What’s a Teacher to Do? Changing the Way We View the Teacher’s Role in Data Analysis Through the Use of Educational Technology
Teachers and school leaders have it regularly drummed into their heads that teachers need more training in data analysis. There is great consensus that teachers should continually pore over the voluminous amount of information available to them regarding student performance, and they are expected to gather more of this information continually throughout the school year. Using finely-honed skills to draw answers from their collected data, teachers are then expected to determine which students need which types of instruction or intervention at any moment in order to keep students moving forward, ultimately boosting the student’s academic achievement.
There is no denying that personalized instruction, which refers to instruction in which the pace of learning and the educational approach are optimized for the needs of each learner, can be highly effective in improving student outcomes. However, this has created a huge shift in teacher roles and responsibilities. Just ask any reading specialist: many of these educators are relegated to the role of exam proctor. Just ask any elementary school teacher, who may be required to do multiple benchmarking assessments every quarter that take away from valuable time for instruction.
No one denies that data-driven instruction allows teachers to personalize learning for each of their students. But, data-driven instruction requires data analysis, which is a time-consuming task that many teachers view as yet another distraction from teaching and interacting with students. What’s a teacher to do? Perhaps the solution lies in asking ourselves an entirely new question: What are a teacher’s strengths, and how can we leverage alternative solutions to help bridge the gaps?
Understanding Teacher Strengths
The average education degree devotes little-to-no time on data analysis and interpretation. A Department of Education study conducted in 2008 found that, “teachers’ likelihood of using data in decision making is affected by how confident they feel about their knowledge and skills in data analysis and data interpretation. Once a teacher begins working in the classroom, we count on professional development to fill in the gaps. Unfortunately, teachers don’t usually find PD useful for improving their skills. According to a recent study:
For the most part, teachers did not find assessment focused professional development to be of high quality or useful, which suggests that the related professional development was not aligned with the needs of teachers. Given that 40% of district-level professional development time was focused on the use of student learning assessment data, this result is discouraging as it indicates that teachers are largely dissatisfied with the quality of a major portion of their professional development opportunities related to assessment data.
A survey performed by the Bill & Melinda Gates Foundation found in a recent survey that 91% of teachers report using some form of analytics solution, but only 32% found the data interpretation to be useful, which also can be interpreted to imply that teachers are being made to participate in data analytics when they may not perceive this activity to be relevant for them. They also found that teachers commonly express a desire to spend more time with students, not spend time analyzing data and engaging in other similarly administrative tasks: “the proliferation of data and the growing expectation that they be put to use has added additional tensions and tradeoffs for teachers.”
And, from a common-sense point of view, would it be presumptuous to assume that a teacher’s strength is in their ability to teach? That teachers choose to become teachers to work closely with students, and not data? Just as we expect a surgeon to perform surgery, or a hairdresser to cut hair, we should expect teachers to excel at teaching, not data analytics. The goal is to find a solution that allows teachers to teach in data-driven ways without having to analyze and interpret the data themselves.
Leveraging Technology for Data Analysis and a Brief Case Study of Lexplore
We have been aware of the ability of technology to improve education for quite some time—a 2006 primer for the National Conference of State Legislatures on educational technology states:
A growing body of evidence demonstrates that technology is an effective means for addressing educational needs, goals and requirements. Educators also have identified links between technology and intermediate goals that lead to high achievement, including improved student behavior, engagement and attendance; improved opportunities for educator professional development; increased efficiency in classroom administrative tasks; and improved communication among stakeholders, including parents, teachers, students and administrators.
The U.S. Department of Education’s stance on technology in education runs along similar lines:
Technology ushers in fundamental structural changes that can be integral to achieving significant improvements in productivity. Used to support both teaching and learning, technology infuses classrooms with digital learning tools, such as computers and handheld devices; expands course offerings, experiences, and learning materials; supports learning 24 hours a day, 7 days a week; builds 21st-century skills; increases student engagement and motivation; and accelerates learning.
So why not use a tool that has a track record of improving educational outcomes to solve the problem of data analytics? The steady progression and integration of technology into classrooms has created a market for educational technology that can help to solve the problem of data analytics. A wide range of companies, such as industry giants ConnectEDU, Edmentum (formerly PLATO Learning) and the ubiquitous textbook publisher McGraw-Hill, or small-and-medium-sized firms such as Knewton or Lexplore, have developed technologies that serve as data analysts and interpreters, allowing teachers to focus on classroom learning instead.
Lexplore is a rapid reading assessment that uses eye-tracking technology and artificial intelligence to determine overall reading ability, risk for dyslexia, fluency level, comprehension level and recommendations for instruction. This technology offers educators a digital, evidence-based way of measuring student reading levels that is fast and objective, allowing educators more time to provide the personalized supports that students need to help them grow as readers.
The development of Lexplore is based on over 25 years of research (including a longitudinal study) at the Karolinska Institute in Stockholm, Sweden. Known as the Kronobergsprojektet, this study, which was launched almost 30 years ago, examined reading difficulties and tracked the eye movements of children—both those with difficulties and those without. Eye-tracking during reading was conducted on children in 3rd grade, and reading difficulties were assessed until adulthood. What makes the Kronobergsprojektet study unique is that children were monitored over a long period of time and the integrity of the recorded eye movement data was able to be maintained. Something that could not be done at the time but which is possible now—advanced statistical analysis—is key to the Lexplore method. By analyzing eye movement patterns from the Kronobergsprojektet study, Lexplore professionals were able to show that statistical models based on the data could, with a high degree of accuracy (95.6%), predict which students were experiencing difficulties after as little as 30 seconds of reading. The analysis also showed that the method provides a good balance between sensitivity (ability to find all of the children who are struggling) and specificity (ability to distinguish which children are not struggling).
While Lexplore is only one type of educational technology focused on solving the issue of data analysis, it provides an example of the type of experience educators want to have when utilizing technology: an experience that allows them to tailor instruction without being data analysts.