Collecting data

1Summary 重點摘要

Good data makes good conclusions: define the purpose, choose a method (questionnaire, interview, observation), sample big and random, never lead the answer.

2Vocabulary cards Vocabulary

3Step by step Steps

1. 1. What is data collection?Data collection is the planned process of obtaining information, and it is the first step of all statistics. To find the class's favourite subject, ask everyone and record the answers; to track the weather, record the temperature daily; to know an object's weight, measure it directly. Raw data is the raw material of every statistic: collect it well and the charts and conclusions that follow are trustworthy; collect it carelessly and no clever analysis afterwards can repair the damage.
2. 2. How to collect data?There are three main collection methods, each suited to a different situation. A survey asks people questions, good for opinions and preferences; observation watches and records on the spot, good when asking is impossible or inappropriate; an experiment measures under controlled conditions, good when precise values are needed. Match the method to the purpose: opinions call for a survey, traffic counts call for observation, and plant growth under different light calls for an experiment. The wrong method yields data that cannot answer the original question.
3. 3. SurveyA survey collects data by asking people questions, in three common forms: a questionnaire written on paper or online that respondents fill in themselves and that gathers many responses at once; an interview with face-to-face questions that can probe details; and a phone survey that is quick but brief. To survey the class's favourite subject, one closed question — 'Which do you prefer: Chinese / English / Mathematics?' — collects the whole class in a single lesson. Keep questions neutral and clear, since leading wording biases the answers.
4. 4. ObservationObservation records facts without interfering with the subject. Examples: noting the temperature every hour, counting vehicles passing a junction every five minutes, or watching how much classmates eat at lunch. Its strength is authenticity — the data reflects what naturally happens; its cost is time, and factors like weather and crowds are hard to control. Design the recording sheet first, with columns for time, vehicle type and count, so on the spot you only tally strokes and the tidying afterwards is easy.
5. 5. ExperimentAn experiment obtains measurements by operating under controlled conditions, most common in science. Examples: weighing and measuring different objects, testing plant growth under different light levels, comparing the insulation of different materials. Experiments are accurate and repeatable; the price is equipment, time, and conditions that cannot always be fully controlled. Write the steps beforehand, read every measurement twice and average, and write units with every number — these details decide how trustworthy the data is.
6. 6. Population vs sampleThe population is the whole group under study; a sample is the part selected for the survey. If the school has 1000 students, the population is all 1000; surveying a chosen 100 makes them the sample. Surveying the entire population is ideal but expensive, so samples are used to infer the population — provided the sample is representative: selection must be random and must cover every level and group. Asking only school-team members about the whole school's sport habits gives a badly biased sample and an untrustworthy conclusion.
7. 7. Organize dataCollected data must be organised before it shows any meaning. Take the class's favourite subject: Chinese 25, English 20, Mathematics 30. Three organising tools: a bar chart turns the counts into bars that compare instantly; a pie chart shows each part's share of the whole — Mathematics is 30 out of 75, or 40%; an average summarises the overall level. After organising, the ranking is obvious and the report persuasive; an untidied list of raw answers lets no reader draw any conclusion.

4Diagrams Diagrams

5Worked example Example

6Common mistakes Mistakes

7Quick check Quick check

8Recap Recap

• Data collection: process of getting data
• Methods: survey, observation, experiment
• Survey: ask people
• Population: all; sample: part
• After collecting, organize (chart, average)