Research MethodsVolume IMeasurement, Flight Data, and Evidence
Volume I of two · by Andres Peña
Avionics Research Methods
Volume I: Measurement, Flight Data, and Evidence
From a physical quantity to a defended technical claim.
- Edition
- First edition, 2026
- Length
- 430+ pages
- Format
- Price
- $35
Order by request. Nothing is charged until you confirm.
Who it's for
Written for three readers.
As the preface names them
Advanced secondary students
With a firm grasp of algebra.
Undergraduates
In aerospace, electrical or systems engineering.
Early-career engineers
Who work with flight data and were taught the statistics without being taught what the statistics license.
The only prerequisite
Algebra.
- Rearrange an equation.
- Read a graph with labelled axes.
- Carry a unit through an arithmetic operation.
That is the whole prerequisite.
Built from nothing
- Probability
- Statistics
- Estimation
- Control
- Signal processing
- Flight-test engineering
Python appears from Part VII onward, as a calculator that keeps a record of what it did, not as a subject.
What it will not do
- Put a solution beside its exercise.
- Describe a synthetic record as flight data.
- Present an invented incident as having happened.
- Report a statistic without its sample count, a value without its unit, or a chapter without its limitations.
Eight claims the book sets out to demonstrate
None of them is original. What the book supplies is the demonstration.
- 01A number without a unit is not a measurement.Chapters 1 and 6
- 02A statistic without a sample count is incomplete.Chapters 17 and 19
- 03Data is not automatically evidence.Chapter 2
- 04Deviation is not automatically cause.Chapter 22
- 05A claim must survive counter-evidence.Chapters 40, 42 and 44
- 06Not knowing is an acceptable scientific state.Chapter 67
- 07A record has limits.Chapters 5, 41 and 65
- 08A reproducible result is stronger than an untraceable one.Chapters 53, 55 and 62
Inside Volume I
Chapter numbers run across both volumes; Part IV is printed in Volume II.
- Part I0–5Foundations
- What Is a Measurement?
- From Data to Evidence
- Four Failure Signatures: Spike, Freeze, Gap, Drift
- Part II6–10The Measurement Chain
- Sensor Output and Calibration
- Quantization, Resolution, and Information Loss
- Sampling, Time, Latency, and Measurement Age
- Part III16–23Statistics for Evidence
- Why n−1 Appears
- Thresholds and False Decisions
- Why Thresholds Detect Deviation, Not Cause
- Part V36–45Investigation Method
- How to Investigate Without Knowing the Answer
- Counter-Evidence
- What Would Change Your Mind?
- Part VI46–52Experiment Design
- Independent Variables and Controls
- Confounding Variables
- Designing an Avionics Experiment
- Part VII53–60Reproducibility
- Dataset Identity
- Hashes and Data Integrity
- The Reproducibility Capsule
- Part VIII61–67Defending a Result
- From Result to Claim
- Stating Limitations
- When the Correct Answer Is “The Evidence Is Insufficient”
- Part IXA–CCases
- Single-Sensor Bias
- Slow Drift, and What Late Detection Costs
- A Dropout That Left No Gap
- Part XBOOK-CAP-V1Capstone
- A Blind Investigation
The datasets, worked examples and case studies are synthetic and labelled ILLUSTRATIVE throughout. Every quantitative claim names its dataset and window.
An independent technical publication, not issued, endorsed or approved by any aviation authority, standards body or academic institution.