Computational Thinking, Data Acumen, Science Teaching, STEM Education, Data Tools

1 Introduction↩︎

Data and computing have become instrumental to all facets of life in the 21st century nasemk12:2026?. While K–12 school districts have, to varying degrees, begun to adapt their curricula to teach data and computing related skills to students rosenberg_big_2022?, nasemwork:2023?, dslp?, nasemk12:2026?, the question of how best to foster these skills practically and equitably remains unsolved.

In this paper, we propose that spreadsheets—despite well-documented limitations nashjohn:1996?, piment:2022?, eric:2022?—can play a meaningful role in developing K–12 data and computing skills. We argue that spreadsheets are part of the toolkit students already encounter, and that their flexibility and prevalence in K–12 settings make them worth taking seriously as a mechanism for skill development.

Recent reports have emphasized the importance of developing data acumen and computational fluency for K–12 students nasemk12:2026?. Data literacy, which some argue encompasses statistical literacy, can begin early in schools gould_2022?. Data acumen is a foundational set of skills and understandings that an individual should have to effectively leverage data science NAP25104?. Computational thinking facilitates problem solving and algorithmic thinking wing_2006?, vanb:2023?, and is a critical component of data science and statistics nasemk12:2026?, pcmi2016guidelines?.

The goal of this paper is threefold:

1. Demonstrate the potential utility of spreadsheets in a K–12 data and computing curriculum;

2. Describe how widely adopted standards documents already outline related data and computing skills that can be taught and reinforced using spreadsheets; and

3. Suggest a set of proposed skills that complement other data and computing learning experiences.

We acknowledge that other data science tools commonly used in K–12 settings (e.g., CODAP, GeoGebra, Desmos, TuvaLabs, and even R or Python) should serve as complements to spreadsheets, each bringing its own strengths to data exploration moon_2023?. Our focus on spreadsheet tools such as Google Sheets and Excel reflects their near-universal availability in school districts and the foundational role they can play in a broader problem solving toolkit (which complements statistics and data science analysis tools).

Toward these goals, we first evaluate spreadsheets against an existing framework for data science tools to better assess their advantages and limitations (Section 2). We then review relevant standards and frameworks that inform data and computing skills K–12 students might develop (Section 3). After addressing how K–12 students might develop data and computing skills, we turn to what they should ultimately gain from working with spreadsheets. In Section 4, we propose five data skills, each illustrated with sample activities that could be integrated across a variety of courses (including but not limited to science, math, and computer science). Finally, we outline challenges, limitations, and the tangible benefits of teaching with spreadsheets (Section 5).

2 Spreadsheets—A Part of the Data and Computing Toolbox?↩︎

Spreadsheets are a common way that data are entered, stored, analyzed, and visualized brom:woo:2018?. These tools are widely used in K–12 settings. A sample of 330 K–12 science educators showed 89% use spreadsheet tools with their students rosenberg_big_2022?. Of those, the vast majority use Google Sheets—either alone or alongside Excel, since students and teachers often have licensing by default rosenberg_messy?, moon_2023?. Hohenwarter_2014? notes that K–12 mathematics software tools often combine a spreadsheet component with a dynamic geometry system and a computer algebra system. Hudson03072025? describe how spreadsheets can be used to support computational thinking and data acumen (e.g., data moves) in secondary school statistics lessons.

Past work has considered guidelines and best practices surrounding the use of spreadsheets for data science. For instance, hertz_eleven_2024? provide a number of tips to effectively utilize a spreadsheet tool for analysis, focused on the importance of making spreadsheets machine-readable (including by having one data point per cell), using computer-friendly characters, defining null values, and being thoughtful around datetime records. brom:woo:2018? share detailed guidance for how data should be organized in spreadsheets. They call for consistency, a thoughtful choice of names for columns and files, a prohibition on calculations within a data spreadsheet, and a plea that data be stored in a rectangular format wick:2014?.

Should spreadsheets be used beyond just as a data organization tool? We believe that they are a component of a broader toolkit and that repeated learning experiences with spreadsheets will deepen data and computing skills. To justify this claim we begin by reviewing frameworks for teaching tools.

There is a considerable history of evaluating the strengths and weaknesses of teaching tools in statistics, beginning with biehler1997?. mcnamara_key_2018? builds on this prior structure to describe key attributes of modern statistical computing tools, relevant for novice and expert users.

piment:2022? extend and adapt the frameworks proposed by biehler1997? and mcnamara_key_2018? to evaluate K–12 data science tools. These tools are classified into four categories: Spreadsheets, Visual Interfaces, Scripting Languages, and Other Interfaces.

piment:2022? conclude that “relying solely on spreadsheets as a data analysis tool is likely to lead to a frustrating experience for those learning and engaging in data science.” While we concur with that assessment (relating to data analysis), and are not suggesting that spreadsheets are the sole tool students should utilize, we believe that spreadsheets have a number of strengths that make them a valuable part of the data and computing toolbox for K–12 students, particularly in the areas of data organization, making data visible and interactive, and fostering computational skills and thinking wing_2006?.

In this section, we expand upon the strengths and weaknesses of spreadsheets in K–12 data and computing education. We elaborate upon each of the elements of the piment:2022? framework (reprinted in Table 1), arguing that while the conclusion of piment:2022? has merit in some areas, it understates the strengths spreadsheets bring in others, particularly computing and data organization skills, and the ability to make data visible and interactive for students. For each of the elements, we provide an informal grade for how well spreadsheet tools address each item in the framework.

2.1 Accessibility↩︎

piment:2022? note that Accessibility is a strong suit of spreadsheets in a K–12 context. We concur. Most school districts provide access to either Google Sheets or Excel to all students. Sheets is fully web-based and has associated apps, so it can be used on any device with an internet connection. These programs can usually be used with screen readers.

Given the widespread existing adaptation of Google Sheets in many school districts rosenberg_big_2022? and its flexibility and power, especially when used with connectors, we believe this tool is well poised to assist in the development of data science and computing skills for K–12 students across the country.

Pre-existing licensing agreements minimize concerns about potential safety, security, and professional development implications of acquiring new software data-security? and leveraging existing licenses.²

2.2 Data as a first-order object↩︎

With spreadsheet tools, data are brought to the fore—they are presented clearly and directly in front of the user, and they are easy to interact with. This is an example of Data as a first-order object and allows for a rapid development of student conception of data. leemojica2022? describe how expert data scientists often explore data by scrolling through a spreadsheet to understand structure, outliers, and other features of the data. This is perhaps one of the strongest features for spreadsheets. This may be the reason why spreadsheets are so widely used in K–12 settings, as they allow students to directly interact with data and develop intuition about what data look like and how they are structured. It may also account for how other tools (e.g., CODAP, GeoGebra, Desmos) have incorporated spreadsheet-like data views to allow users to interact with data (see Section 2.7).

2.3 Ease of entry↩︎

By making data visible to users and interactive, spreadsheet tools foster Ease of entry. Getting started with the basics of data entry is quite easy, with a nontrivial learning curve to get from data entry to numerical analysis, and an even greater learning curve to be able to perform more sophisticated statistical analysis. The simplest functionality can be taught early on, with more advanced skills and intuition being built over time. This is another relative strength of spreadsheet tools.

2.4 Data analysis cycle↩︎

Taken together, the range of functionality that spreadsheets possess encompasses components of the Data analysis cycle. Work by lee_investigating_2022? describes frameworks that outline core components of the data analysis cycle. They look to a related idea: the “Data Investigation Process” (see also Competency 1 of nasemk12:2026?). This process begins by framing a problem; then, an analyst will collect and record data witte_2025?, witte_data_2025?. Next, these data will be processed, explored, and visualized ridgway_2017?. Finally, students can undertake additional analysis and simple modeling, then communicate results and findings.

Using a spreadsheet, students in elementary school could examine individual records, enter data from simple experiments, and develop basic intuition about what data look like. Middle school students, with just a few mouse clicks, can create simple charts to visualize their data. Basic formulas, such as SUM() and AVERAGE(), can be taught to generate numerical summaries. Later, students can begin to use increasingly sophisticated functionality that more closely mimics the Data Moves outlined by erickson_data_2019? (and extended to the spreadsheet context by moon_2023?). Specifically, these Data Moves include: filtering, grouping, summarizing, calculating, merging/joining, and making data hierarchical.³ Finally, some basic modeling is offered by spreadsheets.

How can students communicate their results? Although they are able to copy results from a spreadsheet into a written report, this “cut and paste” process is not ideal or recommended baumer2014?. More generally, version control beckman:2021? is an important aspect of data science and is an area where there is a need for improved tools NAP25104?. While recent improvements in Google Sheets version history allow for a user to manually trace through some granularity of edits and adjustments, without significant effort and thought, it is quite challenging to produce a fully Reproducible workflow. The situation is worse for other common spreadsheet tools, which led to our informal grade of C for this item in the framework.

2.5 Interactivity and flexible plot creation↩︎

Interactivity and Flexible plot creation are two places where spreadsheets themselves fall short of other data analysis tools. While spreadsheets are commonly used to create graphical representations of data prodromou?, they don’t have built-in support for multivariate displays. As another example, there is no built-in way to make a simple dot plot (a common graph used in early grades to help students understand distributions, konold:2007?). (CODAP, on the other hand, provides interactive and easy-to-use features to produce these and other visualizations.) ridgway_2017? provides a review of tools for visualizing data.

One solution to the problem of generating multivariate displays is provided by a connector between Google Sheets and Google Data Studio (formerly Looker Studio). While Data Studio is an additional tool for flexible and interactive dashboard creation, it is included with any Google account. It provides support for a wide range of column types, the creation of custom metrics and fields (“calculated fields”), multivariate displays, interactive filters, cross-filtering interaction across charts, and more.⁴

Further interactivity and automation, though it adds complexity, can be developed in the Google Workspace using Google Apps Script, a JavaScript-based programming language specific to Google. This language allows tools such as Sheets, Docs, Slides, Calendar, Mail, and more to be woven together in unique and useful ways. While it may be more challenging because of the need for more complex coding, engaging in learning and projects that employ Apps Script could allow students and teachers to develop more tool sense, understanding the larger ecosystem of Google and how different software tools can interact.

2.6 Inferential reasoning and non-standard data↩︎

Finally, Inferential reasoning and support of Non-standard data, while important for data analysis, fall largely outside of the scope of this paper (and are areas where spreadsheets are less well-suited). We note, however, that both Google Sheets and Excel have built-in support for regular expressions (allowing the processing of text data), and Google Data Studio has robust support for map-based data. Google Apps Script can also be used to process unstructured data, such as JSON, XML, or HTML files.

2.7 Other Related Tools↩︎

Spreadsheets are by no means the only tool that can be used to develop data and computing skills for K–12 students. A number of complementary tools have been proposed and are being used in K–12 settings, including CODAP fris:bieh:2021?, Hudson03072025?, GeoGebra geogebra?, Mousoulides2011?, prodromou?, Desmos bourassa:2019?, TuvaLabs, and even R or Python moon_2023?. Some of these tools feature spreadsheet-like interfaces and data views. fris:podw:2022? describe how CODAP and TinkerPlots use spreadsheet-like tables in effective ways. fris:bieh:2021? raise important questions about choice of tools (e.g., educational software such as CODAP, spreadsheets, or professional programming tools). While we focus our attention on how data and computing skills can be facilitated using spreadsheets, we believe that such tools are only a part of a broader set of tools and technologies that students in K–12 should encounter, and that educators should not adopt a “one-size-fits-all” approach to data and computing solely using spreadsheets.

3 Data and Computing Standards and Frameworks↩︎

Having covered why we believe spreadsheets (and Google Sheets in particular) have a role in data and computing, we now describe what we believe these students might learn. We begin by reviewing relevant standards and frameworks that relate to the teaching of statistics, mathematics, data science, and computing as a way of demonstrating that related topics are already part of the K–12 curriculum, albeit with varied uptake.

3.1 “K–12 Computer Science Standards”↩︎

In 2017, the Computer Science Teachers Association (CSTA) updated their K–12 Computer Science Standards, a “core set of learning objectives designed to provide the foundation for a complete computer science curriculum and its implementation at the K–12 level.” csta2017? Their framework lays out high-level concepts (such as “Computing Systems,” “Networks & The Internet,” or “Data & Analysis”), as well as sub-concepts (e.g., within “Data & Analysis,” one finds “Storage,” “Collection, Visualization, & Transformation,” and “Inference & Models”). This structure is meant to develop an enduring understanding of principles and best practices for working with computers. Some ideas from the concepts “Data & Analysis” and “Algorithms & Programming” are relevant to developing data acumen and computing skills and can be fostered by learning to use a spreadsheet.

“Data & Analysis” includes “Collection, Visualization, & Transformation” of data, ranging from the basics of data collection and simple visuals in elementary school to data transformation and processing in middle and high school witte_2025?. The “Inference and Models” subconcept (while less relevant to a spreadsheet-driven activity) also speaks to the importance of using data and visualizations to find patterns, communicate them, and predict potential future outcomes.

The “Algorithms & Programming” subconcept refers to variables as being “like a container with a name, in which the contents may change, but the name (identifier) does not.” This idea could be understood as quite similar to a spreadsheet cell: when you use a cell reference (e.g., A2), you can change the content of that cell, and the reference will return whatever value is currently in it.

Finally, the idea of “Modularity” is useful for the data analysis process, by decomposing an analysis into a “precise sequence of instructions.” In the real world, one is often presented with data and asked a question; it is then the job of the analyst to figure out how to identify (and document) steps to utilize the data to answer that question.

3.2 Data Science Learning Progressions↩︎

The Data Science Learning Progressions dslp? is a recent framework for K–12 data science education that includes a set of learning progressions across five strands: “Dispositions and Responsibility”, “Creation and Curation”, “Analysis and Modeling Techniques”, “Interpreting Problems and Results”, and “Visualization and Communication”.

Several of the data skills outlined in Strand C (Analysis and Modeling) reference spreadsheet tools, including grades 3-5 C.2.3b “use spreadsheets to visualize trends and relationships”, C.2.3c 3-5.C.2.3c “use no-code or low-code data science tools. e.g., CODAP, Desmos, Google sheets” and grades 6-8 C.5.2b “use simple mathematical or computational models (e.g., statistical summaries, spreadsheet formulas) to describe patterns and relationships in data”. In Strand D (Interpreting Problems and Results) in grades 9-10 D.1.3b students “solve a real-world comparison problem using a digital spreadsheet, such as selecting insurance policies or entering different lotteries.”

3.3 Standards for K–12 Mathematics↩︎

According to the Common Core Standards for Mathematics common-core-2023?, “Modeling is the process of choosing and using appropriate mathematics and statistics to analyze empirical situations, to understand them better, and to improve decisions.” They provide Figure 1 to demonstrate the standard workflow used to create a model. The National Council of Teachers of Mathematics “High School Mathematics Reimagined, Revitalized, and Relevant” report nctm2024? reiterates the importance of mathematical modeling.

Models can be as simple as writing a total purchase price as a product of the cost per item and the quantity purchased. They can be more complex, as well: modeling financial data, selecting driving routes, predicting the weather, and more. Several potential applications of modeling exist that can be demonstrated using a spreadsheet.

To demonstrate an important concept in mathematics of function composition (applying a function to the result of a second function, i.e., \((f \circ g)(x) = f(g(x))\)), students can use formulas to represent different functions across multiple columns. Properties such as function associativity and lack of commutativity can be explored and visualized with real numbers.

Sequences and series provide another straightforward example of how one might integrate mathematical modeling and spreadsheets. Formulas for recursive and explicit expressions of geometric series could be introduced. Consider a scenario where someone deposits $1,000 into a bank account each year and accrues annual interest at a rate of 5%. They can calculate the total balance of the account at the end of 50 years geom_series_applications?. This task could first be explored recursively using a spreadsheet; this may then help students visualize the series component. Students can then derive an explicit formula and confirm it matches. More guidance about mathematical modeling in K–12 can be found in the Guidelines for Assessment and Instruction in Mathematical Modeling Education (GAIMME) report gaimme?.

3.4 Guidelines for Assessment and Instruction in Statistics Education↩︎

The GAISE (Guidelines for Assessment and Instruction in Statistics Education) II K–12 report outlines skills that students need to make sense of data gaise_ii?. The report highlights the importance of the statistical problem-solving process, the process of collecting and wrangling data, and the important role of technology in data analysis.

As part of the GAISE II K–12 framework, Level B “Collect Data/Consider Data” includes understanding “that data are information collected and recorded with a purpose and can be organized and stored in a variety of structures (e.g., spreadsheets)” gaise_ii?. The report also discusses how students need to “interrogate” the data to learn about data production, types of variables, and how data are represented.

3.5 A Framework for K–12 Science Education↩︎

The national_2012? published a framework for science education for K–12 students that relates closely to the data analysis cycle described in gaise_ii? and formalized by lee_investigating_2022? for the “Data Investigation Process”.

Chapter 3 notes the importance of “developing students’ knowledge of how science and engineering achieve their ends while also strengthening their competency with related practices.” They focus on practices they see as both relevant and historically underemphasized in typical science education: modeling, writing explanations, and “engaging in critique and evaluation.”

The third practice listed in Chapter 3 is “Planning and Carrying Out Investigations.” It is suggested that from an early age, students should be led to think about how data are collected and recorded in scientific investigations.

The fourth practice listed in this chapter is, “Analyzing and Interpreting Data.” Students need to be able to use spreadsheets, databases, and other data tools to store and analyze data and relationships between variables.

Finally, the eighth practice is “Obtaining, Evaluating, and Communicating Information.” Students are encouraged to “write accounts of their work, using journals to record observations, thoughts, ideas, and models. They should be encouraged to create diagrams and to represent data and observations with plots and tables, as well as with written text, in these journals.” This goal is consistent with the use of a lab notebook and the by-hand recording of data, information, and reflections. Ideally, technology tools could replace these written recordings to facilitate improved reproducibility baumer2014?, nasem2019?.

3.6 Developing Competencies for the Future of Data and Computing: The Role of K–12↩︎

A recent report from the National Academies provides a set of data and computing competencies for K–12 data and computing education nasemk12:2026?. The report proposes a set of seven competencies that all K–12 students should develop, beginning in elementary grades:

1. Problem Posing and Problem-Solving Processes

2. Producing and Working with Data

3. Abstraction, Algorithmic Thinking, and Automation

4. Probabilistic and Inferential Reasoning

5. Models and Representations

6. Technology and Society

7. Data and Computing Systems.

A number of these competencies are relevant to the use of spreadsheets in K–12 education, including “Producing and Working with Data” (Competency 2), “Abstraction, Algorithmic Thinking, and Automation” (Competency 3), and “Data and Computing Systems” (Competency 7). The report highlights ways that spreadsheets can be a useful way for K–12 students to structure and organize data, particularly early in their interactions with data.

3.7 Park City Report and Data Science for Undergraduates: Opportunities and Options↩︎

At the undergraduate level, both the Park City report pcmi2016guidelines? and the National Academies “Data Science for Undergraduates” report NAP25104? provide guidance for data science programs. The definition of “data acumen” in the “Data Science for Undergraduates” NAP25104? consensus study report includes numerous concrete goals that may be relevant at a K–12 level. Chapter 2 of the report explores the foundational set of skills and understandings that someone should have to effectively leverage data science. Data acumen comprises a number of concepts, including “statistical foundations” (e.g., exploratory data analysis), “data management and curation” (e.g., data preparation), “data description and visualization” (e.g., visualizations). The section on “computational foundations” places emphasis on student ability to “be able to learn new data technologies.”

To that end, we believe that developing the ability to self-source solutions to problems that arise and understanding the reasons for data collection, storage, and manipulation/wrangling is more important than just following steps to use a specific spreadsheet tool.

3.8 Dana Center Course Frameworks↩︎

Parallel to this research base, the Dana Center’s course frameworks emphasize that all students should be able to critically interpret and engage with data, attend to data ethics, and use technology to conduct investigations in authentic contexts danastats?. The Data Science framework in particular includes as a learning outcome, “Access relevant online datasets/data information (i.e., via spreadsheet imports or API calls).” There is also emphasis on the importance of using “technology to wrangle, visualize, and explore data to develop conceptual understanding.”

4 Data and Computing Skills↩︎

Are there ways to support students’ movement across an investigation process—from data creation/collection and structuring, to exploration/visualization and communication—rather than treating tool use as a set of isolated procedures?

The standards and frameworks introduced in Section 3 provide guidance for the data and computing skills K–12 students might find valuable and indicate where in the existing and future curriculum these skills might be developed. We believe that spreadsheets are a vehicle through which one can teach general concepts that will build data and computing skills that complement other data tools benefiting students into the future.

For instance, a student should understand why they would choose a specific spreadsheet setup to record data from a given experiment, or why they might represent a relationship with a scatterplot rather than a line graph. Such experiences may be more valuable than a “cook-book” approach where they exclusively follow detailed instructions on how to create these outputs given a certain type of input.

In this section, we propose a set of five aspirational data and computing skills derived from existing learning standards and frameworks for undergraduate and K–12 students. We describe how these skills might be introduced or practiced with a set of sample activities and how they connect with existing data and computing standards and frameworks.

1. Data entry, aggregation, and visualization

2. Data consistency checking

3. Mathematical applications within spreadsheets

4. Algorithmic thinking and implementation

5. Scripting, automation, and tool integration

We now examine each of these data skills in turn and propose several potential ways in which spreadsheets could be integrated into the K–12 curriculum.

Given the connected applications offered by Google and the added benefit of Google being entirely web-based (thus standardizing the user experience across different computer types), Google Sheets is a common spreadsheet software environment of choice for K–12 districts. For the remainder of this paper, spreadsheet functions will be explained or demonstrated as they are written in Google Sheets (small modifications are required to implement these examples in Excel).

4.1 Data entry, aggregation, and visualization↩︎

The merits of a spreadsheet using data as a first-order object become immediately clear when simple data entry is employed. This could happen in a middle school class where students are conducting a science experiment, estimating the value of \(\pi\) in pre-algebra, or learning the basics of probability with a coin flip experiment. Instructors can demonstrate this process on a projector, and students can follow along themselves.

These types of activities easily lend themselves to demonstrating arithmetic operations, as well as basic aggregation functions, such as SUM() or AVERAGE(). In a middle or high school science lab, once data have been collected, graphical displays can be created to visualize results gaise_ii?. This could be anything from scatter plots to understand force in physics to bar charts comparing average plant growth under different light conditions in a high school biology class national_2012?.

Further, science classes could have individual or small groups of students collect their own experimental results and then pool these results across a full class roadless?. Those who teach the same class for multiple years could even pool results across time. This would allow students to better understand sampling error and develop intuition on how the Law of Large Numbers (the distribution of means is less variable than the distribution of individual observations) works. Such spreadsheet activities could complement work done using applets or similar tools applets2012?, gaise_ii?.

Science and math classes are not the only ones that can develop this skill. In one 9th-grade English class, students who read The Catcher in the Rye worked in small groups to identify textual evidence of Holden Caulfield’s depressive symptoms, recording results in a spreadsheet. The teacher aggregated these results and made a bar graph to spark a data-driven discussion on the severity and context of these behaviors ssht-classroom-ex?.

We now present two concrete examples of demo spreadsheets, created to illustrate how one might go about exploring this first data skill. Each is an actual Google Sheet that can be accessed via the supplemental materials. (Both rely on fabricated names and data to avoid privacy concerns.)

4.1.1 Paper Airplanes↩︎

This first example provides data one might collect from an experiment on paper airplanes, measuring the distance a plane flies considering two different wing structures and three different types of paper boger?. The data itself could be experimentally collected by students, and data entry into a spreadsheet of this form can be utilized. A snapshot of the data sheet is presented in Figure 2. We believe that this form of spreadsheet use is quite common and facilitates teaching several of the nasemk12:2026? competencies (including “Problem Posing and Problem-Solving Processes” and “Producing and Working with Data”).

A number of summary visualizations and numerical summaries may be included as part of such an investigation. For instance, perhaps a student could create a histogram of flight times. Or, summary statistics aggregated by paper type could be derived. The generation of metrics by paper type (e.g., average flight distance by paper type) can be efficiently implemented using a pivot table. A bar chart could be used to display comparative results (see the “Results” tab).

It would be straightforward to move from a spreadsheet tool used for data collection and organization to a more flexible and interactive visualization tool, such as CODAP. Such an approach would allow students to compare and contrast how effectively tools are able to solve particular tasks see Competency 7 “Data and Computing Systems” from nasemk12:2026?.

4.1.2 Library↩︎

The second example, representing a very small school library, provides some more sophisticated conceptual functionality. The first tab is a database of books in the library while the second contains one row for each time a student checked out one of these books. Book IDs are used to uniquely identify each book. Thus, consistent with a relational database, rather than using a book name associated with each checkout record, we incorporate a book ID. These data are then used, in conjunction with a VLOOKUP() formula, to automatically populate the book title in column D.

We provide a number of analytic questions a user might seek to answer using these data, incorporating different types of aggregation and exploratory numerical analysis. These questions, quantitative answers, and the formula used to obtain these answers are available on the tab titled “Questions.” For instance, a simple AVERAGE() formula is used to compute the mean number of pages per book in the database. An AVERAGEIF() formula is used to compute the average length of Fantasy books in the database; this average is higher than the overall average number of pages per book. Finally, the tab titled “Checkouts Pivot Table” demonstrates an application of a pivot table, specifically for tracking the number of times each book has been checked out.

The library activity reinforces competencies “Producing and Working with Data” and “Abstraction, Algorithmic Thinking, and Automation” from nasemk12:2026?, as well as the “Data & Analysis” and “Algorithms & Programming” concepts from the CSTA K–12 Computer Science Standards csta2017?. It also leverages some of the hierarchy concepts central to eric:2022?.

4.2 Data consistency checking↩︎

Real-world data analysis requires that someone looking at raw data possess tools to recognize where that data may be inaccurate or impact analysis, and take steps to remedy that. This component of data acumen is often overlooked in K–12 education, but it is an important one that is included in nasemk12:2026? as part of the “Producing and Working with Data” competency. For instance, suppose a student downloads a data source with a numerical column, but the numbers have commas in them. Subsequent use of functions (e.g., SUM() for aggregation) may not work properly.

The basics of data cleaning and consistency checking might be introduced in several ways. First, a teacher could use an artificial example of fabricated data to demonstrate some typical real-world problems.

Alternatively, real-world data can be used. As an example, a pre-algebra class or algebra class could model real-world data, allowing the students to download potentially messy data and work towards cleaning it.

As a larger scale example, students could take the Census at School survey census-school? and then download data from students all across the country who have also taken the survey. This latter approach provides a personal connection to the data being analyzed: the students will themselves know what the questions are and which questions are most interesting to them. Such approaches are valuable for students to develop a mindset of data wrangling and to begin to inculcate defensive coding techniques McNamara02012018?.

4.3 Mathematical applications within spreadsheets↩︎

High school math classes such as Algebra II and Pre-Calculus contain numerous mathematical concepts that can be implemented using spreadsheets. For instance, a unit on exponential growth might cover compound interest formulas, which students could compare to simple interest. Such basic modeling is at the core of nasemk12:2026?’s “Models and Representations” competency.

A spreadsheet to demonstrate this model could be made fully interactive, allowing students to understand some more sophisticated spreadsheet features: perhaps principal amounts and interest rates can be modified, frequency of interest payments can be adjusted, etc. This type of “what-if” analysis is a powerful way to develop intuition about how different parameters impact the growth of an investment. Such applications are straightforward to implement in a spreadsheet and were some of the earliest use cases visicalc?.

Larger scale projects that may be appropriate for these classes could include further financial and mathematical modeling concepts, such as a project comparing the cost of purchasing and maintaining a gas car compared to an electric car. The cost of home ownership could be compared using a renting and a purchasing structure. In both of these examples, graphs can be used to display findings, with students encouraged to make these displays as interactive as possible. For example, a user might be able to adjust the initial parameters and, if formulas are properly implemented, see those changes permeate through the spreadsheet.

Finally, units on the basics of probability, at the center of nasemk12:2026? competency on “Probabilistic and Inferential Reasoning” present an attractive opportunity to incorporate spreadsheets into the classroom. Expected value calculations and randomness are two extremely useful applications. The recent film Jerry and Marge Go Large depicts a couple who found a lottery with a positive expected value under certain conditions frankel_jerry_2022?. Students can use spreadsheets to show the positive expected value and then be provided with a project or activity to assess this on a lottery of their own choosing. Later explorations could use applets or other technologies. Such simulations may help students explore how expected values can be misleading when not paired with measures of variability buffett?.

4.4 Algorithmic thinking and implementation↩︎

Spreadsheets provide an opportunity to deepen computational thinking skills vanb:2023?. After multiple exposures and opportunities to practice, students might find themselves implementing more complex, algorithm-based formulas in Google Sheets. For instance, suppose a student makes a spreadsheet to store their grades for the class. In cell A1, the student records a numerical grade; in cell A2, they seek to implement a formula for the letter grade associated with their numerical grade. Such a formula is fairly convoluted if created using basic Google Sheets functions—it requires multiple IF formulas nested within one another. For instance, such a formula for only letter grades A, B, C, and fail might read: IF(A1>%,"A",IF(A1>%,"B",IF(A1>%,"C","Fail"))). Implementing this type of nested logical structure in code reinforces nasemk12:2026? competency “Abstraction, Algorithmic Thinking, and Automation” and related concepts vanb:2023?.

In 2009, the programming language Google Apps Script was released noauthor_launched_2009?. Apps Script is a lightweight, JavaScript-based scripting language allowing for the automation and connection of multiple tools within Google Workspace (e.g., Drive, Sheets, Docs, Calendar, Gmail, etc.). The scripting language is well-documented, and being JavaScript-based, would be accessible to high school students with prior programming experience bastug_effect_2017?. Additionally, Apps Script is freely available with any Google account.

While not all K–12 students may need to learn to program or code, and while there is a need for professional development to support its use, Apps Script provides an accessible jumping off point for those who choose to do so.

One powerful application of Google Apps Script, suitable for a basic introduction to the purpose of the language, is coding custom functions for a Google Sheet apps_script_quickstart?. We return to the prior example of a grading spreadsheet. The Apps Script implementation of this formula might include if and else-if statements or a switch statement. Example source code is provided in Appendix A (Example 1).

With examples like these (and with increasingly sophisticated use cases over time), students can be introduced to more sophisticated algorithmic thinking and its applications to data.

4.5 Scripting, automation, and tool integration↩︎

As students dive deeper into Google Workspace, they may begin to encounter problems that can’t be solved using the built-in functionality. This provides an opportunity to deepen their understanding of the ecosystem of tools and how they can be integrated, as well as to develop skills in scripting and automation (nasemk12:2026? Competencies 3 “Abstraction, Algorithmic Thinking, and Automation” and 7 “Data and Computing Systems”.)

For instance, suppose a student creates a note-taking template for a certain class within Google Docs, and they want to make one copy for each day of the week, renaming them accordingly. Google Drive doesn’t even allow for bulk copying of documents, let alone dynamic re-naming. Example source code to run this simple automation is provided in Appendix A (Example 2).

As a more powerful and complicated example, perhaps a student creates a sign-up form for an event RSVP they are co-hosting with another student. However, this event only has space for 20 attendees. Thus, the two students wish to receive a notification once they reach 10, 15, and 20 sign-ups, so they can track their progress towards capacity but not have to manually check a response spreadsheet. Associated code is provided in Appendix A (Example 3).

As a third example, let’s suppose a student sets up a spreadsheet as a job application tracker. This may have a column for the company they are applying to, a second column for the application deadline, and a check box for whether they have submitted the application. One week before the application is due, it might be useful to get an email notification stating that their deadline is coming up, but only if they haven’t submitted the application already. As before, code to perform such actions is provided in Appendix A (Example 4).

These computing skills, consistent with those that are now commonly taught in many districts and states ds4e_state?, could be developed at a basic level for all students, with potential extensions and deeper exploration for students with more specialized interests.

All three of these examples reflect problems that can’t be solved by default within Google tools, but Apps Script can and does allow for them to be accomplished, leveraging the earlier investment in use of Google Sheets. Activities along these lines could provide students with practical examples of the uses of Apps Script, tangible use cases for programming, and a powerful, eye-opening investigation into the potential for integration of multiple tools.

The integrations that can extend the power of Google Workspace go beyond just programming. Google Data Studio is a free, web-based drag-and-drop data visualization software that allows any user with a Google account to create interactive, permissioned dashboards. It includes a built-in connector to Google Sheets, and also allows for the easy upload of Excel files and CSVs. Thus, although Google Sheets itself may lack the qualities of interactivity and flexible plot creation (mentioned above), by combining Sheets with Data Studio, both of these features are readily accessible and can be implemented at a high level very quickly. Such tools might be used to complement more traditional K–12 data analysis tools such as CODAP and provide structure to deepen computational thinking skills csa2z?.

5 Discussion and Conclusion↩︎

Nearly 20 years ago, konold2007? asked: what are the core ideas in statistics in data analysis, what statistical capabilities do citizens need, and “how do we start early in young people’s lives to develop these capabilities.” The nasemk12:2026? K–12 data and computing competencies report called for a comprehensive rethink of how K–12 students’ learning about data and computing could be integrated more effectively and begun much earlier.

We believe that spreadsheets have a role in K–12 data and computing education and can serve as a foundation for work in other tools. They are not the only tool that should be used to develop data and computing skills, given the many weaknesses that they have for analysis, visualization, reproducibility, and modeling (to name just a few). But they have a number of features that make them particularly well-suited for this purpose (e.g., bringing data to the fore, ease of use, interactivity, etc.), are already widely used in K–12 settings, and are an important workplace tool.

Our aim in this paper is deliberately focused. We do not set out to build a comprehensive K–12 curriculum or provide empirically validated lesson plans; rather, we hope to provide some examples that could allow others to lay conceptual groundwork on which such efforts can build.

We believe that students can achieve deeper data and computing skills through repeated exposure and thoughtful integration of spreadsheets across a K–12 curriculum. While spreadsheet use could theoretically be woven into almost any class nasemwork:2023?, nasemk12:2026?, it may be most straightforward to integrate them across science, math, and computer science courses.

Given a fast-changing technology landscape, foundational concepts may be more useful than knowledge of any specific technology. Learning both the “how” and the “why” allows spreadsheets to be a vehicle through which one can teach general concepts, building skills to benefit students into the future NAP25104?. We believe that the approach we suggest here has potential to help integrate and leverage data and computing into the K–12 curriculum.

In their framework for statistics courses, danastats? suggest that students “use technology and statistical software and applications that are freely available, easy to use, accessible across platforms and devices, and robust to multiple contexts and data set sizes, with ongoing accessibility to students after the class ends.” Tools such as Google Sheets meet these criteria in many districts due to existing licensing.

While we have described how spreadsheets could help develop data and computing literacy, fluency, and acumen in K–12 students, they are still subject to a number of notable limitations. A key challenge for spreadsheets is how they fall short on portions of the data analysis cycle. For instance, Google Sheets does not have the default ability to create multivariate scatter plots (though Google Data Studio provides this feature, albeit at the cost of introducing more complexity and an additional tool). Additionally, while spreadsheets may be part of a reproducible workflow, they don’t possess strong default features to allow for the reproducibility of an analytic workflow. Tools better equipped for a reproducible workflow (e.g., quantitative analysis programming languages like R) do exist but feature a steep learning curve.

While spreadsheets can be beneficial in transitioning students from working with small data sets to medium-sized data sets, they have some limitations when handling “big data” as the term is typically used in the contemporary context. Specifically, Google Sheets has a maximum of 10 million cells allowed per spreadsheet. More importantly, Google Sheet functionality and performance degrade as the number of cells and formulas in a spreadsheet increases rahman_benchmarking_2020?.

Numerous conversations between the authors and selected K–12 educators shed further light on pedagogical considerations surrounding the use of spreadsheet tools within these curricula. One of the biggest challenges may be a lack of teacher knowledge: teachers who don’t themselves know how to use spreadsheets will likely be reluctant to teach them in classes, and/or may get frustrated with trying to teach these technologies to students and give up on them rosenberg_messy?. Thus, robust professional development is needed to prepare teachers for this type of work with students nasemk12:2026?. These issues are important areas for future work.

In addition to a lack of teacher knowledge, practitioners often reported how little students themselves know about spreadsheet tools—even in high school. Starting from the very basics is crucial. One instructor has turned formula writing into a verbal call-and-response process for students. He prompts students: “What do we start with?” They are expected to respond by saying, “Equals sign.” Then, a prompt of “How do we save our formula?” is replied to with, “Click return.”

It is difficult, though important, to help students go from a “cookie cutter” plug-and-chug approach to solving problems more dynamically. For instance, students may struggle to figure out how to jump over hurdles they face in the process of working with spreadsheets. There may be ways for instructors to artificially manufacture a stumbling block in an in-class activity and demonstrate to students how to look up and resolve a question. After developing some basic skills with spreadsheets, students could be given a small problem to solve, which requires some more creative thinking but it is not a huge conceptual leap from what they’ve already seen. This allows students to build foundational knowledge and a set of skills alongside problem-solving strategies.

While much less common in K–12 education, we note that a number of courses focused directly on spreadsheet tools exist in colleges and universities bhcc?, spc?, agr-6932?. Typical topics often included in such a course include:

• Basic terminology and key concepts

• Formulas and functions

• Formatting

• Data management

• Charting and graphing

• Dealing with multiple sheets

• Macros and programming

• Data analysis tools

• Data exchange

While this approach may be appropriate at the undergraduate level, we concur with the advice of nasemk12:2026? that it is important to integrate study of spreadsheet tools into a broader curriculum and not just teach in a standalone elective course.

In any discussion of teaching technology in the present world, we need to consider the impact of Generative AI ellis?. AI has the potential to bring technology and programming skills closer to people campbell_using_2024?, and allows individuals to pick up new skills from whatever level they are at. Existing AI tools are often effective when it comes to basic formula creation in spreadsheets thorne_experimenting_2023?. For example, AI tutors may be a way to support students coding using Apps Script.

Similar tools could also be helpful from a Professional Development (PD) standpoint—instead of a school needing to engineer, say, a series of PD courses depending on teacher background in Google Sheets, Generative AI could be used to tailor information to a specific background and complement more traditional PD modalities.

Of course, Generative AI also has the potential to be detrimental to learning outcomes—for instance, by leading students to skip foundational concepts and never develop a robust “mental model” of the material. Further, these tools pose challenges for academic integrity, as students may submit AI-generated work as their own; among the possible responses, in-class activities that ask students to demonstrate their learning in real time may be particularly useful. A thoughtful conversation about how to address these merits and drawbacks would be important when seeking to incorporate spreadsheets more deeply into the curriculum.

When it comes to skills that students will develop, it is important to somehow distinguish specific technical skills (such as the details for how to apply conditional formatting, add dropdown menus, etc.) from more general and transferable data science skills (e.g., being given a data set and some questions to answer, then utilizing a spreadsheet to do so). There is likely need and room for both types of elements in a K–12 curriculum. However, the exact balance between the two needs to be determined.

Assessing learning success can be difficult and may take a variety of different forms. Some approaches may include: written reports, trying to see if students can use data and graphs to back up their claims; blog-style articles describing findings from a spreadsheet project; or in-class evaluations, where students work with their laptops out and must complete some tasks in a spreadsheet.

In closing, we believe that spreadsheets have a part to play in K–12 data science education and can serve as a foundation for work in other tools. The interactive and visually appealing elements of working with spreadsheets have the potential to activate a creative side in students, turning math or science from something they have to do to something fun and interesting. This may allow for the engagement of an entirely different group of students in STEM classes.

They play an important role as a component of a more general data science toolbox. They have a role for data science practice brom:woo:2018?. A recent survey of Master’s graduates in statistics reported that 69% used Excel on a weekly or daily basis asa-survey?. Anecdotal evidence agr-6932? suggests that many graduate students lack sufficient spreadsheet skills.

We proposed five spreadsheet data skills that students should develop and described how they might be integrated into a variety of courses. Spreadsheet tools, when thoughtfully integrated into a K–12 curriculum across science, math, and technology classes Nasir_2024?, can help K–12 students develop data acumen. In a world increasingly dependent on data, it is critical to ensure that students graduate from high school equipped with the knowledge and resources that will allow them to succeed throughout their lives—whether in further education and/or professionally.

6 Disclosure statement↩︎

The authors have no conflicts of interest to report. We thank the editors and reviewers for their thoughtful and constructive feedback on this manuscript, which has helped to improve the clarity and quality of the paper. We gratefully acknowledge the contributions of teachers Zachary Boboth (Sora Schools), Jed Dioguardi (Hackley School), Diana Kaplan (Hackley School), Seth Karpinski (Hackley School), Daniel Lipin (St. Mark’s School of Texas), Bill McLay (Hackley School), and Keshena Richardson (Hackley School) for providing rich and insightful discussions on effective strategies, challenges, and considerations—both benefits and limitations—related to incorporating spreadsheets into teaching and learning within the K–12 setting.

7 Data Availability Statement↩︎

Resources and sample activities have been made available at the following URL: Google Drive

8 Appendix (Code Examples)↩︎

As described in Section 4, Google Apps Script is a JavaScript-based programming language built into the Google Workspace noauthor_launched_2009?. It is edited entirely in the browser, with scripts executed on Google’s servers—no software installation is required. Apps Script is available with any Google account, including education accounts.⁵ Google also provides an official quickstart guide for using Apps Script gas_quickstart?.

There are two core ways in which Apps Script is used; these two use cases speak to different goals in the classroom. Custom functions extend the functionality of Google Sheets. They are similar to the formulas students are already familiar with, such as AVERAGE() or ROUND(), but allow for more customized logic. In a Google Sheet, students can create a custom function via “Extensions” > “Apps Script”. There, a student could write a short function called genLetterGrade() and then enter =genLetterGrade(0.85) in a cell to convert a numerical score to a letter grade.

Automation scripts serve a different purpose. Rather than returning a value, they carry out actions across Google Workspace services—for instance, a script might send emails through Gmail, duplicate documents in Drive, or respond to submissions in Forms. These scripts can be run manually from a menu or button, or launched automatically through triggers tied to events (such as a form submission) or schedules (such as a daily timer).

Example 1 demonstrates a custom function supporting algorithmic thinking (Skill 4, Section 4.4). Examples 2–4 demonstrate automation scripts that integrate multiple Google Workspace tools, supporting Skill 5 (Section 4.5) (scripting, automation, and tool integration).

8.1 Example 1: Custom Letter Grade Function↩︎

This custom function converts a numerical grade to a letter grade using conditional logic. Once saved in the Apps Script editor, it can be called from any cell as =genLetterGrade(A1).

A more robust version of this function might incorporate input validation—checking that the input is numeric and falls within a plausible range. This kind of defensive programming is a useful habit to develop McNamara02012018?, and provides practice with the data consistency checking described in Skill 2 (Section 4.2).

8.2 Example 2: Automating File Duplication↩︎

This script automates the duplication and renaming of a Google Docs template. Given a template document URL, it creates one copy for each weekday, naming each copy after the corresponding day. Here the function does not return a value but instead performs its work as a side effect when run.

8.3 Example 3: Form Response Notifications↩︎

This script, bound to a Google Form, sends email notifications when the number of responses reaches specified thresholds (e.g., 10, 15, or 20 submissions).

8.4 Example 4: Deadline Reminder Emails↩︎

This script, bound to a Google Sheets spreadsheet, sends daily email reminders for upcoming deadlines. It checks each row of the tracking spreadsheet and sends a reminder if the deadline is within a specified number of days and the task has not yet been marked complete.

---

1. nhorton@amherst.edu↩︎

2. For instance, schools must comply with the Children Online Privacy Protection Act and may be looking for companies to sign the “K–12 Education Technology Secure by Design Pledge” from the Cybersecurity and Infrastructure Security Agency.↩︎

3. In consideration of these major data moves (which are similar to verbs proposed by wick:2014? and discussed in chance:2015?): the UI-driven filter feature can be employed; more complex summary functions can be taught for aggregation (including pivot tables to more closely mirror the group-by/summarize approach); and transformations and variable manipulation can occur in new columns. Additionally, joins can be taught using the XLOOKUP() family of functions. We note that some important aspects of data wrangling (e.g., hierarchical relationships) are challenging to implement using spreadsheet tools eric:2022?.↩︎

4. It is worth noting that Power BI, offered by Microsoft, has many similar features to Google Data Studio. However, Power BI is only available for free as a desktop download, thus making Data Studio more accessible in a school environment.↩︎

5. Some Apps Script features may be restricted by a district’s Google Workspace IT configuration.↩︎