new materials like ours.

COVID also had some effects on the design of our materials and how our lessons played out in classrooms. For example, we had to eliminate an activity in which students learn about air pressure and wind by blowing up and releasing air from balloons in the classroom (sadly, we quickly realized how impossible that activity was during COVID!).

In some classrooms, it took much longer for teachers to complete individual units because the pandemic would send entire classes home for a week or two of remote instruction during peak infection rates, or multiple students would be quarantined at home for many days. We also designed our lessons with the intention that students would investigate data and discuss weather and climate concepts in pairs and small groups. Unfortunately, for large stretches of our unit testing, students had to sit masked at socially distanced desks/tables, sometimes behind plexiglass barriers, investigating data individually, with discourse typically volleying back and forth between the teacher and students and not so much among students. Conditions have certainly improved this spring, but most of our data collection has occurred under classroom conditions that we never imagined when we began.

Thank you for your comments and questions on our video Anne. In addition to impacts on classroom implementation, COVID conditions impacted the way we designed our professional learning sessions and research data collection process. We iteratively revised our professional learning design from the long day sessions we held at the Mount Washington observatory. We relied on shorter evening/late afternoon sessions and utilized virtual ways of engagement. Since most of our data is collected in virtual environments, we have learned effective set-ups and resources for our participants and for us. 

Hi Joi,

Thanks for your questions, and for providing us the opportunity to clarify our project context! Yes, I neglected to make clear in our intro that our units are designed for middle-school science classes. Over the course of two (COVID-filled) years, we have tested two iterations of each unit with between 4-6 science teachers and what’s now approaching 500 students. Although we originally intended to target our units for grade 8 science classes, our teacher collaborators have used them with students in grades 6-8. Many of our partner teachers are the only science teacher in their rural middle school and teach multiple grades – often covering a weather unit before grade 8. Some teachers have used our units with all of their classes across each grade level – because they wanted to give all of them the opportunity to work and learn with authentic weather data from their local area and from Mount Washington!

You pose another great question about how data science is seen as a topic in our classroom settings. Interestingly, some have shared with us that they think it’s critically important and exciting to give their students opportunities to learn to analyze data in their science classes. Some of these teachers previously taught mathematics or have worked with data in their own college/graduate school days and have seemed more comfortable with the data concepts.

Other teachers appear to have been attracted to our work for its inquiry-oriented nature and the opportunity to connect with scientists at Mount Washington. In interviews and written reflections, some of these teachers felt that there may have been too much “math” and data in our materials and asked: “Where is the science?” Although we intersperse data investigations with lessons that focus on learning weather and climate concepts (to help students explain data patterns and the causes of extreme weather events), it seems that our extended data investigations didn’t feel like “science” learning for some teachers. We have found this reaction very interesting and may try to write about it – and its implications for science education.

Finally (sorry for the long response, but you asked such great questions!), yes, conversations about climate change do arise. Teachers have relayed (and we have observed) that when students share out findings from their interviews with family/community members about their experiences with weather, a recurrent theme emerges: temperatures have gotten warmer and snowfall has diminished over time. Family and community members are noticing these patterns, and students are learning about these changes from these trusted contacts. Our lessons do not focus on the causes of climate change, and we don’t currently have information on whether and how teachers discuss the causes. (I’m thinking that we should add this question to our interview protocol as we wrap up our final conversations with teachers!)

Hi Joan,

Thanks so much for your great questions. Yes, our research questions focus on examining whether and to what extent students demonstrate growth in their understanding of basic data concepts (as well as their academic and career interests in working with data), using established pre- and post-assessment measures. We are not trying to measure quantitative growth in student learning of specific weather or climate concepts, but we have collected all the written work products and graphs that students create in CODAP as they progress through each lesson. Questions that we are interested in exploring include how students describe “extreme” weather, what may cause changes in weather, and how students’ descriptions/understandings evolve over the course of each unit.

You then ask some interesting questions about students’ interviews with family/community members. We don’t know how interviewees feel about the activity – but you’ve sparked the idea that perhaps we should find out! Teachers have told us that when parents learn about our project, some have said that they wish they could participate in WeatherX too.

In focus groups and surveys, students have told us that one of their favorite activities in WeatherX is the interview with a family/community member and sharing the stories they’ve heard in class. We provide students with a set of questions that they can use and adapt for their interviews, and in an early lesson, they practice conducting the interview with a classmate before conducting the interview for real. We designed a structured activity for students to share their interview results in class through both written responses to prompts on chart paper and whole-class discussion, but we haven’t considered a way to share results back with family/community members. This is another great idea!

Thank you, Joan, for your comment and questions on our video. Adding to Jo's response on using students' artifacts to look at their weather and climate conceptions, student focus group interviews also showed how some students developed an improved understanding of weather components. For example, they realize how the temperature is typically low in the evening, locations that are closer to the lake, etc. This understanding helps them understand why weather observers measure temperature at certain spots and during certain times. Therefore, these preliminary qualitative finds can inform how using data practices in science classrooms can help students learn how to do science in complex weather and climate science topics. More on climate and climate change, Jo mentioned that we are looking at conceptions of "extreme weather." Our focus has been on resources students use to describe/explain extreme weather, their positioning, and the references to spatiotemporal practices. One thing that evolved from that study was to compare students' conceptions to those of scientists, our interdisciplinary project team, and teachers. These comparisons help us understand where confusion and uncertainties might occur in learning about extreme weather.

Hi Tichina,

Thanks for your great question. We had originally considered organizing community "science nights" at our local schools (which often serve as community centers in rural areas) to gather local knowledge and input into the design of our weather units. But COVID nixed all that. Instead, our collaborating teachers were major sources of initial input into the weather events for students' data investigations. We worked with local teachers from the very beginning to learn more about their personal weather experiences in the local areas. Our project partners at Mount Washington Observatory and at the University of Maine also live in the areas where we are working. In addition, As students began interviewing family and community members about their local weather experiences, we took that information to inform later lesson revisions. As you can see, our development process was very place-based, which is now forcing us to think creatively (and to consult existing research) on how to transfer our learning materials to other locations across the country. Is it possible to listen deeply within local communities and do truly collaborative place-based work that engages students in locally relevant ways, at scale?

Hi Ravanasamudram,

Thanks for checking out our video and for your questions! We recruited participants through multiple channels. I'm at EDC, and our organization has deep knowledge of and strong connections with school districts and rural regional networks throughout New England. Our project partners at Mount Washington Observatory and the University of Maine also have extensive connections with schools and educators in the region. We were able to tap into these networks to advertise and recruit for participants. Most (but not all) of our teacher recruits have been the only science teacher in their school for grades 6-8.

We did not intend for our units to take 3 weeks to implement; our original goal was 2 weeks. There have been a couple cases where teachers finished each of our units in two weeks - either because they omitted some lessons, or their classes were much faster in understanding and completing the materials. (Some of our research will try to probe why there has been such variability among schools and classrooms.) In other cases, teachers took three weeks or more to complete a single unit. Although all of our teachers worked to implement our materials during consecutive days and weeks, those who took longer to complete the units reported frequent disruptions (Covid-related student or teacher absences, or shifts to remote learning; other scheduled field trips or school events, etc.).

As for science and mathematics teacher collaboration: We never stipulated that such collaboration was a requirement, but we had hoped some teacher pairs might sign up. In fact, with our first teacher cohort, we in fact did have such a pair. Sadly, Covid hit right after we had provided them with PD and they weren't able to implement the unit together (which is what they had planned). None of the other participating science teachers have worked with mathematics teachers to deliver the units, but several of them have taught mathematics previously. We also designed the lessons to ensure that the statistics and data practices would be accessible for students in the middle grades, including concepts that are in the Common Core State Standards for mathematics in the elementary and early middle grades (and providing lots of scaffolds and supports for those students who, sadly, hadn't seen or weren't strong in those concepts and practices, like basic graphing).

Hi Leigh,

Thanks for watching our video, and for being a great friend to our project! Yes, it has been very interesting to discover that some of our teachers felt that there was too much data and not enough "science" in our units. Within our project team, which includes former science teachers, mathematics teachers, researchers, etc., we've certainly worked from the start to develop lessons and units that have a balance of CODAP-based data investigations and other types of learning activities (e.g., hands-on examination of warm and cold water masses to provide insights into hot and cold air mass movement; community interviews; outdoor weather observations; short videos and readings; etc.). But for some teachers, we didn't seem to get the balance right, particularly in our earlier unit iterations. We are indeed very intrigued by this finding, and its implications for science teacher education and practice, especially as big data enters ever more arenas of science practice. 

As for students, we learned early on from surveys and focus groups that students start complaining when they spend too many consecutive days looking at data on a screen - even when the tool is interactive and dynamic like CODAP. When we've asked students how we might improve the units, they inevitably say they want more "hands-on" activities. When we ask for examples, we've heard answers like "exploring with slinkys" or "slime!". We are in the process of analyzing our pre/post data for evidence of student learning and growth in data analysis interests, but whatever our findings, it's quite possible that we haven't yet designed data investigations to out-compete slime.

100% consistent with what we hear, Jo. 

Thanks, John.

There are a bunch of contributors to that learning curve that range across several areas. Few of them have a major impact on the learning, but all can cause (unnecessary) confusion, especially to students (and teachers) who haven't had much experience working with real, uncurated data sets.

The primary issue comes down to selecting the right site to get data from and being prepared for any inconsistencies that may arise:

• Selecting a station. There are hundreds of weather stations that report data. Not all collect (or report) the same data. Some, for instance, will have monthly and daily data, but no hourly data. Some will have historical data, but nothing recent. So making a plan to look at data from a particular site requires going down a checklist to make sure that you don't start planning on one site, only to find later in the unit that the chosen site doesn't have the additional data you need to continue on. We've been working with Concord Consortium to tweak the portal that you use to select data to make it easier to see in advance what you're getting.
• Understanding how weather data is reported. A few peculiarities have tripped us up from time to time during the development, such as hourly data can include "special report" data records, which don't always include all variables (and can mess up the count of records... if you're pulling hourly data for a 3-day period, you expect 72 records, but sometimes get more). Another finding was that, for instance, daily reports include three temperature readings: minimum, maximum, and average, but hourly reports only show a single measure (the temperature at time of reading). Further, the norm for weather scientists is to report the average as a simple mean of max and min. So there's potential for inconsistencies, especially if you have CODAP calculate statistics (max, min, mean) off the hourly temperature readings and compare them to the daily data. In most cases it's minor, but if you're exploring around extreme events, the differences could be more significant.

The local unit definitely has potential to work in other communities, and we will be exploring how to address the questions above and provide supports for that to happen. We're also curious about the Mount Washington focused lessons, and whether that location has more universal interest, or whether we can partner with other institutions at different sites regionally to tailor that part of the curriculum as well.

Hi John,

Thanks for checking out our video! I love your video and the work you are doing in Colorado.

Kevin provided more background on some of the challenges that we've tackled to bring a wealth of NOAA data to students in ways that are locally relevant, engaging, and tractable to support key learning objectives within available timeframes in science classrooms. Because there are so many weather stations whose data students can investigate, and because different stations report data with different sets of weather variables and timeframes, we ended up curating data sets for students to ensure they could focus limited classroom time on making sense of data patterns rather than cleaning data or looking for stations with more complete data.

We have created some extension activities, however, for some teachers with students who proceed through our materials much faster than their peers, to allow them to explore data from weather stations anywhere in the country and to hunt for and do follow-up research on extreme weather events. We'll find out how students respond when some of them are set free to explore the data in the NOAA weather portal!

In addition, you asked what students have done with the data stories they have collected. I will assume you're referring to the interviews that students conducted with family/community members about their weather experiences. After students conducted their interviews, their assignment was to summarize what they heard in their own written reflections, and teachers led a structured lesson where students shared their findings by writing responses to specific prompts on chart paper (e.g., Who did you interview? What did you learn about their work and activities and how they are affected by weather?). Students conducted a gallery walk to read and comment on what their peers wrote, and teachers led whole-group discussions of students' findings afterwards. We haven't structured into our lessons any activities where students might do more with their stories and findings, but your question (like another comment above) raises the excellent point that there may be opportunities for more synthesis and sharing -- particularly back to the community -- if teachers feel there is time and it is an activity they can work into their schedules. The students loved learning about weather experiences through the people they interviewed -- I can envision how meaningful it could be to bring students and families/community members together for joint sharing and learning.

Hi Kristin,

Thanks for watching our video, and for your comments! I am inspired by the passion in your project and video. :) In fact, there is lots of overlap between your project and our other project, which we shared in a video last year: https://stemforall2021.videohall.com/presentati...

For our WeatherX project, we've been working with teachers and students in rural middle school science classes. We're not aware of too many electives in our partner schools, and we're not tracking students' coursetaking patterns. However, we have administered pre- and post-surveys and assessments to students, and we will be examining whether students' academic and career interests in working with data increases after completing our units. We know from student focus groups that many students have said they feel much more confident with graphing and data analysis after completing WeatherX lessons. We know from our other project (shown in the video link above), which focuses on high school students in non-AP mathematics and statistics classes, that students displayed statistically significant growth in their learning and interests in analyzing data after completing our statistics/data literacy units on social justice issues. We're itching to finish data collection and to start data analysis for WeatherX to see if we have similar findings from our rural middle-school youth.

Thank you for sharing your video from last year. I am not sure how I overlooked it last year. I did not see a link to your resources for that project. Can you share it?

We have found much the same in terms of increase in student attitudes to statistics, interest in follow-up courses, and their confidence toward different concepts and processes.

Hi Kristin -- Sure, here is a link to our website for the social statistics project, where we have released one full module and will release a second one this summer: https://sites.google.com/view/uss-data/home

We will also be releasing our two WeatherX units later this year. In the meantime, the Microdata Portal that provides access to U.S. decennial census and American Community Survey data (from 1850-2017), and the NOAA Weather data portal that provides access to weather data from across the U.S. (spanning decades), are both free plug-ins within CODAP (built by our partners at the Concord Consortium) for anyone to use at codap.concord.org.

Thank you!