Quantifying measurement : the tyranny of numbers

Print version:

Introduction

1. The tyranny of numbers

1.1. Why we measure things

1.2. A little history

1.3. Surveying

1.4. Other surveys. 2. The error in all things

2.1. Introduction

2.2. Méchain's 'error' in greater detail and least-squares

2.3. The metric survey

2.4. Least-squares

2.5. Statistical methods. 3. A language for measurement

3.1. Introduction

3.2. The quality of measurements

3.3. Measurement errors. 4. What is it that we measure, and what does it tell us?

4.1. A classic laboratory experiment

4.2. Precision measurements made infrequently

4.3. An overabundance of uncertain data

4.4. What makes the world go around? 5. Measurement uncertainty

5.1. Uncertainty

5.2. Uncertainty in measurements

5.3. Type A and Type B uncertainty

5.4. Propagation of uncertainty

5.5. Uncertainty evaluation

5.6. Probability

5.7. Expected value. 6. Guide to the expression of uncertainty in measurement (the GUM)

6.1. Introduction

6.2. Basic definitions

6.3. Evaluating uncertainty components

6.4. Uncertainty derived from some assumed distribution

6.5. Combining uncertainty components

6.6. Expanded uncertainty and coverage factor. 7. Clinical trials

7.1. Introduction

7.2. Sample size

7.3. Statistical hypothesis testing. 8. Direct measurements : quadrupole moments and stray light levels

8.1. Introduction

8.2. Measuring the quadrupole moments of molecules

8.3. Experimental details

8.4. How many measurements do you need? 9. Indirect measurement : the optical Kerr effect

9.1. Introduction

9.2. The optical Kerr effect. 10. Data fitting and elephants

10.1. Introduction

10.2. Regression analysis

10.3. Over-fitting data

10.4. Avoiding over-fitting