Teaching Students to Analyze Data

Author: Arlene Vinion Dubiel

“ Data analysis is the intersection of science and math, purposely integrating both subjects.”

I was surprised when engaging my pre-service elementary teachers in an activity using radioactive decay to determine the age of the Earth. My students did not know how to design a data table. After getting over my initial shock, I took a step back and talked them through the process of how to design and organize data tables. Likewise, I often needed to explain to in-service teachers when to use different types of graphs depending on the question and data. But, it wasn’t until I listened to Diane Johnson, Master Teacher with MSUTeach at Morehead State University in Kentucky, that I realized not knowing how to analyze data is a universal issue and that we, as instructors in STEM, can and should do to teach our students how to analyze data. 

At the National Science Teaching Association (NSTA) Engage 2020 virtual conference, Diane shared how her preservice secondary math and science teachers were challenged in how to analyze data in their research methods class. This led her to make it a mission to teach students to analyze data. Diane’s talk inspired me to also take up this cause, hence this blog. In full disclosure, many of the resources shared here come from Diane’s Engage 2020 conference presentation also entitled “Teaching Students to Analyze Data.” 

Data analysis gap

If we think back to our science labs, we usually engaged what we call cookie-cutter labs. Students conduct investigations to collect data and fill in a data table. Then they are told how to analyze and represent the data in a graph before answering questions that address the meaning of the data. Providing students with directions for analysis is done in the interest of time so more content can be covered. In math courses, students will link data tables, graphs, and equations so they can visualize relationships between variables. The data points exactly fit graphed lines so equations can be determined. But this data is usually dissociated from real-world variables and so the relationships lack meaning. 

 Data analysis is the intersection of science and math purposely integrating both subjects. Because we traditionally separate math and science content in schooling, students lack the opportunities to make decisions with analyzing and representing data. Additionally, it takes time for students to grapple with data and try and sometimes fail with the analysis. With our current system of pushing through content, the argument is that we do not have time to allow students to do this. But it is in grappling with data that deep learning can occur.

Why teach data analysis

Providing students with opportunities to grapple with data is supported through our math and science standards. The first Common Core Math practice is “Make sense of problems and persevere in solving them.” Within the standard, it states that “Mathematically competent students...make conjectures about the form and meaning of the solution and plan a solution pathway. This standard can apply to the use of data with students making conjectures about the form and meaning of data and planning how to use it. The need for data analysis is more obvious in the Next Generation Science Standards with one of the Practices of Science being “Analyze and Interpret Data.” Supporting statements from A Framework for K-12 Science Education, upon which the NGSS are based, include “a major practice of scientists is to organize and interpret data through tabulating, graphing, or statistical analysis” and “data must be presented in a form that can reveal any patterns and relationships.” 

In a previous blog, I shared that the fastest growing occupations are those in mathematics. In particular, there is a need for data analysts. We gather trillions of small pieces of data every day. Every time you make a purchase, do a web search, or use an app on your phone, data are gathered and recorded. Gathering data is the easy part. It is what to DO with that data that we need to prepare our students for.  

Begin with questions

If these arguments have convinced you that students need opportunities to grapple with data, then the question becomes, how do we do this? How do we address data analysis skills and scaffold our instruction so our students are able to engage in these practices? 

Good lessons begin with good questions. While we eventually want students to ask questions themselves, we must first model this practice by asking good questions that can be answered through data analysis. For the intersection of science and math, we want to ask questions about the relationships between variables. For example, how does early social environment affect aggressiveness in adult bees? Starting students off by providing them with the question gives them a clear directive of what to do. When they get lost in the weeds, you can point them back to that question. Some resources asking good questions are found in a previous blog. 

Authentic data sources

Once we have the question, we need to let students grapple with the data to answer that question. As our focus here is on analysis rather than investigation, it is best to provide students with an authentic dataset rather than have students analyze their own data from an investigation.  Ideally, the datasets we provide are not perfect. The messier the data, the more questions students have and the more thinking, engagement, and learning that can occur. 

Data Nuggets is an excellent resource that uses data from authentic investigations. Lesson plans for educators elementary through post-secondary are provided along with levels to scaffold graphing skills. With these lesson plans, students must make decisions on how to analyze and then represent the data. The example question above comes from the Data Nugget “To bee or not to bee aggressive.” Data Nuggets is a good place to start if you are new to teaching with data.

If you feel comfortable teaching data analysis and are looking for datasets rather than lesson plans, the Science Education Resource Center (SERC) at Carleton College has curated locations for data sources. Tips and training for supporting educators as they teach with data are also found on this site. 

Authentic datasets for climate data with classroom ready data sources can be found through the National Oceanic and Atmospheric Administration (NOAA). National Geographic is another resource for climate data with lesson plans that include datasets. If you are teaching statistics or social science, the ICPSR is the go-to site for archived survey data. This resource is part of the Institute for Social Research at the University of Michigan. This site is for advanced students who are capable of analyzing very large datasets of over 1,000 cases or more. 

There are a myriad of resources involving the agent of our current pandemic found on Data Spire. Other data sources on Covid-19 were also shared in a previous blog. As the pandemic is ongoing, more data is added each day. Finding your favorite Covid-19 data source now could be a benefit in future years as we look back to deepen our analysis and understanding of the disease and its spread. 

We need to support, not direct our students on how to analyze data. Providing students with opportunities to try, and fail, with analyzing and representing data leads to deep learning that can prepare them for our data-rich world. 

Students need more opportunities to make data analysis decisions. To scaffold this opportunity, give students datasets and ask them a focus question to answer using that dataset. Check out some of the resources provided here to begin. 

Ready To Jumpstart Your Math Redesign Momentum?

Create effective, equitable & quantifiable change in your math program built on the good work you've already done