Part 1: 3 Small Tweaks to Set Students Up for More Success with Data in the New Year

Data Moves & Topics December 1, 2022

It is hard to believe but we are almost to the end of 2022 (ack, where did the time go?!)

The end of the year comes with a lot of different things – racing to get through the last couple components of our unit, students longingly thinking more about vacation than school, family planning and/or stress for upcoming get togethers…sound familiar?

The end of the year can also be a time to pause, reflect, and consider some adjustments going forward. Why else do you think New Year’s Resolutions are SO popular? Well…we LOVE starting something new, yes. But in order to start something new we need to know what it is we want to change.

So, let’s take a moment to pause and reflect on how our students are doing with data.

>>>>> PAUSE <<<<<

Regardless of where you feel your students are at with data at this point in the year, let’s cover first things first. We absolutely DO NOT need to throw everything out and start new. Nor is it a foregone conclusion that these students “just won’t get it” and we will try again next year.

So in that framing, we want to share three small tweaks that you can use NEXT WEEK regardless of whatever curriculum you are using AND regardless of whatever ways your students are graphing (as a note see “How Are/Should We Make the Graph?” blog post on ways to graph and “Benefits & Limitations of Different Graphing Tools” blog post on our perspective on graphing tools).

We will discuss each tweak and what it could look like in your classroom in this 3-part series.

Tweak #1: Move away from one-&-done graphing

Here’s a story of Joshua Stevens, the lead data visualizer for NASA Earth Observatory. His team wanted to use NASA satellite data to help the general public, policy makers, and farmers understand the extent of the 2022 drought in California’s central valley (2022 was a nasty year for rice growers due to drought and the subsequent water shortages).

Natural-color satellite images from NASA’s Operational Land Imager (OLI) on Landsat 8.

Looking at natural-color satellite images from NASA satellites you can quickly see the difference in how much area is growing plants in September 2021 (image on left) as compared to September 2022 (image on right).

Wow! That is A LOT less green.

But how much less?

Josh’s team wanted to be able to better quantify just how different 2022 was from previous years so they dove further into the satellite data to see what they could find to share and communicate out.

o to review, the NASA team:

1. Had heard of a phenomenon that was happening (high drought conditions + water shortages = less rice growing in California)

2. Could see using the natural-color satellite images that the amount of green in the same area was much less in 2022 than 2023.

3. Wanted to determine how much of an annual change there was in vegetation in 2022 and compare that with previous years.

4. Wanted to be able to communicate this story to the general public in ways that would quickly help people see how much of an annual change there was in 2022 compared to other years (once they figured that out themselves :)).

They decided to use a metric called the “Normalized Difference Vegetation Index” as a way to quantify the amount of healthy plants year to year. Here is what Josh shared with me about the NDVI:

• Uses satellite observations of red (which plants absorb) and infrared (which plants reflect) light to calculate the index.

• Always ranges from -1 to +1

• Higher (positive) values are associated with healthier crops

• Lower (negative) values are associated with poor crop health—or the lack of crops

** If you want to learn more about NDVI, please check out a great article on GISGeography titled “What is NDVI (Normalized Difference Vegetation Index)?”

Now onto the data exploration and graphing!!

After calculating August NDVI for each year, Josh first graphed the NDVI values for each year. As a note, he used August because that is typically when rice crops are quite healthy.

Looking at this graph he saw that 2022 was indeed lower than other years.

But he still wondered what “normal” NDVI was during this time period. And he wondered how 2022 varied from that “normal”.

So, he kept exploring the data to see what he could see further.

Looking at this graph he saw that most years generally hover around 0 (aka the long-term average), some years the NDVI anomaly is a bit higher (more healthy plants) and some years the NDVI anomaly is a bit lower (less healthy plants).

But what Josh and his team got really excited about in this step of look at the data is that 2022 stands out more clearly. It is not only lower than any other year, it is substantially lower than the long-term average over the full time series of the dataset.

They felt like this was a clear story in the data.

So, now onto the data explaining/communicating and graphing!!

In this next graph, Josh made the following adjustments to make it easier for the audience to make sense of:

• Restricted the y-axis to just the relevant range of values (no bar reached 0.3 before, so there is no reason for the chart to go so high)

• The color of each bar to reflect that healthier values are green and lower, poorer values are brown

These changes make it easy to read the data in terms of the bar size, direction, and color. Josh also wanted to draw readers’ attention to 2022—the year of interest— by using a dark brown that catches your eye.

So, he kept adjusting the data to explain and communicate.

To make the graph ready to share with the general public, Josh added a title, subtitle, and a context map.

He chose a title that would reinforce the point they wanted to make and the subtitle to clarify the type of measurement being shown (August NDVI anomalies).

He included a small, subtle map to show readers the area these NDVI values were recorded from.

This chart is now stronger on its own and it fits with the overall visual language in the story.

And then because general public, policy makers, and farmers often think in terms of the space that we live, work, and grow things in they made a map. This covers a similar area to the natural-color images we first looked (aka “Extent of Landsat imagery”), but with the NDVI anomalies overlaid for each area.

Now, wow this pops even more how that area from the natural-color satellite images is REALLY lower than in previous years.

Interestingly this map also helps us see that not everywhere in the Central Valley of California has had such a dramatic decrease in the NDVI. This is something we would NOT have seen from the bar chart of the annual average alone.

Hmmm, seems like maybe something to think more about…like maybe in Tweak #2 next week ;)

But back to this week’s tweak. Why did I want to share this story?

Because think about all these graphs and maps that Josh made as he and his team were making sense of the data they had and how best to explore it to understand what was going on. And then how many graphs and maps that he made to best communicate the story they found in the data.

A. LOT. MORE. THAN. ONE.

Ok, but why all the capitalized words and periods? Well, because more often than not when I had my students make graphs back in the day and when I am working with teachers across the country…we have students make ONLY ONE GRAPH.

As in, we are expecting them to gain an understanding of the process of working with data and the process of doing science, but we are not actually giving them opportunities to see what it actually takes to work with data or do science.

Now, I am NOT suggesting that your students need to dive into calculating indices from NASA satellite data, integrating across areas, and calculating anomaly metrics. Gosh no (well at least not as the expectation of all of our students).

But, I AM suggesting that your students could make a graph, talk about it with a partner, and then revise their graph. Even that those two small extra steps of 1) talking to someone else while you are in the middle of working with the data, and 2) making adjustments to your graph based on that conversation would be HUGE in terms of helping students understand far better what goes into working with data and doing science.

So my challenge to you (should you choose to accept it :) ) is: Where next week/month/calendar year can you have your students make 2 rather than 1 graph of the same data for the same question?

Share your thoughts, comments, wins, and flops! We would love to hear.