Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for statisticians

Sensory Panel Analysis Agent

A consistent sensory analysis in which unreliable panelists are identified and their effect on conclusions is shown.

AnalysisScience and research
See how it worksGet access
Sensory Panel Analysis Agent: what goes in, what the agent does and what you get

What it does

A sensory session produces scores from 12 panelists and one of them rated everything the same, while another reversed the scale. This agent checks panelist agreement and replicate consistency first, using panel performance statistics. Then it runs the planned tests, such as analysis of variance and post hoc comparisons, and flags panelists or products that behave oddly. It reruns the analysis without the flagged panelists and compares the conclusions. If conclusions change, it reports both. It writes up the results with the test details. You approve the report. Edge case: when a product is only tested by some panelists, the analysis uses a design that allows incomplete data.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueYes, continueApprovedNoNo 1 STARTS WHEN Panel data exported 2 USES A TOOL Load scores and check data completeness 3 DOES Calculate panelist agreement and replicateconsistency 4 CHECKS THE RESULT Is every panelist above the performance thresholds? If not: flag the panelists and note patterns such asscale reversal or no variation. Back to step 3. 5 DOES Run the planned tests on all data 6 DOES Rerun the tests without the flagged panelists 7 CHECKS THE RESULT Are the conclusions the same with and without theflagged panelists? If not: report both results and mark the conclusion asdependent. Back to step 5. 8 DOES Write the results with test details and productdifferences 9 YOU APPROVE Sensory scientist approves the report 10 RESULT Sensory analysis report
Read the steps as a list
  1. Panel data exported
  2. Load scores and check data completeness
  3. Calculate panelist agreement and replicate consistency
  4. Is every panelist above the performance thresholds?If not: flag the panelists and note patterns such as scale reversal or no variation. Back to step 3.
  5. Run the planned tests on all data
  6. Rerun the tests without the flagged panelists
  7. Are the conclusions the same with and without the flagged panelists?If not: report both results and mark the conclusion as dependent. Back to step 5.
  8. Write the results with test details and product differences
  9. Sensory scientist approves the reportThe agent waits here for your OK.
  10. Sensory analysis report

How it decides

A panelist is flagged when agreement or replicate consistency falls below the thresholds. The conclusion is robust if it holds with and without flagged panelists.

  • Panelist agreement below the threshold: flag
  • Panelist gave the same score to every sample: flag no discrimination
  • Conclusion changes without flagged panelists: report as dependent
  • Missing scores over 10 percent: use an incomplete design analysis

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Performance thresholds
  • Tests to run
  • Significance level (default 0.05)
  • Panel size
  • Report template

What keeps you in control

It always asks you first

  • The sensory report
  • Any removal of a panelist from future panels

Hard limits

  • Never remove a panelist from the panel list
  • Never hide the all-data result

It stops when

  • Done: report approved
  • Stop: too few panelists after exclusion for the test

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensIn a session with 12 panelists and 5 juices, panelist 7 had a negative correlation with the panel and panelist 4 gave 6 for every sample. The check failed for both. The agent compared analyses. With all data, juice C was higher in sweetness. Without those two, juice C and D were tied. It reported both. The scientist approved.

More agents for statisticians

Quality control · Product development

Product Experiment Sample Integrity Agent

Experiment samples that are valid before anyone draws conclusions.

8 steps2 checks3 jobs
Quality control · Science and research

Research Figure Reproduction Agent

A reproducible figure bundle whose values and labels match the analysis.

7 steps2 checks3 jobs
Quality control · Decisions

Research Analysis Assumption Agent

Every analysis has its assumptions checked, with violations decided by the investigator.

8 steps1 check11 jobs
Planning · Decisions

Landing Page A/B Test Planning Agent

A single-change test plan that is launched correctly and read at the right time

11 steps2 checks6 jobs