SA101: مبادئ الإحصاء 1
يوليو 2026

The Shape of Information

A list of forty wages tells you almost nothing. You cannot see a pattern in a column of numbers. You cannot feel where the data is dense, where it thins out, where it stops entirely. The techniques in this unit exist for one reason: to give data a shape you can actually read. Every choice you make — how many classes, how wide, which chart type — is a decision that changes what the data appears to say. That is not a flaw in the method. It is the method.

Interactive 09The Class Width Problem

When organizing continuous data into a frequency distribution, the most impactful decision you make is the class width. Make the classes too wide, and you lose all the detail—everything clumps together into a featureless block. Make them too narrow, and you lose the shape entirely, getting a jagged mess of individual data points.

Start Value10
Max Value98
Required Span88
Width = ceil(Span / k)15
180
10
10
18
25
6
40
3
55
2
70
1
85
100
Daily Wages (£)
Class Width Calculation
WMaxStartk=98106=14.67    15 W \approx \frac{\text{Max} - \text{Start}}{k} = \frac{98 - 10}{6} = 14.67 \implies 15

Notice how changing the number of intervals alters the "story" the distribution tells.

Interactive 10Histogram vs. Bar Chart

They look similar, but they represent fundamentally different types of data. A bar chart displays categorical data (like eye color or car brands), where the gaps between bars emphasize that the categories are separate. A histogram displays continuous, quantitative data, where the bars touch because the number line is unbroken.

Gaps matter. If the data is continuous, the bars must touch to represent the flow of the number line.

Interactive 11The Cumulative Ogive
While a histogram shows how many items fall within a specific interval, an Ogive (cumulative frequency polygon) shows how many items fall below a certain value. It answers questions like "How many students scored below 80?" instead of "How many students scored between 70 and 80?"
010203040450102030405060708090100110Cumulative Frequency (Workers)Daily Wages (£)

Preset Questions

Hover over the chart to read specific cumulative values.

Ogive Calculation
F(x)=P(Xx)=Cumulative FreqN F(x) = P(X \le x) = \frac{\text{Cumulative Freq}}{N}

The Ogive never drops down—it only accumulates until it reaches 100% of the population.

Interactive 12Pie Chart vs. Bar Chart

Pie charts are ubiquitous, but human eyes are notoriously bad at comparing angles and areas. A bar chart is almost always a superior choice because we are much better at comparing straight-line lengths.

Pie Chart

Carrefour(36%)
Hyper(33%)
Spinneys(16%)
Metro(15%)

Bar Chart

52,921
47,348
22,939
21,886
Carrefour
Hyper
Spinneys
Metro
Calculations
θ=360×ff=360×52921145094=131.3 \theta = 360^\circ \times \frac{f}{\sum f} = 360^\circ \times \frac{52921}{145094} = 131.3^\circ

See how difficult it is to rank similar-sized slices in a pie chart compared to the exact same data presented as bars.

Interactive 13Comparing Groups

Visualizations become truly powerful when you use them to compare distributions side-by-side. By plotting two different groups on the same scale, you can instantly see differences in central tendency, spread, and shape that summary statistics might hide.

Group A

n = 200
1200
10
30-40
30
40-50
50
50-60
60
60-70
40
70-80
10
80-90
Student Grades

Group B

n = 400
1200
40
30-40
50
40-50
50
50-60
60
60-70
80
70-80
120
80-90
Student Grades

Observation:Group B appears much larger across almost all grades. It's difficult to see which group performed better on average because Group B's sheer size dwarfs Group A.

Overlapping histograms or dot plots reveal exactly where the two populations diverge and intersect.

Exercise: Reading Data

It's time to put your visual literacy to the test. Given a specific question and data type, select the most appropriate visualization method to tell the correct story without misleading your audience.

Step 1: A professor collected the final grades of 10 students. Fill in the Cumulative Frequency column to build the Ogive.

ClassFreqCum. Freq
0 - less than 32
3 - less than 63
6 - less than 92
9 - less than 122
12 - less than 151
024681003691215
Complete table to unlock

Choose carefully: the wrong visualization can turn a clear insight into a confusing mess.

Act Progress3 / 7
Unit 3: The Shape of Information
NEXT Unit 4: Locating the Center

"Every dataset has a center. The problem is that 'center' means three different things depending on what you ask."