Graphs are visual representations of data used to identify patterns, trends, and relationships in statistical datasets efficiently.
## Core concept
A graph is a pictorial method of presenting statistical data. Unlike tables (which show exact values), graphs emphasize visual comparison and highlight patterns at a glance. In CA Foundation, you study graphs as tools for organizing univariate data (single variable) and bivariate data (two variables).
Key purposes: - Identify trends and seasonal variations - Enable quick comparisons between datasets - Make data interpretation accessible to non-technical audiences - Detect outliers and anomalies
## Main types of graphs (within CA Foundation scope)
### 1. Histograms - Used for continuous, grouped data (e.g., height, weight, time) - X-axis: class intervals (no gaps between bars) - Y-axis: frequency or frequency density - All bars are adjacent—no space between them - Class width need not be equal; use frequency density if unequal
### 2. Frequency polygons - Line graph connecting midpoints of class intervals - Can overlay multiple distributions for comparison - Cumulative frequency polygon = Ogive (shows cumulative frequencies)
### 3. Bar charts - Used for discrete, categorical data (e.g., regions, products, job categories) - X-axis: categories (bars separated by gaps) - Y-axis: frequency or count - Bars can be vertical (column chart) or horizontal
### 4. Pie charts - Circle divided into segments proportional to frequencies - Angle of each segment = (Frequency ÷ Total) × 360° - Useful for showing composition/parts of a whole
### 5. Scatter diagrams (Scatter plots) - Used for bivariate data - Plot points (x, y) without connecting lines - Shows correlation between two variables - Visual assessment of positive, negative, or zero correlation
### 6. Time series graphs - Line graph showing data over time - X-axis: time periods; Y-axis: values - Reveals trends, seasonality, and cyclical patterns
## Formula / rule
For histograms with unequal class widths:
Frequency Density = Frequency ÷ Class Width
Height of bar in histogram = Frequency Density
For pie chart segments:
Angle = (Frequency ÷ Total Frequency) × 360°
## Common exam applications
- Drawing a histogram: Given frequency distribution table with class intervals, construct histogram. Remember: no gaps between bars for continuous data.
2. Frequency polygon: Plot class midpoint (x) against frequency (y). For cumulative frequency polygon (ogive), use upper class boundaries.
3. Reading graphs: Extract information—e.g., "How many observations fall between 20–30?" or "What is the mode class from the histogram?"
4. Pie chart construction: Convert raw data into angles and draw segments.
5. Bivariate relationships: Analyze scatter plots for strength and direction of correlation (positive/negative/zero).
## Common mistakes
- Confusing bar charts with histograms: Histograms are for continuous grouped data (no gaps); bar charts for categorical data (gaps present).
- Unequal class widths: Plotting frequency directly instead of frequency density on y-axis; always use frequency density for unequal intervals.
- Incorrect class boundaries: Using midpoints instead of upper/lower boundaries when drawing ogives.
- Misinterpreting scales: Axes must be clearly labeled with units and start from zero (or note a break).
- Overgeneralization from scatter plots: Correlation visible in plots doesn't prove causation.
## Worked example
Construct a frequency polygon from this data:
| Marks | 0–10 | 10–20 | 20–30 | 30–40 | 40–50 | |-------|------|-------|-------|-------|-------| | Frequency | 5 | 12 | 20 | 10 | 3 |
Class midpoints: 5, 15, 25, 35, 45
Plot points: (5, 5), (15, 12), (25, 20), (35, 10), (45, 3)
Connect these points with straight lines. The highest peak at (25, 20) shows the modal class is 20–30.