Pearson's Decoding Adaptive
- Nimisha Asthagiri (Deactivated)
Owned by Nimisha Asthagiri (Deactivated)
Intro
This document captures notes from Pearson's Decoding Adaptive.
"The tools, however, are not a panacea. For several reasons, it’s unlikely that a single tool will ever be able to take over a student’s education and direct them to every single thing they should do. Nor is it likely that we would want it to, as a critical part of education is building student agency – helping students own their learning, make decisions, become lifelong learners, and develop their metacognitive skills."
Goals
"... helping every student achieve his or her maximum potential through differentiation"
Benefits
- Maximize Learning (Efficiency)
- Optimize time - what students need when they need it, with deeper meaningful interactions with teachers and peers.
- Real time - students know now, rather than what they knew in the placement test.
- Real data - used to automatically respond to a student’s needs.
- Precision (Effectiveness)
- Precise recommendations for students.
- Deeper understanding of students levels.
- Individualized interventions (when and how).
Challenges
- Choice and Control
- Flexibility to override mastery levels, student skill levels, recommendations, etc.
- Customize to teacher preference for sequencing and pacing.
- Question: Is a dynamically created learning path actually more effective and efficient for students over time?
Features
- Individualize: Capacity to select the best content for students.
- Analyze: Capacity to collect data on how students learn.
- Pedagogy: Capacity to use collected data to reveal how students learn.
Expectations
"The next generation of adaptive learning tools .. help students improve their learning processes .. such as motivation, creativity, perseverance, and self-regulation."
- Start with a clear educational vision.
- The best adaptive learning includes teachers and technology.
- Teacher: provides personal and social learning.
- Technology as teaching assistant: provides support and captures data.
- Teachers need user-friendly data, across multiple products.
Adaptive Learning Tools
Education technologies that can respond to a student’s interactions in real-time by automatically providing the student with individual support.
- What are digital adaptive learning tools?
- Education technologies that can respond to a student’s interactions in real-time by automatically providing the student with individual support.
- What are NOT digital adaptive learning tools?
- Tools that mark an answer as correct or incorrect and then provide one singular path for learning, regardless of the student’s response.
- Tools that don’t collect data in real-time.
- Tools that collect data through one singular assessment and prescribe a path of learning, but don’t collect data or provide support in real-time.
Adaptive Learning Tool Types
Adaptive Content
- Targeted feedback and hints aligned to student's specific misunderstanding or based on student's previous responses.
- Additional review material (e.g, videos, text, private tutoring services).
- Scaffolding, without changing the original sequence of skills.
Adaptive Assessment
- Assessments change and respond based on
- Correctness of student's responses.
- Difficulty level of questions.
- Assessments used as
- Practice engines, which come after a lesson, where students continue to answer questions until they reach mastery level before moving on.
- Benchmark tests, given periodically to reassess the students academic progress; may create forward learning path; may analyze large groups to ensure reliability and validity of the tests.
Adaptive Sequence
- Continuously collect real-time data on performance and use it to automatically change a student’s learning experience, using algorithms and predictive analytics.
- In contrast to differentiated content, where new learning paths are assigned based on performance on on interleaved assessments.
- Three step process:
- Collect data (answering questions, clicking on hints, or using virtual manipulatives)
- Analyzes data
- Adjusts the next content for the student
Collect Data
- Types of Data (to create a more complete picture of their abilities)
- Academic performance: e.g., answers to problems - what students know
- Learning process: e.g., number attempts, resources used (calculator, etc) - how students learn
- Student interest: e.g., repeated access to a learning resource - why and how students learn
- Less commonly used: social behaviors, ratings, mood
- Difficulty Level
- Webb’s Depth of Knowledge
- Bloom’s Taxonomy
- simply: easy, medium, and hard
- Granularity: general topic, specific concept or skill
- Learner History: tool creates a profile of the learner’s interactions with the content, which continues to grow as the student uses it.
Analyze Data
- Learner Analysis: analyzing student performance data
- Weighting categories of data
- Applying thresholds of mastery
- Comparing groups of students data, by similar profile
- Calculating probability of mastery
- Applying rules for correct and incorrect responses
- Skill Selection: how many skills to present next
- One option: Predesigned content map (of sequences)
- Few options: Return to previous units or skills
- Infinite options: Any skill successfully completed by students of a similar profile
- Content Analysis: selecting content to present next
Adjust Content
- Delivery
- Assigned content
- Recommended content
- Amount
- Individual piece of content
- Group of content
- Design
- Interrelated content (e.g., in a sequence)
- Independent content (e.g., in a playlist)
Adaptive Sequence Example Types
| Phase | The All-in-one | The Comparison | The Recommendation | Educator Do-It-Yourself (DIY) | |
|---|---|---|---|---|---|
| Description of example type | Adjusts content based on skills the student should work on next uses weights and thresholds. | Adjusts content by selecting a skill based on skills that similar students previously needed/used. | Adjusts content by recommending activities that similar students found useful. | Adjusts content as predetermined by the teacher for each response. | |
| Collect | Type | academic performance data, learning process data | academic performance data, learning process data | academic performance data, learning process data, interest data | academic performance data |
Granularity + | discrete skill, specific concept, difficulty level | specific concept, difficulty level | general standard or topic | determined by the author | |
| History | learner’s profile over time | learner’s profile over time | learner’s profile over time | -- | |
| Analyze | Learner analysis | weighting categories, applying thresholds of mastery | comparing groups of learners’ data | comparing groups of learners’ data | correct or incorrect response |
| Skill selection | few options | infinite options | -- | one option | |
| Content analysis | -- | -- | usage, interest, effectiveness | -- | |
| Adjust | Delivery | assigns content | assigns content | recommends content | assigns content |
| Amount | group of content | individual content | group of content | determined by the author | |
| Design | related content | independent content | independent content | determined by the author | |
| Examples | DreamBox, Mathspace | Knewton, Waggle, CogBooks, SuccessMaker | Fishtree, Brightspace LeaP, MyLabs | Smart Sparrow | |