Sign in to save

Bookmark this page so you can find it later.

Sign in to save

Bookmark this page so you can find it later.

Sampling Distribution & Standard Error cheat sheet - grade 10-12

Click image to open full size

Statistics Grade 10-12

Sampling Distribution & Standard Error Cheat Sheet

A printable reference covering sampling distributions, standard error, Central Limit Theorem, confidence intervals, and standard error formulas for grades 10-12.

TopicsStatisticsData AnalysisProbabilityAlgebraStudy Guide
Download PNG

Related Tools

Related Labs

Related Worksheets

Related Infographics

Study as Flashcards

This cheat sheet explains how sample statistics vary from sample to sample and how that variation is measured. Students need these ideas to understand why different samples from the same population can give different results. Sampling distributions and standard error are the foundation for confidence intervals, hypothesis tests, and many real data investigations.

The sheet helps students connect formulas to the meaning of uncertainty in statistics.

The core idea is that a statistic, such as a sample mean or sample proportion, has its own distribution across many possible samples. The standard error measures the typical distance between a sample statistic and the true population parameter. For means, the standard error is SExˉ=σnSE_{\bar{x}} = \frac{\sigma}{\sqrt{n}} when the population standard deviation is known.

For proportions, the standard error is SEp^=p(1−p)nSE_{\hat{p}} = \sqrt{\frac{p(1-p)}{n}} when the true population proportion is known.

Key Facts

  • A sampling distribution is the distribution of a statistic, such as xˉ\bar{x} or p^\hat{p}, from many random samples of the same size.
  • The standard error of the sample mean is SExˉ=σnSE_{\bar{x}} = \frac{\sigma}{\sqrt{n}} when the population standard deviation σ\sigma is known.
  • If σ\sigma is unknown, the estimated standard error of the sample mean is SExˉ≈snSE_{\bar{x}} \approx \frac{s}{\sqrt{n}}.
  • The standard error of a sample proportion is SEp^=p(1−p)nSE_{\hat{p}} = \sqrt{\frac{p(1-p)}{n}} when the population proportion pp is known.
  • For confidence intervals using sample data, the estimated standard error for a proportion is SEp^≈p^(1−p^)nSE_{\hat{p}} \approx \sqrt{\frac{\hat{p}(1-\hat{p})}{n}}.
  • The Central Limit Theorem says that the sampling distribution of xˉ\bar{x} becomes approximately normal as nn increases, especially when n≥30n \ge 30.
  • Increasing the sample size nn decreases standard error because SESE is divided by n\sqrt{n}.
  • A common confidence interval form is statistic±critical value×SE\text{statistic} \pm \text{critical value} \times SE.

Vocabulary

Sampling Distribution
The distribution of a statistic calculated from many random samples of the same size from one population.
Statistic
A number calculated from sample data, such as xˉ\bar{x} or p^\hat{p}, used to estimate a population parameter.
Parameter
A fixed but often unknown number that describes a population, such as μ\mu, σ\sigma, or pp.
Standard Error
The standard deviation of a sampling distribution, measuring the typical sampling variation of a statistic.
Central Limit Theorem
A theorem stating that many sampling distributions become approximately normal for large enough sample sizes.
Sample Size
The number of observations in a sample, represented by nn.

Common Mistakes to Avoid

  • Confusing standard deviation with standard error is wrong because standard deviation measures spread in individual data values, while standard error measures spread in sample statistics.
  • Using SExˉ=σnSE_{\bar{x}} = \frac{\sigma}{n} is wrong because the sample size affects standard error through n\sqrt{n}, so the correct formula is SExˉ=σnSE_{\bar{x}} = \frac{\sigma}{\sqrt{n}}.
  • Assuming every sampling distribution is normal is wrong because normality depends on the population shape, sample size, and conditions such as independence.
  • Forgetting to check random sampling or independence is wrong because standard error formulas rely on samples that are reasonably random and observations that are not strongly dependent.
  • Using pp when only p^\hat{p} is available can be wrong in real sample work because the true population proportion is usually unknown, so use SEp^≈p^(1−p^)nSE_{\hat{p}} \approx \sqrt{\frac{\hat{p}(1-\hat{p})}{n}} for estimation.

Practice Questions

  1. 1 A population has standard deviation σ=12\sigma = 12, and samples of size n=36n = 36 are taken. Find SExˉSE_{\bar{x}}.
  2. 2 A survey finds p^=0.40\hat{p} = 0.40 from n=100n = 100 students. Estimate the standard error using SEp^≈p^(1−p^)nSE_{\hat{p}} \approx \sqrt{\frac{\hat{p}(1-\hat{p})}{n}}.
  3. 3 If the sample size increases from n=25n = 25 to n=100n = 100 while σ\sigma stays the same, how does SExˉSE_{\bar{x}} change?
  4. 4 Explain why a larger sample size usually gives a more reliable estimate, even though it does not guarantee the sample statistic equals the population parameter.

Understanding Sampling Distribution & Standard Error

Imagine a school has every student write down their daily screen time. The full set of answers is the population. If you repeatedly draw random groups of fifty students and calculate the average for each group, the averages form a new pattern.

Individual screen times may range widely, yet group averages tend to stay much closer together. Extreme values partly balance out when observations are combined. This is why an average from a larger random group is usually more stable than an average from a small group.

A sampling distribution is often studied through a simulation because collecting thousands of real samples is rarely practical. Software can repeatedly sample from known data and show how the resulting statistics behave.

The Central Limit Theorem helps explain why a bell shaped model appears so often in statistics. It concerns the averages, not necessarily the original measurements. For example, household incomes can be strongly right skewed because a small number of households earn very high incomes.

Averages from many sufficiently large random samples can still be close to normally distributed. The rule of thirty is a useful classroom guide, not a guarantee. Very skewed data or data with major outliers may need much larger samples.

Independence matters too. A sample should be a small part of the population, commonly less than ten percent when sampling without replacement. This condition prevents one selected person from strongly changing the chance that another person is selected.

Standard error is used to build a margin of error. The margin tells how far a sample result may reasonably be from the population value under a chosen confidence level. A ninety five percent confidence interval uses a critical value that is roughly two standard errors on each side for many common situations.

The confidence level describes the long run method. If researchers repeated the same sampling process many times, about ninety five percent of intervals made by that method would contain the true parameter.

It does not mean there is a ninety five percent chance that a fixed population value is inside one completed interval. The population value is already fixed, while the interval changes from sample to sample.

Precision is not the same as accuracy. A large sample can have a small standard error but still give a misleading result if the sampling method is biased. An online poll about school lunch may miss students who do not see the link.

Asking only students in one lunch period can miss patterns in other grades. Random selection reduces this problem, while a bigger sample mainly reduces random variation. For proportions, check that the expected numbers of successes and failures are large enough before using a normal method.

Students should notice the square root rule for sample size. To cut standard error in half, the sample size must be multiplied by four. Collecting more data improves precision, but the cost and time can rise quickly.