Solved Exercises on Sampling Methods in Grade 10 Statistics

Master sampling methods: simple random, stratified, and cluster sampling through these 5 detailed exercises with comprehensive explanations.

Solution: Exercises 1 to 3
1 Simple Random Sampling
Exercise 1
A school has 500 students. Describe how to select a sample of 50 students using simple random sampling. What are the advantages and disadvantages of this method?
Definition:

Simple Random Sampling: Every member of the population has an equal chance of being selected, and every possible sample of size n has an equal chance of being selected.

Simple Random Sampling Method:
  1. Assign a unique number to each member of the population (1 to 500)
  2. Use a random number generator or table to select 50 unique numbers
  3. Select the corresponding members for the sample
Step 1: Assign numbers to population

Number all 500 students from 1 to 500

Step 2: Generate random numbers

Use a random number generator to select 50 unique numbers between 1 and 500

Step 3: Select sample

Select the 50 students corresponding to the randomly generated numbers

Step 4: Verify randomness

Ensure each student had an equal probability of selection (50/500 = 10%)

Solution:

Assign numbers 1-500 to students, generate 50 random numbers, select corresponding students. Each student has 10% chance of selection.

Advantages:

• Unbiased: Each member has equal chance of selection

• Simple: Easy to implement with small populations

• Statistically valid: Allows for probability calculations

Disadvantages:

• May lack representativeness: Could miss important subgroups

• Requires complete list: Need access to full population

• Logistical challenges: Difficult for large, dispersed populations

Tip: Use computer software for truly random selection.
Key Note: Sample size affects precision of results.
2 Stratified Sampling
Exercise 2
In a school of 1000 students, 600 are female and 400 are male. Design a stratified sample of 100 students that maintains the gender proportion. Explain why this method might be better than simple random sampling.
Definition:

Stratified Sampling: Population is divided into non-overlapping groups (strata), then samples are taken from each stratum proportional to its size in the population.

Step 1: Identify strata and calculate proportions

Female: 600/1000 = 60%, Male: 400/1000 = 40%

Step 2: Calculate sample size for each stratum

Female: 60% × 100 = 60 students, Male: 40% × 100 = 40 students

Step 3: Apply simple random sampling within each stratum

Select 60 females randomly from 600, and 40 males randomly from 400

Step 4: Combine samples

Total sample: 60 females + 40 males = 100 students

60 females (from 600) + 40 males (from 400) = 100 total students
Final answer:

Select 60 females using simple random sampling from 600, and 40 males using simple random sampling from 400.

Formula:

Sample size for stratum = (Stratum size / Total population) × Total sample size

Why stratified is better:

• Guaranteed representativeness: Ensures all subgroups are represented

• More precise estimates: Reduces sampling error

• Better comparisons: Allows analysis within subgroups

3 Cluster Sampling
Exercise 3
A researcher wants to survey 500 high school students across 50 schools in a district. Explain how to use cluster sampling, and compare it to simple random and stratified sampling.
Definition:

Cluster Sampling: Population is divided into clusters (usually geographic or organizational units), some clusters are randomly selected, and all members of selected clusters are included in the sample.

Step 1: Define clusters

Each school represents one cluster (total of 50 schools)

Step 2: Randomly select clusters

Randomly select a number of schools (e.g., 10 schools) to ensure approximately 500 students

Step 3: Include all members of selected clusters

Survey all students in the selected 10 schools

Step 4: Calculate total sample size

If each school averages 50 students, 10 schools = 500 students

Select 10 schools randomly, survey all students in those schools = 500 total
Final answer:

Randomly select 10 schools from 50, then survey all students in those schools (approximately 500 total).

Comparison Table:
Method Process Advantages Disadvantages
Simple Random Individual selection Unbiased, fair Expensive, may miss subgroups
Stratified Proportional subgroup sampling Representative, precise Requires detailed population info
Cluster Group selection Cheap, convenient Less precise, potential bias
Sampling Methods Overview
\(\text{Stratified Sample Size} = \frac{\text{Stratum Size}}{\text{Total Population}} \times \text{Total Sample Size}\)
Stratified Sampling Formula
🎯 Simple Random Sampling
Every individual has equal chance of selection. Like drawing names from a hat.
Assign numbers
Generate random numbers
Select individuals
📊 Stratified Sampling
Divide population into groups, then sample proportionally from each group.
Identify strata
Calculate proportions
Sample within strata
🏢 Cluster Sampling
Divide population into clusters, randomly select clusters, sample all in selected clusters.
Define clusters
Randomly select clusters
Sample all in clusters
When to use Simple Random: Small, homogeneous populations where complete list is available.
When to use Stratified: When important subgroups exist that need representation in the sample.
When to use Cluster: Large, geographically dispersed populations where cost efficiency is important.
Solution: Exercises 4 to 5
4 Multi-stage Sampling
Exercise 4
A city has 20 districts with 100 schools each. The city wants to survey 500 teachers. Design a multi-stage sampling plan combining cluster and stratified sampling.
Definition:

Multi-stage Sampling: A complex sampling method that combines multiple sampling techniques in stages, often starting with cluster sampling followed by other methods within selected clusters.

Step 1: Stage 1 - Cluster Sampling

Select 10 districts randomly from 20 districts

Step 2: Stage 2 - Within Selected Districts

Identify teacher categories (e.g., subject areas) as strata within each selected district

Step 3: Stage 3 - Stratified Sampling

Within each selected district, randomly select 50 teachers proportionally from different subject strata

Step 4: Stage 4 - Verification

Total: 10 districts × 50 teachers per district = 500 teachers

Stage 1: 10 districts → Stage 2: Strata within districts → Stage 3: 50 teachers per district → Total: 500
Final answer:

First randomly select 10 districts, then within each district identify subject strata and randomly select 50 teachers proportionally from each stratum.

Advantages of Multi-stage:

• Cost-effective: Reduces travel and administrative costs

• Practical: Suitable for large, complex populations

• Representative: Maintains diversity while reducing complexity

Considerations:

• Increased complexity: More difficult to analyze statistically

• Potential bias: Each stage introduces potential sources of bias

• Sample size calculations: More complex due to multiple stages

5 Comparative Analysis
Exercise 5
A company with 5000 employees across 5 departments wants to survey 250 employees. Compare the effectiveness of simple random, stratified, and cluster sampling for this scenario. Which method would you recommend and why?
Definition:

Comparative Analysis: Evaluating different sampling methods based on criteria like representativeness, cost, feasibility, and statistical validity.

Step 1: Analyze the scenario

Population: 5000 employees across 5 departments, Sample size: 250 (5%)

Step 2: Evaluate Simple Random Sampling

Advantages: Unbiased, simple; Disadvantages: May not represent all departments equally

Step 3: Evaluate Stratified Sampling

Departments as strata: 5% of each department (proportional allocation)

Step 4: Evaluate Cluster Sampling

Select 1-2 departments as clusters (non-proportional representation)

Step 5: Make recommendation

Stratified sampling recommended for departmental representation

Recommend STRATIFIED SAMPLING: 5% from each department ensures representation
Final answer:

Recommend stratified sampling with departments as strata. Select 5% of employees from each department to ensure all departments are represented proportionally.

Method Comparison:
Method Representation Cost Complexity Best For
Simple Random May vary Medium Low Homogeneous populations
Stratified Guaranteed High Medium Heterogeneous populations
Cluster Limited Low Low Large geographic areas
Tip: Choose sampling method based on research goals and population characteristics.
Key Note: Sample size affects precision regardless of sampling method.
Important: Document your sampling method for result interpretation.
Sampling Methods: Laws, Methods, Rules, Definitions
\(\text{Sampling Fraction} = \frac{n}{N}\)
Sampling Fraction Formula
Key definitions:

Population: The entire group of individuals we want to study

Sample: A subset of the population selected for study

Parameter: A numerical characteristic of a population

Statistic: A numerical characteristic of a sample

Complete Sampling Methodology:
  1. Define the population: Clearly specify who/what you want to study
  2. Choose the sampling method: Based on population characteristics and research goals
  3. Determine sample size: Balance between precision and practical constraints
  4. Implement the sampling: Follow the chosen method precisely
  5. Verify representativeness: Check that sample reflects population characteristics
Tip 1: Always document your sampling procedure for transparency.
Tip 2: Larger samples generally provide more precise estimates.
Tip 3: Consider practical constraints when choosing sample size.
Tip 4: Use computer software for random number generation.
Common errors: Convenience sampling, voluntary response bias, undercoverage.
Exam preparation: Know differences between sampling methods, recognize when each is appropriate.
Essential Formulas:

• Sampling fraction: \( \frac{n}{N} \)

• Proportional allocation: \( n_h = N_h \times \frac{n}{N} \)

• Standard error depends on sampling method and sample size

Questions & Answers

Question: I'm confused about when to use stratified vs. cluster sampling. Both involve dividing the population into groups. Can you clarify the key difference?

Answer: Great question! The key difference lies in how sampling occurs within the groups:

  • Stratified sampling: We divide the population into strata (groups), then take a sample FROM EACH STRATUM. The goal is to ensure representation from all important subgroups.
  • Cluster sampling: We divide the population into clusters, then randomly select ENTIRE CLUSTERS and include ALL members of selected clusters. The goal is to reduce costs and increase efficiency.

Analogy: Think of a pizza. For stratified sampling, you'd cut slices from each topping area (representing different strata). For cluster sampling, you'd randomly select whole slices (clusters) and eat everything on those slices.

When to use: Use stratified when you want guaranteed representation from important subgroups. Use cluster when the population is naturally grouped and you want to reduce logistical costs.

Question: In stratified sampling, why do we maintain the same proportion in each stratum? Couldn't we take different proportions?

Answer: You raise an excellent point! There are actually TWO types of stratified sampling:

  • Proportional allocation: Sample size from each stratum is proportional to its size in the population (the standard approach)
  • Disproportional allocation: Different sampling rates for different strata, used when some strata are more important or variable

For example, if you're studying a rare disease, you might oversample the affected population to get enough cases for meaningful analysis, even though they're a small percentage of the total population.

However, for most basic applications (like your grade level), proportional allocation is the standard because it maintains the same representation as the population.

Formula for proportional allocation: \( n_h = N_h \times \frac{n}{N} \), where \( n_h \) is sample size for stratum h, \( N_h \) is population size of stratum h, n is total sample size, and N is total population size.

Question: How does sample size affect the accuracy of our results? Is bigger always better?

Answer: Generally, larger samples provide more accurate estimates, but there are important nuances:

  • Law of diminishing returns: Accuracy increases with the square root of sample size, so doubling sample size only improves accuracy by about 40%
  • Sampling method matters: A well-designed smaller sample can be more accurate than a poorly designed larger sample
  • Cost considerations: At some point, the marginal gain in accuracy doesn't justify the increased cost

Mathematical relationship: The standard error of a sample mean is proportional to \( \frac{1}{\sqrt{n}} \), where n is the sample size.

Practical advice: Focus on proper sampling method and reducing bias rather than just increasing sample size. A representative sample of 1000 is usually much better than a biased sample of 10,000.