Understanding which type of analysis to apply is essential for proper interpretation:

2.1 Descriptive Analysis

This answers "What happened?" by summarising past data using statistics such as mean, median, mode, and standard deviation. For example, calculating the average monthly sales of a shop in Dodoma helps understand past performance.

2.2 Diagnostic Analysis

This answers "Why did this happen?" by investigating the causes behind observed patterns. For instance, if sales dropped suddenly, diagnostic analysis might reveal that prices increased or a competitor opened nearby.

2.3 Predictive Analysis

This answers "What is likely to happen?" by using historical data and statistical models to forecast future outcomes. For example, predicting next month's ice cream sales based on past seasonal trends.

2.4 Prescriptive Analysis

This answers "What should we do?" by recommending optimal actions based on predictive outcomes. For example, a prescriptive system might recommend ordering more bread and milk while reducing sugar stock based on sales forecasts.

3.1 Bar Charts

Bar charts compare categorical data across different groups. They are ideal for showing quantities side by side.

Example: A bar chart showing sales of three products (Bread: 500 units, Milk: 350 units, Sugar: 150 units) makes comparison immediate.

3.2 Line Graphs

Line graphs display data changing over time, making trends easy to track.

3.3 Histograms

Histograms show the distribution of numerical data by grouping values into bins. They reveal the shape of data—whether it is skewed, normal, or uniform.

3.4 Pie Charts

Pie charts show proportions or percentages of a whole. They work well for representing market share or budget allocation.

3.5 Scatter Plots

Scatter plots display the relationship between two numerical variables. Each point represents one observation with x-axis and y-axis values.

Example: Plotting study hours (x-axis) against exam scores (y-axis) reveals whether more study time correlates with higher scores.

From a scatter plot, you can identify:

• Positive relationship: Both variables increase together
• Negative relationship: One increases while the other decreases
• No relationship: No clear pattern between variables

Numerical summaries complement visual analysis:

Worked Example:

Given student scores: 5, 7, 8, 7, 3

Mean calculation: $\text{Mean} = \frac{5 + 7 + 8 + 7 + 3}{5} = \frac{30}{5} = 6$

Median calculation: Ordered data: 3, 5, 7, 7, 8 Median = 7 (the middle value)

Standard Deviation calculation:

Step 1: Find differences from mean (6) and square them:

• (5-6)² = 1
• (7-6)² = 1
• (8-6)² = 4
• (7-6)² = 1
• (3-6)² = 9

Step 2: Calculate variance: $\text{Variance} = \frac{1 + 1 + 4 + 1 + 9}{5} = \frac{16}{5} = 3.2$

Step 3: Standard deviation: $\text{Standard Deviation} = \sqrt{3.2} \approx 1.79$

A low standard deviation indicates scores cluster near the mean (6), while a high standard deviation shows greater variability.

5.1 Excel Pivot Tables

Pivot tables quickly summarise large datasets by grouping and aggregating data.

Example: Sales by Region and Product

Given data:

Creating a Pivot Table:

1. Select the data range
2. Insert → PivotTable
3. Drag Region to Rows, Product to Columns, Sales to Values (automatically sums)
4. Result shows totals by region and product

5.2 Python Libraries

Python provides powerful libraries for visualisation:

• Pandas: Organises and analyses data in DataFrames
• Matplotlib/Seaborn: Creates visualisations
• NumPy: Handles numerical operations

Simple Python Example:

import pandas as pd
import matplotlib.pyplot as plt

# Sample data
data = {"Region": ["Tabora", "Tabora", "Ruvuma", "Ruvuma"],
        "Product": ["Mango", "Orange", "Mango", "Orange"],
        "Sales": [20000, 15000, 30000, 10000]}

df = pd.DataFrame(data)

# Group by region and sum sales
sales_by_region = df.groupby("Region")["Sales"].sum()
print(sales_by_region)

# Create bar chart
sales_by_region.plot(kind="bar")
plt.title("Total Sales by Region")
plt.ylabel("Sales (TZS)")
plt.show()

1. Understand the question: Define what you want to learn from the data
2. Prepare the data: Clean missing values, remove duplicates, and format consistently
3. Choose appropriate visual: Select the chart type matching your data and question
4. Create the visualisation: Use tools like Excel or Python to build the chart
5. Interpret findings: Read the visual to identify patterns, trends, or relationships
6. Draw conclusions: Connect insights to the original question
7. Communicate results: Present findings clearly to your audience

When interpreting scatter plots or statistical relationships, remember:

• Correlation means two variables move together
• Causation means one variable directly causes change in another

Example: Ice cream sales and drowning incidents are correlated (both increase in summer), but neither causes the other. The confounding variable is temperature—hot weather causes people to buy more ice cream AND swim more, increasing drowning risk.

Always consider confounding variables before drawing causal conclusions.

In Tanzania, data visualisation and interpretation skills are valuable in small businesses and daily life. For instance, a market vendor in Kariakoo can use a simple bar chart (drawn on paper or created in Excel) to track which items (like rice, beans, or tomatoes) sell best each week. By interpreting this data, the vendor can decide which products to stock more of during peak seasons, ultimately improving profits and reducing waste. Similarly, a school administrator might use a line graph to monitor student attendance trends over the term and take timely action when patterns indicate declining participation.