Design effective surveys and questionnaires. Collect primary data systematically and understand sampling methods and potential sources of bias.
Click on each step to learn how to collect reliable and unbiased data:
Start with a clear question about what you want to find out. This guides all your data collection decisions.
Decide whether to use surveys, observations, interviews, or existing data sources based on your research question.
Create clear, unbiased questions that will give you the information you need. Avoid leading questions and confusing language.
Choose who you'll collect data from. Your sample should represent the larger group you want to understand.
Gather your data systematically and check for errors, missing responses, or unusual patterns that might indicate problems.
Data collection is the foundation of all statistical analysis. The quality of your conclusions depends entirely on the quality of your data.
Design your own survey question and see how different wordings can affect responses:
Preview: What is your favorite subject in school?
Learn to spot and avoid bias in your survey questions. Biased questions lead to unreliable results.
"Don't you think mathematics is the most important subject?"
Problem: Leading question that suggests the "right" answer.
"Which subject do you think is most important for your future?"
Better: Neutral wording that doesn't influence the answer.
"How much do you love playing video games?"
Problem: Assumes everyone loves playing video games.
"How often do you play video games?"
Better: Asks about frequency without assuming feelings.
Click on dots to select a sample from this population of 50 people. Try different sampling methods:
Data you collect yourself through surveys, interviews, or observations.
Advantages: Exactly what you need, current, reliable
Disadvantages: Time-consuming, can be expensive
Data that already exists from other sources like reports, websites, or databases.
Advantages: Quick to access, often covers large populations
Disadvantages: May not fit your exact needs, could be outdated
Numerical data that can be measured and counted.
Examples: Height, age, test scores, number of pets
Analysis: Can calculate averages, create graphs, find patterns
Descriptive data about qualities, characteristics, or opinions.
Examples: Favorite colors, opinions, descriptions
Analysis: Look for themes, categories, common responses
Research Question: "What lunch options do students prefer?"
Method: Survey all Year 6 students (120 students)
Sample Questions:
Results: 85% prefer hot meals, pizza most popular
Research Question: "How do families in our area recycle?"
Method: Interview 30 families (systematic sampling)
Sample Questions:
Results: 70% recycle, paper/plastic most common
Research Question: "How much time do students spend on devices?"
Method: Survey + diary tracking for 1 week
Sample Questions:
Results: Average 3.5 hours/day, games most common
Research Question: "How do students travel to school?"
Method: Observation + survey at school gates
Data Collection:
Results: 45% walk, 30% car, 25% bus/bike
The school wants to improve the library. Design a survey to find out what students think.
Identify which question is most likely to give unbiased results:
You want to survey 50 students from a school of 500. Which sampling method is best?
What should you do before collecting data from other students?
You want to know about students' breakfast habits. Which question is best designed?
Example: "How much do you love our amazing school lunch?"
Problem: Assumes students love lunch and pressures positive answers.
Fix: "How would you rate our school lunch?" (with balanced scale)
Example: Only surveying students in the computer lab about technology use
Problem: Sample doesn't represent all students - only tech-interested ones.
Fix: Use random sampling from all students in the school.
Example: "How often do you not avoid eating vegetables?"
Problem: Double negative makes question hard to understand.
Fix: "How often do you eat vegetables?" (Simple and clear)
Example: "What's your favorite sport?" (Only lists team sports)
Problem: Doesn't include individual sports or "I don't like sports"
Fix: Include comprehensive options plus "Other" and "None"
Example: Surveying only 3 students about school-wide issues
Problem: Results can't represent the whole school population.
Fix: Use larger, representative sample (at least 30 for basic statistics)
Example: "How much money does your family earn?"
Problem: Too personal, may get dishonest or no answers.
Fix: Ask less directly or use ranges if really necessary
You have mastered data collection and survey design! Here's what you can now do:
Create clear, unbiased questions that gather reliable information
Select representative samples using appropriate sampling methods
Identify and eliminate sources of bias in questions and sampling
Gather primary and secondary data systematically and ethically
You've mastered the fundamentals of data collection and survey design
You can now gather reliable data to answer important questions!
Ready for the ultimate challenge? In Chapter 9.6, you'll combine all your statistics skills to solve complex real-world problems!