Round 115

Redesigning Middle School Math Grouping

A public school district's middle schools currently track students into 'advanced' and 'standard' math classes starting in 6th grade based on a single placement test taken in 5th grade. Data show this system correlates strongly with family income and produces widening gaps by 8th grade, but the advanced track also seems to accelerate genuinely capable students who might otherwise be under-challenged. The district curriculum committee must recommend a single model to pilot next year across all middle schools. Goals: reduce inequities tied to income/testing bias, avoid demotivating either high- or low-performing students, keep teacher workload and training demands feasible, and maintain rigor for students heading toward advanced high school math (calculus-track). Constraints: limited funding for extra teacher aides or curriculum redesign, parent/community attachment to existing 'advanced' label, and a need for a decision within one budget cycle. Tradeoffs include equity vs. perceived rigor, simplicity of implementation vs. personalization, and short-term parent backlash vs. long-term outcome data that will take years to confirm.

Dashboard

Status

DECIDED

Winner

B · Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks.

Majority

B · Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks.

Duration

5999 ms

Topic

Options

B Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks.
2.30 of 3 PPV winner
C Adopt untracked classes for 6th and 7th grade to build shared foundations, then introduce an optional accelerated pathway in 8th grade open to any student who requests it.
0.70 of 3
A Replace tracking with a within-class rotating model where students move fluidly between skill-based groups on a unit-by-unit basis depending on demonstrated mastery of that specific topic.
0.00 of 3
D Eliminate tracking entirely and place all students in mixed-ability classes, using in-class differentiation (tiered assignments, small-group instruction) supported by teacher training.
0.00 of 3
E Keep the current two-track structure unchanged but add supplemental tutoring and enrichment programs targeted at low-income students to narrow the placement gap without altering the tracking mechanism itself.
0.00 of 3

Checks

Baselines

PPV no gold answer
Majority no gold answer
Comparison PPV matched majority
k samples8
Agents3
Parsed samples 24 / 24
Quality near unanimous split agent disagreement

PPV

Top Influences

GPT 5.4 mini · delegate 1.150
Gemini 3 Flash Preview · delegate 1.150
Claude Sonnet 5 · delegate 1.000

Samples

Answer Distribution

Claude Sonnet 5 C: 6 B: 2
GPT 5.4 mini B: 8
Gemini 3 Flash Preview B: 8

PPV

Influences

Label Role Score
GPT 5.4 mini delegate 1.150
Gemini 3 Flash Preview delegate 1.150
Claude Sonnet 5 delegate 1.000

Voting matrix

Constructed Delegation Matrix

To / From Claude Sonnet 5 GPT 5.4 mini Gemini 3 Flash Preview
Claude Sonnet 5 0.0000.0000.000
GPT 5.4 mini 0.1500.0000.000
Gemini 3 Flash Preview 0.1500.0000.000
A 0.0000.0000.000
B 0.0001.0001.000
C 0.7000.0000.000
D 0.0000.0000.000
E 0.0000.0000.000

Embeddings

Sample Embeddings

24 samples
Claude Sonnet 5 sample #0 · C · 1536 dims C Claude Sonnet 5 sample #1 · C · 1536 dims C Claude Sonnet 5 sample #2 · B · 1536 dims B Claude Sonnet 5 sample #3 · B · 1536 dims B Claude Sonnet 5 sample #4 · C · 1536 dims C Claude Sonnet 5 sample #5 · C · 1536 dims C Claude Sonnet 5 sample #6 · C · 1536 dims C Claude Sonnet 5 sample #7 · C · 1536 dims C GPT 5.4 mini sample #0 · B · 1536 dims B GPT 5.4 mini sample #1 · B · 1536 dims B GPT 5.4 mini sample #2 · B · 1536 dims B GPT 5.4 mini sample #3 · B · 1536 dims B GPT 5.4 mini sample #4 · B · 1536 dims B GPT 5.4 mini sample #5 · B · 1536 dims B GPT 5.4 mini sample #6 · B · 1536 dims B GPT 5.4 mini sample #7 · B · 1536 dims B Gemini 3 Flash Preview sample #0 · B · 1536 dims B Gemini 3 Flash Preview sample #1 · B · 1536 dims B Gemini 3 Flash Preview sample #2 · B · 1536 dims B Gemini 3 Flash Preview sample #3 · B · 1536 dims B Gemini 3 Flash Preview sample #4 · B · 1536 dims B Gemini 3 Flash Preview sample #5 · B · 1536 dims B Gemini 3 Flash Preview sample #6 · B · 1536 dims B Gemini 3 Flash Preview sample #7 · B · 1536 dims B
Claude Sonnet 5 GPT 5.4 mini Gemini 3 Flash Preview

Samples

Agent Responses

Claude Sonnet 5 openai_compatible · anthropic/claude-sonnet-5
#0 C · Adopt untracked classes for 6th and 7th grade to build shared foundations, then introduce an optional accelerated pathway in 8th grade open to any student who requests it. 1536 dims

Option C balances equity and rigor best: untracked 6th–7th grades reduce early income/testing bias and prevent premature labeling, while an opt-in accelerated 8th-grade pathway preserves a calculus-track option and addresses parent attachment to advanced oppor...

#1 C · Adopt untracked classes for 6th and 7th grade to build shared foundations, then introduce an optional accelerated pathway in 8th grade open to any student who requests it. 1536 dims

Option C phases in shared foundational instruction to reduce early income/testing-based sorting, while preserving a rigorous, opt-in accelerated pathway by 8th grade that maintains calculus-track access and addresses parent attachment to advanced options. It r...

#2 B · Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks. 1536 dims

Option B addresses the core equity problem (a single biased 5th-grade test) while preserving the "advanced" label parents value and maintaining a rigor pathway for calculus-track students, making it politically feasible within one budget cycle. It requires onl...

#3 B · Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks. 1536 dims

Option B directly addresses the root inequity (single high-stakes 5th-grade test tied to income) by broadening placement criteria and enabling annual movement, without requiring a full curriculum overhaul or heavy new staffing—fitting the one-budget-cycle cons...

#4 C · Adopt untracked classes for 6th and 7th grade to build shared foundations, then introduce an optional accelerated pathway in 8th grade open to any student who requests it. 1536 dims

Option C best balances the constraints: it delays high-stakes tracking until 8th grade, reducing early income/testing bias while giving all students shared foundational rigor in 6th-7th grade. Making the accelerated pathway open by request (rather than gatekep...

#5 C · Adopt untracked classes for 6th and 7th grade to build shared foundations, then introduce an optional accelerated pathway in 8th grade open to any student who requests it. 1536 dims

Option C offers a balanced pilot: shared untracked foundations in 6th-7th grade reduce early income/test-driven sorting and stigma, while an opt-in accelerated 8th-grade pathway preserves rigor and calculus-track access without requiring costly full-scale diff...

#6 C · Adopt untracked classes for 6th and 7th grade to build shared foundations, then introduce an optional accelerated pathway in 8th grade open to any student who requests it. 1536 dims

Option C phases untracked instruction early to reduce income/testing-based sorting and build equitable foundations, while preserving a rigorous accelerated pathway in 8th grade for calculus-track students, satisfying parent attachment to advanced options. It's...

#7 C · Adopt untracked classes for 6th and 7th grade to build shared foundations, then introduce an optional accelerated pathway in 8th grade open to any student who requests it. 1536 dims

Option C balances equity and rigor by delaying high-stakes tracking until 8th grade, giving all students two years of shared, unbiased foundational instruction that reduces income-driven gaps. It preserves an "advanced" pathway (addressing parent attachment an...

GPT 5.4 mini openai_compatible · openai/gpt-5.4-mini
#0 B · Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks. 1536 dims

The best choice is the option that most directly reduces income-linked placement bias while still preserving a route for advanced students and staying feasible within one budget cycle. Option **(B)** keeps the familiar two-track structure, but improves fairnes...

#1 B · Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks. 1536 dims

The strongest choice is the one that most directly reduces income-based sorting while still preserving a pathway for students who are ready for advanced math. Option (B) keeps the familiar track structure and advanced label, but makes placement broader and mor...

#2 B · Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks. 1536 dims

The best choice is the option that most directly reduces income-linked bias while still preserving a path for students ready for advanced math. A yearly, multi-measure placement system with movement between tracks is more equitable than a single test, keeps th...

#3 B · Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks. 1536 dims

The best choice is the option that most directly addresses income-based bias while still preserving a path for advanced students and staying feasible to implement. Option (B) keeps the familiar two-track structure, reduces reliance on a single biased test by u...

#4 B · Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks. 1536 dims

The committee needs a model that directly reduces income-linked placement bias while still preserving an accelerated path for students who need it. Option (B) keeps the basic two-track structure for rigor and parent familiarity, but improves fairness by using ...

#5 B · Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks. 1536 dims

The committee needs a model that reduces income-based bias, preserves a path for advanced students, and stays feasible to implement within one budget cycle. Option (B) keeps the familiar two-track structure for community acceptance but makes placement and move...

#6 B · Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks. 1536 dims

The best choice is the option that most directly reduces income-linked sorting while still preserving a pathway for students who are ready for advanced material. Option (B) keeps the familiar advanced/standard structure for parents and rigor, but makes placeme...

#7 B · Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks. 1536 dims

The committee wants a model that reduces income-based bias, preserves rigor for advanced students, and stays feasible within limited funding and one budget cycle. Option (B) best balances these goals: it keeps a familiar two-track structure to reduce backlash,...

Gemini 3 Flash Preview openai_compatible · google/gemini-3-flash-preview
#0 B · Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks. 1536 dims

Option (B) addresses the issue of testing bias by using multiple measures and allows for mobility between tracks, satisfying the goal of reducing inequities while maintaining the "advanced" label parents value. Unlike option (A), (C), or (D), it keeps teacher ...

#1 B · Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks. 1536 dims

Option (B) addresses the issue of testing bias and income correlation by using multiple entry metrics and annual reassessment, allowing for upward mobility while maintaining the existing "advanced" label parents value. This model is more feasible within one bu...

#2 B · Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks. 1536 dims

Option (B) addresses the goal of reducing income-based inequities by replacing a single high-stakes test with multiple data points and allowing annually reassessed movement between tracks. This approach maintains the "advanced" label and rigor that parents val...

#3 B · Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks. 1536 dims

Option (B) addresses equity concerns by using multiple assessment measures to reduce bias from a single test, while also providing a "reassessment" mechanism to fix initial misplacements. It honors the community attachment to the "advanced" label and maintains...

#4 B · Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks. 1536 dims

The school district faces a conflict between maintaining perceived rigor and reducing socioeconomic inequities, all while working under strict budget and time constraints. Option (B) directly addresses the bias of the single placement test by introducing multi...

#5 B · Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks. 1536 dims

Option (B) addresses the issue of testing bias and income correlation by moving away from a single-point-in-time assessment to a more holistic, recurring evaluation. It maintains the "advanced" label and tracking structure, which satisfies parent demands and k...

#6 B · Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks. 1536 dims

Option (B) balances equity and rigor by replacing a biased single-point entry with a more holistic, annual assessment process, allowing for mobility without the high costs of a total curriculum redesign. This approach addresses income bias through multiple mea...

#7 B · Keep the two-track system but replace the single 5th-grade test with multiple measures (grades, teacher recommendation, and a test) reassessed each year, allowing movement between tracks. 1536 dims

Option (B) offers the most practical solution within the given constraints by addressing testing bias through multiple measures while preserving the "advanced" label that parents value. It minimizes teacher workload and budget demands compared to radical redes...

Timings

Stages

Sampling5347 ms
Embedding641 ms
Signals10 ms
PPV1 ms

Errors

Agent and Round Errors

No errors recorded.