Experimental design
Introduced| English | Meaning |
|---|---|
| independent variable/ˌɪndɪˈpendənt ˈveərɪəbl/ | An input or factor whose value is chosen or varied to examine its effect on another quantity. |
| dependent variable/dɪˈpendənt ˈveərɪəbl/ | The measured or calculated outcome whose response to changes in another variable is studied. |
| control variables/kənˈtrəʊl ˈveərɪəblz/ | Factors kept the same in an investigation so that the effect of the independent variable can be assessed. |
| fair test/feə test/ | fair test |
| anomalies/əˈnɒməliz/ | anomalies |
Doing experiments well
- A good experiment gives results you can trust and repeat.
- That takes careful experimental design.
Experiment design route
Follow a fair test from question to reliable conclusion.
The independent variable is the one that you:
Difficulty 1/5 · 100 coins + a small bonus on first completion today
You change the independent variable.
The dependent variable is the one you:
Difficulty 1/5 · 100 coins + a small bonus on first completion today
You measure the dependent variable.
Variables kept the same to make a test fair are called ______ variables.
Difficulty 1/5 · 100 coins + a small bonus on first completion today
Control variables are held constant.
Variables
- The independent variable 自变量 is the one you deliberately change.
- The dependent variable 因变量 is the one you measure.
- Control variables 控制变量 are kept the same so the test is fair.
Simple distillation separates a solvent from a solution by evaporating then condensing it
For a fair test, you should change:
Difficulty 1/5 · 100 coins + a small bonus on first completion today
Only one variable changes in a fair test.
A fair test 公平测试
- To be a fair test, only the independent variable should change.
- This makes sure any change in the result is due to that variable alone.
A flame test: the metal ion gives the flame a colour
Repeating measurements and averaging improves reliability.
Difficulty 1/5 · 100 coins + a small bonus on first completion today
Repeats reduce the effect of random error.
Reliable results
- Repeat measurements and take an average to improve reliability.
- Spot and discard obvious anomalies 异常值 (odd results) before averaging.
Worked example. Testing how temperature affects reaction rate: temperature is the independent variable, rate is the dependent variable, and concentration and volume are controlled. Repeat each temperature three times for reliability.
You've got it
- independent variable = what you change; dependent = what you measure; control = kept the same
- a fair test changes only the independent variable
- repeat and average for reliability; discard anomalies