Inference from sample statistics and margin of error

Inference from sample statistics allows researchers to estimate characteristics of an entire population without surveying every single individual. The sample statistic (such as a sample mean or sample percentage) serves as an estimate, while the margin of error accounts for the natural variation between samples.

Key Principles of Statistical Inference

• Sample vs. Population:

  • Population: The complete set of individuals or objects being studied.

  • Sample: A smaller subset chosen from the population to collect data.

    Example: Surveying $500$ registered voters (sample) in a city to estimate how all $50,000$ voters (population) will vote.

    
    

• Random Sampling (Generalizability):

  • To make valid conclusions about a population, the sample must be selected randomly.

  • If a sample is biased (e.g., surveying only people at a specific location), the results apply only to that group, not the broader population.

    Example: Asking people at a dog park about their favorite pet creates a biased sample that cannot represent the whole town.

    
    

• Margin of Error and Confidence Intervals:

  • The margin of error defines a range above and below the sample statistic where the true population parameter is likely to fall.

    $$\text{Confidence Interval} = \text{Sample Statistic} \pm \text{Margin of Error}$$

    Example: A poll shows $54\%$ of voters support a candidate with a margin of error of $\pm 3\%$. The true population support is estimated to be between $51\%$ ($54 - 3$) and $57\%$ ($54 + 3$).

    
    

Factors Affecting Margin of Error

The size of the margin of error depends directly on how the sample is collected and how large it is.

• Sample Size Effect:

  • Larger sample size $\implies$ Smaller margin of error (more precise estimate).

  • Smaller sample size $\implies$ Larger margin of error (less precise estimate).

    Example: A sample of $1,000$ voters will have a smaller margin of error than a sample of $100$ voters from the same population.

• Variability in Data:

  • Datasets with more spread/variability produce a larger margin of error than datasets where responses are very similar.

1.  Identify the exact population represented by the sample selection method.

   Example: A researcher surveys $200$ randomly selected students from the school soccer team about their daily water intake.

   • Correct conclusion: The findings can be generalized to all students on the school soccer team.

   • Incorrect conclusion: The findings cannot be generalized to the entire student body, because the sample was drawn exclusively from soccer players.

2.  Calculate the upper and lower bounds of a confidence interval.

   Example: A study estimates that the average weight of a adult dog in a shelter is $45\text{ lbs}$ with a margin of error of $4\text{ lbs}$.

   Bound	Formula	Calculation	Result
   Lower Bound	$\text{Sample Mean} - \text{Margin of Error}$	$45 - 4$	$41\text{ lbs}$
   Upper Bound	$\text{Sample Mean} + \text{Margin of Error}$	$45 + 4$	$49\text{ lbs}$

3. Determine how changing sample size alters the margin of error.

   Example: A pollster wants to decrease the margin of error of an election survey.

   Action	Impact on Margin of Error	Precision
   Increase sample size (e.g., $500 \to 2,000$)	Decreases	Higher precision
   Decrease sample size (e.g., $500 \to 100$)	Increases	Lower precision