Population: The entire group of individuals or items that we want to draw conclusions about
Sample: A subset of the population that is actually observed and measured
Parameter: A numerical characteristic of the population (usually unknown)
Statistic: A numerical characteristic calculated from sample data
Statistical Inference: Using sample data to make conclusions about the population
- Identify the group of interest (population)
- Determine who was actually measured (sample)
- Distinguish between the two based on scope of study
- Understand the relationship between population and sample
The researcher wants to know about all 7th grade students in the city
This is the entire group of interest
The researcher measured heights of students in 5 randomly selected schools
This is the actual group studied
Population: All 7th grade students in the city
Sample: Students from the 5 selected schools
Sample: Students from 5 randomly selected schools
The population is all 7th grade students in the city, and the sample is the students from the 5 randomly selected schools. The researcher will use sample statistics to estimate population parameters.
• Scope Difference: Population is complete, sample is partial
• Measurement: Only sample is directly measured
• Goal: Learn about population through sample study
• Representation: Sample should represent population characteristics
Sample Proportion: The fraction of successes in the sample
Population Proportion: The fraction of successes in the entire population
Point Estimate: A single value used to estimate a population parameter
Statistical Inference: Using sample data to make educated guesses about population parameters
Sampling Error: The difference between sample statistic and population parameter
- Identify population and sample sizes
- Calculate sample proportion
- Use sample proportion to estimate population parameter
- Apply proportion to population size for estimate
Population size: 1,500 employees
Sample size: 150 employees
Satisfied in sample: 90 employees
Sample proportion = 90/150 = 0.6 or 60%
Population parameter being estimated: True proportion of satisfied employees
Point estimate: 60% of the population is satisfied
Estimated satisfied employees = 0.6 × 1,500 = 900 employees
Estimated satisfied employees: 900
Population parameter: True satisfaction rate
The sample proportion is 60% (90/150). Based on this sample, approximately 900 out of 1,500 employees (60%) are satisfied. The population parameter being estimated is the true proportion of satisfied employees in the company.
• Proportional Reasoning: Sample proportions estimate population proportions
• Point Estimation: Sample statistic serves as estimate for population parameter
• Scaling: Apply sample proportion to population size for count estimates
• Random Sampling: Ensures valid statistical inference
Representative Sample: A sample that accurately reflects the characteristics of the population
Bias: Systematic error that favors certain outcomes over others
Sampling Bias: When some members of the population are more likely to be included in the sample
Selection Bias: Bias introduced by the method of selecting the sample
Generalizability: The extent to which sample results can be applied to the population
- Identify the target population
- Examine the sampling method used
- Determine if the sample is representative
- Identify potential biases
- Suggest improvements
The target population is the general population regarding healthy eating habits
Sample was taken from people at a gym, who may have different health priorities
Gym-goers are likely more health-conscious than the general population
This creates selection bias
Results are likely to overestimate healthy eating habits in the general population
Better method: Random sampling from entire population
No, this would not be a good estimate for the general population. People at gyms are more health-conscious and likely to eat healthily, creating selection bias. A better method would be simple random sampling from the entire population to ensure representativeness.
• Representativeness: Sample should mirror population characteristics
• Bias Recognition: Identify factors that make sample unrepresentative
• Generalizability: Results only apply if sample is representative
• Random Selection: Reduces selection bias and improves representativeness
Population: The complete set of individuals, items, or data that researchers want to study and draw conclusions about. It represents the entire group of interest.
Sample: A subset of the population that is actually studied. Researchers collect data from the sample to make inferences about the population.
Population Parameter: A numerical characteristic of the entire population, such as population mean (μ) or population proportion (P). These are usually unknown and are what we want to estimate.
Sample Statistic: A numerical characteristic calculated from sample data, such as sample mean (x̄) or sample proportion (p̂). These are used to estimate population parameters.
Statistical Inference: The process of using sample data to make conclusions about population parameters.
• Size: Population is always larger than or equal to the sample
• Measurement: Only the sample is directly measured or observed
• Goal: Use sample data to learn about population characteristics
• Representativeness: Sample should reflect population characteristics
• Random Selection: Reduces bias and improves representativeness
- Define Population: Clearly identify the group of interest
- Design Sampling Method: Choose appropriate random sampling technique
- Collect Sample Data: Gather information from selected individuals
- Calculate Statistics: Compute sample statistics (mean, proportion, etc.)
- Make Inferences: Use sample statistics to estimate population parameters
- Assess Reliability: Consider sample size and potential bias
Sample: Subset of group
Parameters: Population measures
Statistics: Sample measures
Statistics → Approximate → Parameters
Inference: Sample to Population
Bias: Reduces Validity
• Population is the entire group of interest, sample is the subset studied
• Only sample data is directly collected and analyzed
• Sample statistics are used to estimate population parameters
• Random sampling reduces bias and improves representativeness
• Inferences are only valid if the sample is representative of the population
Questions & Answers
Question: Why can't we just study the entire population instead of using samples?
Answer: Studying the entire population (census) is often impractical because:
- Cost: Too expensive for large populations
- Time: Takes too long to collect data from everyone
- Resources: Requires extensive personnel and materials
- Accessibility: Some populations are impossible to reach completely
- Efficiency: Well-designed samples can provide accurate estimates with less effort
Example: Surveying all 330 million Americans about their health habits would be impossible, but a representative sample of 1,000 can provide reliable estimates.
Question: How do we know if a sample is representative of the population?
Answer: A representative sample should:
- Reflect Diversity: Include all subgroups in proportion to their presence in the population
- Use Random Selection: Give every population member equal chance of inclusion
- Avoid Bias: Not systematically over- or under-represent any group
- Match Characteristics: Have similar demographic and relevant features as population
Example: For a school survey, a representative sample would include students from all grades, backgrounds, and demographics in proportion to their presence in the school.
Question: What's the difference between a parameter and a statistic?
Answer: The key differences are:
- Source: Parameters come from populations, statistics come from samples
- Values: Parameters are fixed (but usually unknown), statistics vary between samples
- Notation: Parameters use Greek letters (μ, σ, P), statistics use Roman letters (x̄, s, p̂)
- Purpose: Statistics estimate parameters
Example: The true average height of all 7th graders in a district (μ) is a parameter; the average height of 50 randomly selected 7th graders (x̄) is a statistic that estimates μ.