This article originally appeared on CoSN’s blog and is reposted here with permission.

CoSN’s Driving K-12 Innovation initiative recently announced that Analytics & Adaptive Technologies as a Top Tech Enabler (tool) for 2025 and it’s appeared on the list five of the past seven years. The Driving K-12 Innovation Advisory Board defines Analytics & Adaptive Technologies as: these are digital technologies that collect and use data related to teaching and learning. Analytics refers to the process of analyzing data collected about student learning and the opportunity to leverage data to inform instructional decision making.

Adaptive technologies are tools that adapt to the student based on their interactions with the technology. ​​​​​​​​​These adaptations could be in the form of suggesting next steps, providing remediation, controlling pacing, or providing feedback based on analysis of the student’s performance.

But what is this topic exactly and why is it important for moving education forward?

During the November 2024 meeting of CoSN’s EdTech Innovation Committee, participants were asked those same questions, and shared their perspectives on what educators should do about the topic. Read on for 10 key takeaways from the discussion.

1. How much is too much data? “When I moved to Virginia [from Ireland], what struck me about the U.S. education system was your engagement with data,” said John Heffernan (CEF Professional Development, Ireland). “In Ireland, we don’t use data in the same way. We don’t collect the same amount of data, and I just wonder, is this a U.S.-centric thing that everything has to be measured?
2. We need common language and understanding around these terms. During the conversation, there was a need to define terms and shared language. For example, Analytics & Adaptive Technologies or Personalization? Personalized or Individualized Learning? Committee member Ruben Puentedura (Hippasus, Massachusetts) explained that we also need to distinguish between traditional data, like test scores, and the much bigger world of data that “can be used creatively and usefully applied by students and teachers that can include a range of things from students’ interests and type of social networks that exist in a classroom.” Puentedura added: “The data can be used, not just as a question of how to assess something, it can be used to scaffold and underpin what happens in a classroom. AI allows you to take rich worlds of quantitative data and qualitative, narrative data and do qualitative analysis on a scale that you couldn’t do before.”
3. Just as EdTech innovators need common terminology, we then need to be able to communicate it with leadership. “As we look at what the messaging should be, how can it be simplified for both school leaders and for teachers in understanding what it is? And then, what workload that we can eliminate from districts from an already full agenda,” asked Andrew Fekete (Community Consolidated School District 93, Illinois). “There’s a lot of resistance in saying that we don’t have time for this conversation. And my pushback is, we don’t have time to not have this conversation.”
4. Challenges in data collection and usage.” Right, wrong, or indifferent, we still have significant pockets of school districts that don’t grasp even the basics of data,” said Beverly Knox-Pipes, EdD (Former CTO/Education Consultant, Michigan). “As a result, they are often reactive rather than proactive, failing to plan and strategize effectively for what they truly need. This includes not only setting up and managing their student information systems but also understanding how to gather and use data to drive student achievement—the ultimate purpose of education.”
5. The impact of AI and machine learning on educational data. “With the advent of AI and small language models, we’re going to be more and more dependent on our own data internally,” said Pete Just, CETL (Just Strategics, Indiana). “I’ve been talking to a lot of school districts about this as they’re trying to figure it out. But the number one thing is: you have to have good data. So if you’re going to try to make decisions and try to go deep on using an AI tool to help make those decisions. They’re going to be off if your data is not quality.”
6. The importance of digital literacy. A high school teacher in Committee member Kathleen Stephany (School District of Holmen, Wisconsin)’s district is piloting a new course called Data Science, which will have a math component, content knowledge, and communication. “The teacher, when she proposed it, talked about how much data is created per minute,” said Stephany. “If you think about Venmo transactions, streaming videos – there’s tons of data. [The course is about] how to use that data and what goes into that.”
7. Since the beginning, this topic has been about student agency. “The role of adaptive technology and analytics was to grow student agency in their own learning path, whether it’s mastering outcomes or failing forward. To learn from it and the process,” said Janice Mertes (CDW Education State Level Ambassador). “There is an adult use of the term and a student use of this term, to the point of adding knowledge of data, literacy, and analytics.”
8. Analytics & Adaptive Technologies are about teacher agency, too. “We should also be talking about the need for teacher agency at the same place where we need student agency in conversations like these, especially when we’re talking about professional development of staff and being able to choose their pathway and have more choice in the targeted professional learning that we offer. We need ways that allow our educators to follow those paths based on the needs that they’re identifying and thinking about how we design those,” Nick Stoyas (Elmhurst Community Unit School District 205, Illinois).
9. Current concerns about data privacy. “I worry about how the data will be used after the student’s assessment. I’m a big believer in analytics and adaptive tech. I think that there’s some definite positives, but I just want to make sure we don’t lose track of that data privacy piece that protects our kids,” Ryan Cox (Osseo Area Schools – District 279, Minnesota).
10. The role of vendors when it comes to ethics and transparency in educational technology. A lively discussion arose about the need for vendors to be transparent about how student data is used and assessed, and Puentedura stated that if the vendor cannot share that information, they could not be considered. Many Committee members agreed. “It should be a checkbox, if they’re not transparent in what they’re doing with student data and supporting their privacy, the school should automatically pass and move on,” Emily Marshall (Vail School District, Arizona).

Thanks to all EdTech Innovation Committee members who participated in this essential discussion!

Key points:

• AI models are only as good as the data that goes into the tool
• Where AI and multimodal learning will go in 2025
• Teachers: It’s time to make friends with AI
• For more news on AI, visit eSN’s Digital Learning hub

It’s never too early for campus or district-wide IT teams to begin planning for upcoming tech upgrades and implementations. Because many of these upgrades happen over summer break, teams can use the upcoming spring semester to ensure their data is in A+ shape to support new AI tools.

AI has already significantly impacted education by improving how students learn, teachers teach, and educational institutions operate. The World Economic Forum’s Shaping the Future of Learning: The Role of AI in Education 4.0 touches upon AI’s extensive potential, from tailored student learning experiences to reducing administrative burdens to using this innovative technology to improve curricula.

For campuses and districts that haven’t started their AI journeys, it’s critical to know that AI models are only as good as the data that goes into the tool.

To ensure data can adequately train AI to improve education-related outcomes, consider these six strategies.

1. Solve for data anomalies

Detecting outliers in your data baseline–like observations, events, or data points that deviate from the standard–is key to optimizing AI in your educational system. Although data anomalies don’t always indicate something’s amiss, it’s wise to investigate them to be sure.

While the exact method depends on data types, distribution, and computational resources, anomaly detection can be handled through statistical, machine learning, and clustering-based methods. By detecting and solving data inconsistencies early, tech teams are more likely to ensure accurate AI models and avoid future problems.

2. Automate data cleansing

Automated data cleansing enhances accuracy and consistency by fixing or removing incorrect, corrupted, duplicate, or incomplete data within a dataset. It’s a critical step toward managing data, ensuring accuracy, and warranting trustworthiness.

This stage is vital because clean, well-prepared data prevents AI from generating distorted results and reduces the computational resources needed by training models. Further, clean, automated data frees educational tech teams to concentrate on developing AI models and other valuable tasks instead of fixing data obstacles.

3. Observe data quality metrics continuously

Identify your campus or district’s key data quality metrics to measure and improve datasets regularly. Monitoring these metrics involves assessing, measuring, and managing data for accuracy, consistency, completeness, reliability, and validity.

Regular audits allow tech teams to be informed and agile, detecting potential problems before they get out of control and negatively affect AI outcomes.

4. Make data governance routine

Setting the rules, roles, and uses of data will help ensure that datasets are clean and accurate before being leveraged for AI. This governance of data processes upholds all teams and tools to the standard needed for successful operation.

Perform and recognize data stewardship for employees promoting your school system’s data governance initiatives. Effective data governance helps decrease data inconsistencies in school-wide systems, which improves overall data integration initiatives.

5. Enhance data security

Since 2005, U.S. educational institutions have undergone 3,713 data breaches, affecting 37.6 million records. Data breaches can damage a school system’s reputation and decrease trust among students, faculty, and the community.

Further, a lack of sufficient data security measures in AI systems could lead to non-compliance, which is often accompanied by a school system being held liable for a breach, having funding withdrawn, or being investigated by the U.S. Department of Education.

To avoid such troubles, secure your data through encryption, access controls, firewalls, content filters, network security, endpoint segmentation, regular backups, continuous updates, and security awareness training.

6. Ensure data is standardized

Finally, data standardization helps AI models learn patterns more effectively and consistently. It is essential for preserving data quality and allows different systems to exchange data in a consistent format.

By practicing the most common standardization techniques–data cleaning, data governance, data normalization, and data transformation–educational institutions can safeguard data consistency, which is critical for training AI and machine learning models.

A+ AI relies on stellar data

AI can help transform school systems by adapting to each student’s learning needs and personalizing their learning experience. By automating clerical tasks, educators’ time becomes free for more hands-on instruction. It can also help to identify strengths and weaknesses in student performance, allowing educators to prepare better-targeted instructional strategies.

If your school district is ready for AI, adopt these six data strategies to ensure that your AI model is optimized for all stakeholders, especially students, teachers, and administrators.

There’s no need to wait for summer break to ensure your data gets a top grade. Start now so your AI program is at the head of its class next fall.

This post on interoperability originally appeared on CoSN’s blog and is reposted here with permission.

Key points:

• As privacy and security become increasingly important, so has the need for interoperability
• 3 strategies to streamline K-12 data management
• 3 ways a modern data solution can improve student outcomes
• For more news on interoperability, visit eSN’s IT Leadership hub

Interoperability is the seamless, secure, and controlled exchange of data between applications. Implementing interoperability can bring information from edtech tools together to holistically contextualize student learning, allowing educators, schools, and districts to understand their students better and support their paths to graduation. Executing data interoperability standards creates a modern data infrastructure by supporting improved development, privacy, and cybersecurity practices while ensuring compliance with industry best practices.

Interoperability has often been seen as an extra step that districts couldn’t necessarily implement due to capacity. However, as privacy and security become increasingly important due to advancing technology, so has the need for interoperable systems.

As Project Director of Project Unicorn, my role has been to show educators and edtech vendors why interoperability is a critical foundational component of a data ecosystem. I often speak with district administrators or edtech vendors who have never heard of “interoperability” or don’t know about using K-12 data standards. This is why Project Unicorn was created; to build a movement for data interoperability and ensure that everyone can access important knowledge about the building and implementation portion of data interoperability.

Project Unicorn works with early adopters who are ready to start their interoperability journey, knowing they have the privacy and security practices to begin. But as new technologies emerge, like generative AI, and as cybersecurity breaches continue to increase (80 percent of school IT professionals reported that their schools were hit by ransomware in 2023, up from 56 percent in 2022, noted in a recent article), educators, solutions providers, and government officials alike have realized that interoperability, privacy, and security enhance each other, and an integrated ecosystem matters.

“The importance of interoperability’s intersection with privacy and security cannot be overstated,” said Paula Maylahn, CoSN, Interoperability Project Director and Principal Consultant at Paula Maylahn Consulting. “Often, the focus on interoperability is on one side of the coin–getting data where it needs to go. Equally as important is the flip side–keeping data from where it should not go.”

How interoperability, privacy, and security enhance each other

Practicing data interoperability can support improved development, privacy, and cybersecurity while ensuring compliance with industry best practices from both districts and vendors in these ways:

Interoperability standards add clarity to data collection and management
Interoperability has the potential to support data portability and ownership
Interoperability standards can improve product usability and security
Supporting interoperability increases transparency around data practices
Interoperability standards can help simplify the process of compliance with privacy laws

(Learn more about Privacy, Security, and Interoperability at https://www.projectunicorn.org/resources/privacy-security-interoperability)

Interoperability at the White House

On Monday, August 7, 2023, a group of school superintendents, educators, and education technology vendors met at The White House with First Lady Jill Biden, Secretary of Education Miguel Cardona, and Secretary of Homeland Security Secretary Alejandro Mayorkas to unveil the new Infrastructure Briefs, a series of three documents by the Office of Educational Technology, one including Cybersecurity and Infrastructure Security Agency (CISA). They spoke on the importance of interoperability and cybersecurity in building safe and resilient data ecosystems in K-12. This “Back to School Safely” event framed how to build an interoperable, secure, and privacy-enabled ecosystem for education technology and support the field.

A national plan for edtech

Following the Infrastructure Briefs, the United States Department of Education released the 2024 National Educational Technology Plan (NETP) on January 22, 2024.

Since 1996, the NETP has served as the flagship educational technology policy document for the United States, setting a vision and plan for K-12 learning enabled by technology across the U.S., territories, tribal lands, and DoDEA schools worldwide. This latest iteration of the NETP delves into the transformative potential of educational technology, highlighting systemic solutions to bridge the digital divides in use, design, and access, particularly in the context of educational equity.

The 2024 NETP aligns with the Activities to Support the Effective Use of Technology (Title IV A) of the Every Student Succeeds Act and continues to promote a vision of equity, active use, and collaborative leadership. It aims to ensure equitable access to technology and the transformative learning experiences that technology enables.

“In the real, day-to-day work of interoperability in a K-12 school district, the goal is to get actionable data in the hands of administrators and educators to make the best possible decisions to ensure all of our students succeed,” said Dr. Chantell Manahan, Director of Technology (MSD of Steuben County & CoSN Board Member) “While we can achieve this through hard work by intentionally designing infrastructure, building local capacity, aligning with vetted data standards, and choosing the EdTech vendors who also champion this vision, it is never easy. And we have an obligation to achieve this interoperability at scale, all while protecting student data privacy. This means ensuring access to student data on a need-to-know basis, following all federal, state, and local laws and community norms. It’s a delicate balancing act of access to the right data for the right decision-makers, and support from the federal level through the infrastructure briefs and the NETP provide the guidance and prioritization of interoperability for achieving this digital equity.”

Don’t fall behind: Steps districts can take

To keep up with emerging technologies and keep your data infrastructure and student data secure, you must ensure the foundational pieces of the data ecosystem are in place: interoperability, privacy, and cybersecurity. With that foundation in place, your technology can advance, benefiting students without worrying about security.

Use the Project Unicorn School System Data Survey tool to assess your data ecosystem. The brief survey is designed to help the education sector better understand current K-12 school system capabilities for leveraging education data. Review your results with free personalized resources and technical support from the Project Unicorn team.

Key points:

• The right data management is make-or-break in student success
• 3 strategies to streamline K-12 data management
• Reaching data readiness: 10 steps to take
• For more news on data management, visit eSN’s IT Leadership hub

We continue to read headlines about how artificial intelligence (AI) can substantially increase labor productivity across the global economy, creating both time and informational efficiencies. Yet there seems to be an argument about where and how this can revolutionize the world of education. As districts become increasingly more data driven, we need to accept that AI is here to stay and leverage information management solutions to drive student outcomes. 

Tracking and managing data plays a critical role in the success of our students, so now more than ever, there is a fundamental need for districts to implement a modern solution partner to do that heavy lifting. Manually building spreadsheets, pulling out pivot tables, and building formulas and dashboards from scratch can waste our valuable time and resources. What might take days, or even weeks, for an administrator to input can be done in seconds by the right software. Intelligent educational software instantly recognizes patterns in data, even when they are not immediately apparent. This can be crucial for identifying trends, correlations, or factors influencing student performance that might be overlooked by manual analysis. 

Ultimately, data management software allows us to shift our focus to having the right conversations, providing personalized learning experiences, delivering targeted support, and making impactful decisions to promote student success. 

Here are three ways a modern data solution can improve student outcomes in your district:    

Create collaboration among stakeholders

A solution that centralizes, consolidates, and synthesizes data can drive conversation among school counselors, psychologists, administrators, intervention specialists, classroom teachers, and anyone else invested in students’ success. It allows the team to support the whole child by seeing students’ data through a variety of different lenses, ultimately enabling classroom teachers to make informed decisions and improve the overall learning experience for their students—and this can be initially accomplished through the use of just a few key dashboards. 

As our district sees the effects of these initial dashboards come to life, we know that our students are more likely to succeed if families are engaged in their learning. A dashboard that tells the story of the whole child allows teachers to share data-driven insights with students and families, providing a clear picture of academic progress. Seeing a report card once a semester isn’t enough for families to understand how their kids are doing, and a modern data solution allows families to see trends over time in academics, mental health, attendance, and more. This provides families with clear, easy-to-understand visualizations of their child’s performance and challenges, enabling collaborative efforts to support student learning both inside and outside the classroom.

Identify patterns for at-risk students to provide real-time support

No student can be evaluated for at-risk behavior by test scores alone. By tracking attendance, behavioral patterns, mental health, grades, and other key markers, educators can pinpoint students who may be struggling and intervene before issues escalate.

The software provides data in real-time—which is crucial for identifying students at risk of academic challenges or dropping out. Data analysis software also creates algorithms and allows for the identification of patterns and trends in student performance. This proactive approach of using consistent dashboards each week allows us to quickly see which students are in the at-risk category, and drive immediate conversations about those specific students.

Create individualized support and pathways 

A modern data solution can create personalized student experiences by tailoring educational content, support, and interactions to meet individual needs and preferences. The software can do a deep dive into each student’s history and instantaneously outline what’s needed to reach personal goals in academics, career readiness, and social-emotional skill building. We can clearly see enabling students to define and track their progress toward goals and plans in our roadmap, as we continue to empower students to take an active role in managing their own future. 

We can also explore using a modern data solution to help students easily connect with community partners, courses, or other resources. This includes students accessing opportunities like: job shadowing, career mentorship, informational interviews, internships, and practicums. As schools are finding value by weaving in a variety of student-centered experiences with real-life and relevant learning activities into their curriculum, students can use the software to log hours and keep them on track. 

Data management platforms have the potential to revolutionize education in many ways, transforming the traditional model of teaching and learning. By leveraging algorithms and analytics, a modern data solution can analyze data in real-time, providing educators with immediate insights to make informed, timely decisions that positively impact student success. Ultimately, it frees educators to get back to doing what they do best—challenging and supporting every student, every step of the way.

Key points:

• K-12 educators have hundreds of data sources, but that data is siloed
• Learning to manage a sea of data can help support your district’s strategic mission
• See related article: Reaching data readiness: 10 steps to take

Think about all the data points available at your average school district.

Standardized test scores. Behavioral assessments. Attendance. And grades, of course. Not to mention, most districts are assessing grit, emotional wellness, and individual personal growth plans as well.

On top of that there are the principal’s surveys, the innumerable Google forms created by individual teachers, and whatever data is gathered by a dizzying array of edtech tools. It’s safe to say most districts have hundreds of data sources. Maybe even thousands.

So, K-12 educators should truly have a full picture. They understand the whole child—or do they?

The problem is that you won’t find all this data in one place in most districts. Everything is siloed. Nothing is centralized. Access is limited. And as a result, districts simply aren’t getting the most out of all that data. They can react to problems in the rear-view mirror, but not necessarily respond, thoughtfully, in real-time.

Here are three ways to tame that sea of data and start turning your overly complicated data ecosystem into something that supports your district’s strategic vision.

Rethink the role of the Chief Technology Officer

At most districts, the Chief Academic Officer plays a huge role in the strategic plan. Districts empower this person to make critical decisions on anything impacting student learning, while ensuring they have all the resources they need. The Chief Technology Officer (CTO) isn’t always empowered in the same way. Maybe it stems from a time, not so long ago, when we thought of the CTO in terms of hardware—the pliers-and-wires person, the person who deals with login problems and broken monitors. Whatever the reason, that has to change if districts truly want to use technology to their students’ advantage.

District leaders should ensure the CTO has far more of a voice in the big-picture plan. The CTO needs to be part of any edtech solution conversation and empowered to provide meaningful input into the strategic direction and decision-making process. The CTO needs to be integrated into the district’s comprehensive vision and must have the right skills to connect all stakeholders around the goal of managing and effectively using whole-child data.

Eliminate data silos

With all the edtech tools on the market, it’s not an easy decision-making process. While it’s tempting to look for a silver bullet—one single edtech solution that solves all your problems—the patchwork approach all but guarantees your data will be relegated to separate silos. You’ll wind up with a situation where the only way to analyze data is to gather it manually. That’s not only inefficient, but it also takes more time, and educators certainly don’t have extra time.

District leaders, with the strategic guidance of the CTO, need to prioritize interoperability to build an ecosystem of tech tools that can seamlessly exchange data. This might involve slowing down the procurement process and strategically considering how a solution fits into the district’s tech stack, how it fits into their larger strategic vision, and how it supports the whole child. K-12 educators have a strong culture, stretching back decades, of putting a tremendous amount of time and thought into adopting new curriculum. They can bring that same level of thought and sophistication to buying edtech tools.

Empower data-informed decision making at all levels

Take a look at the apps you rely on as an administrator to understand how a particular student is doing. Then ask yourself: When’s the last time a student or a parent logged into those platforms? Can teachers use them? Community partners?

Too often, data is seen as something that’s held at the district level, with tightly controlled access to even parents and teachers. Typically, it’s the job of an administrator to merge data from six or eight different places to create a report for students to see how they’re performing in real time—well, sort of.

The desks of top administrators, though crucial to overall success, aren’t where the magic actually happens. It’s not where the relationships are built and interventions are deployed. When teachers, parents, or community partners step in to help a student, they need the complete picture of what’s going on at school, such as attendance, behavior, grades, and services used. That’s why it’s critical to empower educators on the front lines. Using real-time data platforms can allow all stakeholders access to the data they need to identify teaching and learning problems. That’s the way to solve them. 

Align data management with strategic goals

If we take a step back and look at the goals of data management, we can see the true purpose. It’s not to create reports for the state education departments. It’s not data for data’s sake. Good data management isn’t an end goal. It needs to be integrated into your district’s strategic vision–actually, it’s a means to meet your district’s strategic vision. And that’s about ensuring each student has what they need for success.

Related:
3 ways schools can use data management to help students
Data doesn’t talk–people do

In today’s data-driven era, schools and districts must embrace data interoperability. Taking steps to unified K-12 data may seem overwhelming at first, but it’s OK–we’ll guide you through the process in this eSchool News webinar.

In this webinar, you’ll learn how to:

• Identify and consolidate data sources for unified K-12 data.
• Implement measures to break down data silos and improve accessibility.
• Follow the 10-step program for achieving data readiness.

Watch now to leverage the potential of unified K-12 data and make a positive impact on your school or district’s educational outcomes.

Educators across the country are seeing a greater need to collect and use data to inform decisions as they work to help students. The pandemic severely disrupted our schools, and many districts used any student data they had to identify who was struggling and how to best provide support.

As districts continue to respond to the evolving circumstances of the pandemic, making the most of all available data to improve student outcomes remains critical to understanding the factors that most contribute to students’ success.

The power of using data is immense. When used properly, it can help districts make vital decisions about setting goals and providing targeted support for students. Whether you are new to data analytics in K-12 or a seasoned veteran, here are three practical ways to apply data to help drive better student outcomes.

1. Use Data to See a Holistic Picture to Identify and Support At-Risk Students 

Educators can and should use data to gain a holistic view of each student. One data point from a single observation never tells a student’s full story. Capturing a student’s academic, behavioral, attendance, and engagement data can provide a deep, informed understanding of who the student is, where they are succeeding, and where development is needed. Dashboarding data from different areas of interest can often illuminate trends and early warning signs, lending information to identify which students might need support.

A middle school in Mississippi sought to visualize data based on their homegrown at-risk model comprised of three categories: attendance, discipline, and grades. Each category had its own risk score ranging from zero to three. Combining all three categories generated a total possible risk score ranging from zero to nine. See chart Custom At-Risk Criteria below for reference. For attendance, missing five or six days of school would yield an attendance risk of two, trending toward high risk for absences. Assuming that same student missed no additional days of school, had no disciplinary events, and all of his grades were higher than 70, their total at-risk score would remain two.

Specifying a unique and multi-tiered rubric for each risk category provided a rich amount of information and a natural way to parse and analyze data. In this instance, school administration discovered that chronic absenteeism accounted for the most risk among their student population, with 97% of students having at least one risk point attributable to absences. Disciplinary events were overall negligible, with few overall risk points coming from this category. Risk based on low performance in the classroom revealed an interesting but troubling pattern. Though few students were at risk due to having low classroom grades, most students within this group had an overall high-risk score (an average of six). Moreover, this data revealed that students who were failing one classroom subject were usually failing at least one other subject as well.

Filtering and comparing results by grade level and other demographic factors allowed educators to see if differences emerged based on students’ current circumstances (e.g., experiencing homelessness or being in an after-school program). In other words, this data informed whether some students, more than others, were more or less frequently observed as overall high risk or high risk by particular categories.

2. Use Data to Set Goals and Target Interventions

One thing most educators can agree on is that all students learn differently. Thus, student goal setting and interventions are often tailor-made to the unique needs of each student. Seeing where a particular student falls on the at-risk model enables educators to set specific, individualized goals for students and target interventions where most needed. While each school will face unique challenges, data can help all schools spot trends and determine areas of priority.

Continuing with the prior example, having set and applied clear criteria to identify at-risk students, the district could then turn its attention to implementing an intervention. An example could be to reduce the number of students acquiring the highest risk score the following year, and set a goal of meeting a specific (but reasonable and attainable) threshold of success for the next three years (e.g., reduce by 10 percent the first year). One such approach includes selecting one or two areas of focus at one or two campuses initially. Further breaking down results of the at-risk model by individual categories would inform which areas of risk are the highest priority, and would offer the most return on investment if successfully targeted for intervention. In the current example, students with failing grades tend to have the overall highest score, suggesting a potential area to target, or seek further information.

3. Use Data to Support Meaningful Parent-Teacher Communication

Sharing the students’ data with their families can help to drive ongoing meaningful communication between school and home. Using data, educators can inform parents about their students through data-driven conversations.

Offering specific information about a child’s attendance or behavior can help with creating more meaningful relationships with home. Having data enables teachers to talk specifically about the number of attendance or behavior incidents for a student in a month and have a two-way conversation with a parent about supporting the child’s improvement. For instance, being able to see that a student has been chronically absent this year, but has never had an attendance problem in past years suggests the family might be experiencing a unique and stressful situation. Only with data at hand would this insight be possible, allowing the opportunity for the educator to ask if everything is okay at home. On a lighter note, having student-level data that is longitudinal and includes both quantitative and qualitative data offers a wonderful opportunity to see how a student has progressed over time, which is always a fun conversation to have with caregivers.

There is great potential for data to serve as a tool for educators and administrators who want to improve schools and help all students reach their potential. Collecting and analyzing student-data can help create a holistic view of a student, set appropriate, individualized goals and targeted interventions, and support meaningful parent-teacher communication. This can all help students along the way and ensure greater outcomes.  

A media friend was looking into a recent Vanderbilt study on six unidentified school districts across the state. The researchers found that “more students were chronically absent this fall than in previous years, and absenteeism increased the most among English Learners, students of color, and students who are economically disadvantaged.” I told her I was not concerned about the latest findings.

When pressed on the issue, I pointed out that the sample size was a little concerning. Only 6 districts were represented; my guess is they looked at the larger urban areas. I find the research misleading. There are 147 districts in the state, each with unique and distinct issues. If we rush in and try to apply a one size fits all solution to any issue, we would be making a mistake.

By failing to identify the six individual districts, the results from the research were problematic to me—as well as to other stakeholders and policymakers as well. The study has some interesting findings. It would be useful to those 6 unidentified districts. However, I am not certain there is a crossover for other districts.

It is worth focusing on the fact that there was a global pandemic ongoing. A pandemic that nobody saw coming, or that we had ever dealt with previously. Education was largely online, so that may point out why low-income, ELL students, and students of color did not have internet access or had difficulty accessing online school.

In addition, some students were still having an issue getting devices initially. The research period was over by the first semester and did not allow for adjustments that may have occurred. I am concerned about replicability in the study, and subsequent reliability. Even though it was a fascinating read with interesting thoughts, I am not sure I can go much beyond that at this time.

Professor Andrea Jones-Rooy brilliantly stated in Quartz: “Data doesn’t say anything. Humans say things. We’ve conflated data with truth. And this has dangerous implications for our ability to understand, explain, and improve the things we care about.” Data is important, but it can be manipulated and generated in the way questions are asked. The value is often suspect, and conclusions can be pre-determined.

Gavin Freeguard, using a similar theme, pointed out that, “badly presented data will be confusing to people inside government too. If we as the general public are having difficulty understanding messages across (and within) datasets, it suggests those inside government are having similar struggles. Indeed, making sense of the data is made nearly impossible by the sheer number of sources in use.”

In recent years, we have seen an overreliance on research from sole sources. In our state’s public education, much of the education research is done at Vanderbilt or Gates Foundation-funded research through various education nonprofits. That is not meant to be disparaging, but rather an observation.

However, the question we must ask ourselves is, “What is the possibility of the research being replicated or reproduced by other research institutions or organizations? Will the findings be the same?” We must recognize a susceptibility to biases in research. Bias can occur in the planning, data collection, analysis, and publication phases of research. Many biases work subconsciously, are undetectable, and ultimately not correctable. We must be aware of inconsistencies between actual results and preconceived expected outcomes.

This brings us back full circle. Was Vanderbilt’s research accurate and unbiased? Perhaps. Then again, the data may have been distorted by how the questions were asked or by some unforeseen bias. Who was the audience? Who paid for the research? What was the objective of the research? We all want to improve public education, but a little context on the findings would have been helpful—especially identifying the districts used in the study and identifying who paid for the research.

Too often media is off and running with stories about research without critical details. My friend happened to drill down and ask the tough questions. For too long we have been shallow when it comes to facts, and excellent when it comes to entertainment. Don Henley sings “Long Way Home,” which contains a great lesson: “There are three sides to every story–yours and mine and the cold, hard truth.” Such is with some research and news. It is up to all of us to determine what is true, and what is not. Data doesn’t talk, people do!

One of the most powerful tools in fighting the COVID-19 pandemic has been data. Data analytics has informed what we can do, when we can do it, and has kept us safe. As more schools reopen their doors, data is also playing a vital role in ensuring they do so safely.

The Centers for Disease Control (CDC) issued key indicators for dynamic school decision making, which include measures of underlying community transmission as well as a measure of adherence to key mitigation strategies. While these indicators provide a solid foundation for establishing and maintaining in-person plans, they aren’t always inclusive of the data that parents, teachers, and staff need to feel comfortable about returning to the classroom.

To gain buy-in from these stakeholders and help inform plans as schools reopen, schools districts must also consider four under-the-radar-data points. Let’s take a look.

1. Access to high-speed internet

The pandemic and subsequent stay-at-home orders exposed a glaring inequity in access to high-speed internet in the home. An April 2020 survey, spearheaded by the U.S. Census Bureau, found that 3.7 million households with children had internet available “sometimes, rarely, or never.” This disparity is felt most acutely in rural America where only 65 percent of Americans have access to fixed internet.

The consequence for a child’s education is significant. In the past year, students have struggled with failing grades while unprecedented numbers simply disappeared from classes, forcing districts to rethink their approach to getting them to show up.

As schools mull reopening plans, a student’s access to this basic service should be a key consideration. Publicly available data sources such as Broadband USA, the National Broadband Map, and Education Superhighway can help school districts and government leaders learn about connectivity in their state and prioritize in-school learning for those students who can’t participate in virtual education.

2. Availability of vaccine

Until an approved vaccine is available for children under 16, faculty, staff, and teachers are rightly concerned about potential disease transmission in schools and the risk of bringing infection back home to their families as schools reopen.

To minimize health risk and alleviate these concerns, it’s important that school districts gather data on the availability of the COVID vaccine and the percentage of teachers and school staff who opted to receive it. Typically available from state departments of health, this data can be used to prioritize physical reopening decisions.

But it can also be combined with other information to develop a more customized approach based on circumstances and need. For example, staff and teachers who lack broadband internet at home but have been vaccinated may be prioritized for a return to school. Alternatively, if a school is in an area with low vaccination rates but a high percentage of homes with high-speed internet access, then leaders may err on the side of caution and manage learning virtually until vaccination rates increase.

Schools could also give vaccinated teachers the option to return for in-person schooling for those vulnerable student populations who can’t participate in or are struggling with remote learning.  Fortunately, the recent increase in vaccine availability should limit this problem for faculty and staff by the start of the next school year. 

3. Food insecurity

Many students rely on school breakfasts and lunches for their core meals of the day. Some localities have pivoted and offer meals at schools (transportation permitting) or provide meal subsidies to families. While these measures address immediate need, states must also plan for the future.

Millions of jobs were eliminated or put on hold during the pandemic, and many may never come back. It’s also unlikely that schools will be fully operational for some time, placing a continued strain on family pocketbooks.

While ensuring children have access to affordable, nutritious food is not the responsibility of schools, as they plan for the prolonged impacts of the crisis, it’s an important conversation for school boards and social service agencies to have.

This is the case in Virginia. Through data sharing and data analytics, the commonwealth is helping families continue to put food on the table – even those who were previously ineligible. Since many of these recipients weren’t in the social services’ system, the state pioneered an inter-agency data initiative to identify eligible families and quickly issue 440,000 new food benefit cards to those in need.

4. Affordability of care

In the face of COVID-19, the cost of childcare has skyrocketed. Increased demand and the need to meet stringent safety guidelines saw licensed childcare center rates soar by 47 percent while home-based family childcare costs have increased by an average of 70 percent.

Under pressure from parents to reopen, school boards should leverage childcare data to inform their decision-making and prioritize returning impacted students, teachers, and staff. Data points to consider include the availability and cost of childcare in the area, and whether a student has working or non-working parents or guardians.

Government leaders can also leverage data to help families find care. For instance, some states are using data from childcare licensing records and surveys to analyze which childcare facilities are open, at what capacity, and mapping that data so parents can quickly understand their options.

Tracking the metrics that matter most may require policy changes

Guidelines from the federal government and anecdotes from how other schools are managing to reopen are only helpful to a point. In the face of a crisis, school leaders must find ways to track the metrics that matter the most to their communities. In most cases this means uniting previously siloed data from across a variety of government sources to unlock insights and reach consensus about back-to-school plans. In many states this may require changes to data sharing policies that are often codified to keep that data under the control of the “data owner.”

If data is to keep answering the questions that arise during these fluid and dynamic times, it’s important for schools and government leaders to factor these under-recognized considerations into their discussions so as to help build broader consensus among parents, teachers, staff, and families.

Schools worldwide face difficult choices as they try to balance student and staff safety with their educational mission during the pandemic. All-remote learning eliminates the risk of an outbreak at the school, but at-home education doesn’t provide the most productive learning environment for everyone, and it can put at-risk students at a disadvantage for a variety of reasons.

Schools and districts are making decisions about their approach by taking into account factors like virus prevalence in their area and student and family needs. An August 2020 McKinsey report notes that in the U.S., about 75 percent of the 50 largest districts started this school year on an all-remote basis. Others returned to all in-person learning, and some had to shut down again due to outbreaks.

Some districts have adopted a hybrid model where online instruction alternates with in-person teaching in classrooms that have reduced class sizes and social distance protocols in place. Hybrid education can help schools address the needs of vulnerable learners, including younger children who are difficult to engage online, as well as low-income and special education students, along with English-language learners.

Data Is the key to hybrid learning success

A hybrid approach can give districts the flexibility they need to safely transport students to school and provide in-person instruction at a safe distance. But maintaining two instructional models makes it more important than ever to monitor education data closely. Teachers and administrators need to collect and analyze data so they can identify trends and address problems to improve student success.

Critical data in aggregate form might include enrollment numbers, attendance, course completion rates, graduation rates, and demographics like age, race, gender, economic status, and special education needs across the student population. On a classroom level, key data can include grades on quizzes, tests, and education assessment instruments; engagement levels; and teacher observations.

Survey data from students and parents can be critical in building an effective hybrid program or making adjustments along the way. In a recent report on hybrid learning, Bree Dusseault of the University of Washington’s Center for Reinventing Public Education recommends surveying students and parents to gauge preferences and maximize hybrid program effectiveness.

Establish a baseline and use data to monitor progress

Before educators can help students move forward, they have to understand their current academic status. Assessment tools like standardized tests can help. Once they have established a baseline, educators can use data to monitor progress toward educational goals. Analytics tools can also help teachers and administrators spot broad trends, such as a drop in grades during the post-holiday period.

With data that points to clear patterns, educators can take steps to address a negative trend, like providing additional instruction or enlisting parents in the effort to keep students focused. At-home education is tough on working parents, but more data on children’s performance can help teachers keep parents informed and indicate where a student needs extra attention.

Analysis of aggregate data can help educators gain insight into the needs of underserved groups and create more balanced educational opportunities. Surveying parents and students can help schools identify obstacles to at-home learning, such as broadband limits or technology barriers, so the school can plan accordingly and ensure that everyone has the resources they need.

Choose the right data-collection tools

Most educators have access to at least some district, state, and national data that can help them make decisions, but they often need more in-depth information. Secure, cloud-based data-collection tools can help teachers and administrators collect the targeted data they need to optimize hybrid learning and keep track of online and in-classroom activities.

For example, cloud-based forms allow educators to distribute online quizzes, collect signups for extracurricular activities or class registration, and distribute course evaluations. Educators can use online survey forms to gain valuable data from students and/or parents to better understand student needs. They can then use that information to design an effective hybrid program.

Although vaccine distribution is a hopeful sign that educators can return to more normal operations soon, schools around the world are going to be dealing with remote instruction for the foreseeable future, and hybrid arrangements may be here to stay. Schools that collect and analyze data to design a program that meets student needs will be best positioned to adapt to whatever comes next.