Die Druckversion wird nicht mehr unterstützt und kann Darstellungsfehler aufweisen. Bitte aktualisiere deine Browser-Lesezeichen und verwende stattdessen die Standard-Druckfunktion des Browsers.

Sampling and Statistical Questions

Introduction

Sampling and Statistical Questions are central ideas in Statistics. They help you ask questions that can be answered with data, choose a useful group to study, recognize possible bias, and make careful conclusions about a larger group.

Imagine that your school wants to know how students travel to school. Asking every student may be possible, but in a larger town or national study, asking everyone could take too much time or cost too much. A well-chosen sample can provide useful evidence about a larger population. The quality of the conclusion depends on both the question you ask and the way you collect the data.

In this aiMOOC, you will learn to:

Recognize statistical questions: Decide whether a question expects variability in its answers.
Identify populations: State clearly which full group you want to understand.
Identify samples: Describe the smaller group from which data are actually collected.
Use random sampling ideas: Explain why random selection can help make samples more representative.
Detect sampling bias: Find ways in which a sample or survey can systematically favor some responses.
Make cautious inferences: Use sample results as evidence about a population while remembering sampling variability.

Statistical Questions

What Makes a Question Statistical?

A statistical question is a question that can be investigated with data and that anticipates variability in the data. Variability means that the values or categories you collect are not expected to be identical.

For example, the question How many minutes do students in Grade 8 spend reading for pleasure on a typical school day? is statistical. Different students are likely to give different answers, and you can collect data to describe the distribution.

By contrast, How many minutes did Jordan read yesterday? asks for one particular value about one particular case. It does not, by itself, ask you to study variability across a group.

A useful test is to ask yourself:

Does the question refer to a group or repeated observations?
Could the answers vary?
Can data be collected to investigate the question?
Is the group, variable, and time frame clear enough to study?

Improving a Statistical Question

A vague question can often be improved by naming the population, variable, and time frame. Compare these examples:

Too vague: How much do teenagers exercise?

More useful: How many minutes of physical activity do students in Grades 7 and 8 at our school report on a typical weekday?

The improved question tells you who is being studied and what measurement to collect. It also gives a time frame, making answers easier to compare.

A statistical question does not have to use numbers. Which type of school lunch do Grade 7 students prefer? is also statistical because the responses can vary across categories.

Variables and Data

A variable is a characteristic that can take different values. Some variables are quantitative, such as height, travel time, or number of books read. Others are categorical, such as favorite school subject, transport type, or whether a student agrees with a statement.

When you plan a statistical investigation, the question and variable should match. If your question is about travel time, collecting only transport type will not answer it fully.

Populations and Samples

Population

The population is the entire group you want to learn about. A population must be defined clearly. It could be all students in one grade, all households in a city, all trees in a park, or all products made by a factory during a certain week.

If your question is What proportion of Grade 8 students at our school bring lunch from home?, then the population is all Grade 8 students at that school.

Sample

A sample is the smaller group from which you actually collect data. Researchers often use samples because studying a whole population can be slow, expensive, or impractical.

A sample is useful only if the way it was selected allows reasonable conclusions about the population. A very large sample can still be misleading if it systematically leaves out part of the population.

Census and Sample Survey

A census collects data from every member of the population. A sample survey collects data from only part of the population.

A census can provide very complete information, but it may require more time and resources. A sample survey can be faster and easier, but it introduces sampling variability: different samples from the same population can produce different results.

Random Sampling and Representativeness

Simple Random Sampling

In a simple random sample, each possible sample of a given size has an equal chance of being selected. In school investigations, you can approximate this idea by giving every student an identifier and using a random-number method to choose the sample.

Random selection helps protect against personal choice influencing who enters the sample. It does not guarantee a perfect sample, but it makes systematic selection bias less likely when the sampling process is carried out correctly.

Representative Samples

A representative sample resembles the population in ways that matter for the question being studied. For example, if a school has students from several grade levels but a survey includes only one grade, the sample may not represent the whole school.

Representativeness is not achieved simply by choosing people who are easy to contact. Convenience can create bias when the convenient group differs from the population in relevant ways.

Other Sampling Designs

A simple random sample is not the only useful design. In a stratified sample, the population is divided into important groups, called strata, and random samples are taken from those groups. This can help ensure that smaller but important groups are included.

A systematic sample selects members using a regular interval after an appropriate starting point. A cluster sample selects whole groups, or clusters, rather than selecting individuals from across the entire population. These methods require careful design so that the final sample still supports the intended conclusion.

Sampling Bias

What Is Bias?

Sampling bias occurs when the sampling process systematically favors some members of the population over others. Bias is different from ordinary sample-to-sample variation. Random variability can change direction from one sample to another, while bias tends to push results in a particular direction.

Common sources of bias include:

Convenience sampling: Asking only people who are easiest to reach.
Voluntary response: Allowing people to decide for themselves whether to participate, especially when people with strong opinions may be more likely to respond.
Undercoverage: Leaving out part of the target population from the sampling process.
Nonresponse: Selected participants do not respond, and nonresponders may differ from responders.
Response bias: The wording, setting, or method of asking influences responses.

Biased Questions and Neutral Wording

Sampling is only part of good data collection. Survey wording also matters.

Loaded wording: Do you agree that our school should finally improve its outdated playground?

More neutral wording: Do you support, oppose, or have no opinion about the proposed playground changes?

The second question reduces pressure toward one answer. Neutral questions should avoid emotionally loaded words, hidden assumptions, and confusing double questions.

Sampling Variability

Why Samples Differ

Suppose two students independently take random samples of 50 students from the same school. In one sample, 32 students prefer Option A. In the other, 29 prefer Option A. The sample proportions are 64 percent and 58 percent. Neither sample must be wrong. The difference can occur because the samples contain different people.

This is sampling variability. If you repeatedly take random samples from the same population, the resulting statistics usually vary.

Sample Size and Stability

Larger random samples generally show less sample-to-sample variability than smaller random samples. However, a large sample does not automatically remove bias. If a survey of 2,000 students includes only members of one sports club, the sample may be much less useful than a smaller random sample drawn from the full school.

Good investigations therefore consider both sample size and sampling method.

Simulation as a Tool

You can study sampling variability with a simulation. For example, a bag with colored counters can represent a population. Draw a random sample, record the proportion of each color, replace the counters, mix again, and repeat. Comparing results across many samples helps you see how much a statistic can vary simply because of random selection.

From Sample Data to Population Conclusions

Describing the Sample First

Before making a conclusion about a population, describe the data you actually collected. Depending on the question, useful summaries may include proportions, mean, median, range, interquartile range, or graphical displays.

A histogram groups quantitative data into intervals and shows how frequently values occur. A box plot can summarize the center and spread of a quantitative distribution.

Making an Inference

An inference uses sample evidence to make a claim about a larger population. A strong middle-school inference should be cautious and connected to the sampling method.

For example: In a random sample of 80 Grade 8 students, 60 percent preferred a later library closing time. This sample provides evidence that a majority of Grade 8 students at the school may prefer the later closing time.

A weaker statement would say that exactly 60 percent of every Grade 8 student prefers it. The sample result is an estimate, not a perfect measurement of the entire population.

Comparing Two Samples

You can learn more by comparing multiple random samples from the same population. If results are similar, that consistency supports greater confidence in the general pattern. If results differ greatly, you should investigate sample size, sampling variability, and possible bias before making a strong claim.

Planning a Statistical Investigation

A useful investigation can be organized as a cycle:

Ask: Write a clear statistical question that expects variability.
Define: Identify the population you want to understand.
Plan: Choose a sampling method that gives the population a fair chance to be represented.
Collect: Gather the data consistently and ethically.
Analyze: Summarize patterns, center, spread, and variation as appropriate.
Interpret: Use sample evidence to make a cautious conclusion about the population.
Communicate: Explain the method, evidence, limitations, and conclusion clearly.

Before you collect data from people, protect privacy, avoid unnecessary personal questions, and follow your school rules for surveys and interviews.

Worked Examples

Example 1: Statistical or Not?

Question A asks, How many pets does each student in our class have? This is statistical because the answers are expected to vary.

Question B asks, How many pets does Lee have? This asks for one specific value and does not investigate variability across a group.

Example 2: Detecting Bias

A student wants to know whether students at school want more after-school sports. The student surveys only people leaving basketball practice. That sample is likely biased because students at sports practice may have different opinions about after-school sports than the full school population.

A better plan is to build a list of all students and randomly select students from across the school, or to randomly select students within each grade.

Example 3: Interpreting a Sample Proportion

A random sample of 100 students finds that 47 choose cycling as their preferred form of weekend exercise. The sample proportion is 47 percent. You can use this as evidence about the school population, but you should not claim that exactly 47 percent of all students would choose cycling. Another random sample could give a somewhat different percentage.

Interactive Tasks

Quiz: Test Your Knowledge

Which question is statistical? (How many hours of sleep do Grade 8 students usually get on school nights) (!How many hours did Mia sleep last night) (!What is the school phone number) (!How many desks are in Room 12)

What is the population in a survey about all Grade 7 students at a school? (All Grade 7 students at the school) (!Only students who answer the survey) (!Only students in one mathematics class) (!The survey questions)

What is a sample? (A smaller group studied to learn about a population) (!Every member of the population) (!A graph showing the final results) (!A question with only one possible answer)

Why is random sampling useful? (It reduces the chance that personal selection creates systematic bias) (!It guarantees every sample has the same results) (!It removes all sampling variability) (!It makes sample size unimportant)

Which method is most likely to create convenience bias? (Surveying only students sitting near the researcher) (!Randomly selecting names from the full student list) (!Using a random number generator on student identifiers) (!Randomly sampling from each grade level)

What does sampling variability mean? (Different random samples can produce different statistics) (!Every survey question has the same answer) (!Biased samples always give identical results) (!Population values change whenever a sample is taken)

Which statement about sample size is correct? (Larger random samples usually have less sample-to-sample variability) (!A large biased sample is always representative) (!Sample size matters only for categorical data) (!A sample of one is always enough)

Which wording is the most neutral? (Do you support oppose or have no opinion about the proposal) (!Do you support the excellent new proposal) (!Why should everyone accept the needed proposal) (!Do you agree the old plan is obviously unfair)

What should you do before generalizing from a sample to a population? (Check how the sample was selected) (!Assume the sample perfectly matches the population) (!Ignore missing groups) (!Use only the largest number in the data)

Which conclusion is most appropriate from a random sample? (The sample provides evidence about the population but another sample may differ) (!The sample proves every population member has the same opinion) (!The sample result must equal the population result exactly) (!Sampling variability can be ignored)

Memory Game

Population	Entire group you want to understand
Sample	Smaller group from which data are collected
Variability	Differences among data values or sample results
Bias	Systematic influence that can distort results
Inference	Conclusion about a population based on sample evidence
Randomization	Chance-based selection used to reduce selection bias

Drag and Drop

Match the correct terms.	Topic
Statistical question	A question that expects variable data
Representative sample	A sample that reflects important features of the population
Convenience sample	A sample chosen because members are easy to reach
Sampling variability	Natural differences among results from different random samples
Neutral wording	Survey language that avoids pushing people toward one response

...

Crossword Puzzle

Population	What word names the entire group you want to study?
Sample	What word names the smaller group from which data are collected?
Randomization	What chance-based process can help reduce selection bias?
Variability	What word describes differences among data values or sample results?
Representative	What word describes a sample that reflects the population well?
Inference	What word means a conclusion about a population based on sample evidence?

LearningApps

Cloze Text

Open-Ended Tasks

Easy

Question Sort: Create a two-column poster with six statistical questions and six non-statistical questions. For each statistical question, underline the part that shows whose data you would collect.
Population and Sample Hunt: Find four examples of surveys or data studies in news, school materials, or public websites. For each one, identify a possible population and sample.
Bias Detective: Write three short survey scenarios that contain sampling bias. Exchange them with a partner and explain how each sample could be improved.
Data Display Sketch: Collect one small set of class-safe numerical data and create a simple histogram or box plot. Write two sentences describing what varies.

Standard

School Survey Design: Write one statistical question about school life, define the population, and design a random sampling plan that could be carried out fairly.
Sampling Simulation: Use colored counters, cards, or a digital randomizer to take at least ten samples from a fixed population. Compare the sample proportions and describe the variability you observe.
Question Wording Lab: Write four pairs of survey questions in which one version is loaded and the other is neutral. Test the pairs with classmates and discuss whether wording appears to affect responses.
Mini Interview Study: With teacher permission, interview a small, appropriately selected sample about a school-related topic. Record the sampling method, summarize the data, and state one limitation.

Advanced

Competing Samples Investigation: Draw several random samples of two different sizes from the same known data set. Compare how much the resulting means or proportions vary and explain the effect of sample size.
Sampling Methods Video: Produce a short instructional video that demonstrates a simple random sample, a convenience sample, and a stratified sample using the same fictional population.
Local Data Project: Visit a library, school office, museum, park, or other approved place and identify a question that could be studied with a sample. Propose an ethical data-collection plan and justify the sampling method.
Evidence Based Recommendation: Conduct a small statistical investigation, analyze your sample, and write a recommendation for a real audience. Include the statistical question, population, sampling method, evidence, limitations, and a cautious inference.

Learning Assessment

Assessing a Sampling Plan: A school surveys only students in the cafeteria about preferred lunch changes. Explain whether the sample can represent the whole school, identify possible bias, and redesign the plan.
Comparing Sample Results: Two random samples from the same population produce proportions of 42 percent and 51 percent. Explain how both results can be reasonable and what additional information would help you judge the strength of the evidence.
Evaluating a Claim: A website says that 90 percent of teenagers prefer one brand after surveying 500 followers of the brand. Analyze the population, sample, selection method, and whether the claim can be generalized.
Designing an Investigation: Create a statistical question relevant to your class, define the variable and population, select a sampling method, and justify why your design reduces bias.
Transfer to a New Context: A wildlife team cannot count every bird in a large wetland. Explain how ideas from school surveys can transfer to this situation, including sampling, representativeness, variability, and cautious inference.
Communicating Uncertainty: Rewrite an overconfident conclusion from sample data so that it accurately describes what the evidence supports and what remains uncertain.

Evidence of Learning

Knowledge: You can explain statistical questions, populations, samples, random sampling, bias, representativeness, sampling variability, and inference.

Skills: You can classify questions, identify populations and samples, design fair sampling procedures, detect bias, summarize sample data, compare repeated samples, and justify conclusions.

Products: Useful evidence may include a survey plan, sampling simulation, graph, written analysis, poster, interview summary, video explanation, or short statistical report.

Transfer: You can apply sampling ideas to school decisions, public opinion, quality control, environmental studies, health information, and other situations in which a smaller group is used to learn about a larger group.

Communication: You can state what the data support, distinguish a sample result from a population truth, identify limitations, and use cautious language when uncertainty remains.

OERs on the Topic

The English Wikipedia articles on sampling, statistical populations, statistical inference, and sampling error provide useful extensions when you want to explore the ideas in greater depth.

Linked Learning Areas

Sampling and Statistical Questions

aiMOOC Projects