SA101: مبادئ الإحصاء 1
يوليو 2026

The Measurement Problem

Statistics is not about numbers. It is about what numbers mean — and more importantly, what they don't.

Before you calculate a mean, run a regression, or test a hypothesis, you have already made a decision that determines whether your answer is meaningful or nonsense: you decided what kind of thing you were measuring. A student's nationality and a student's exam score are both numbers in a spreadsheet. One of them you can average. The other one you cannot. The difference between the two is the entire foundation of statistical analysis.

Interactive 01The Moving Statistic

The true parameter (like the average height of all humans) is fixed but unknown. The statistic (the average height of your sample) is known but constantly moving. Every time you draw a new sample, the statistic changes.

Population (N=200)
True Parameter (μ\mu)
0100
Sample Statistics (xˉ\bar{x})
0100

Notice how the sample means build their own distribution around the true population parameter.

Interactive 02What Are You Allowed To Say?

The branch of statistics you are using depends entirely on the claim you are trying to make. Are you describing what you already have, or are you trying to infer what you don't know?

Population Space (N = 200)
Total Population
N = 200
True Population
Mean (μ\mu)
34.1
Pop. Std Dev
(σ\sigma)
19.6
Parameter Estimation Axis
0
50
100
μ=34.1\mu = 34.1

Descriptive statements report facts about the sample. Inferential statements make leaps of faith about the population.

Interactive 03The Arithmetic Permission Machine

Not all numbers are created equal. The level of measurement dictates what mathematical operations are legally allowed to be performed on the data.

Theorem Validation Workspace
Theoretical Context
Kilograms possess a "True Zero" (the literal absence of mass). This structural absolute allows us to safely make proportional Ratio statements, such as concluding that 100kg is exactly twice as heavy as 50kg.
100 kg
Drop
50 kg
Available Operators
Identity
>
Magnitude
-
Interval
÷
Ratio
Maximum Confirmed ScalePending Validation

Notice how higher levels of measurement inherit all the permissions of the levels below them.

Interactive 04The Object Scanner

Quantitative data must be further split based on how it is gathered. If it's counted, it's discrete. If it's measured, it's continuous. Use the scanner to see how different lenses extract different data from the same object, then test your intuition.

Variables

A characteristic of a population or sample.

Used Vehicle
Awaiting scan directive...

Exercise: Scenario Classification

"He is 19 years old. His mother is a teacher. He has 3 brothers. He got a 'B' in the course."

Every piece of information in that statement represents a specific type of variable. Drag and drop the extracted concepts into their correct classifications.

19 years oldAge
TeacherMother's Occupation
3 brothersNumber of Siblings
'B' GradeCourse Grade
Categorical
Quantitative Discrete
Quantitative Continuous

Be careful with money and time — they can be tricky to classify depending on the context.

Act Progress1 / 7
Unit 1: The Measurement Problem
NEXT Unit 2: The Methodology

"How to actually do statistics, step by step."