When scientists start to investigate something they usually have a hypothesis that they are testing. This means they have an idea about what will happen when they explore something or take some readings, but they need the evidence to either confirm their thinking or suggest they need to think again.
From this they can make a prediction. It is easy to get mixed up between hypotheses and predictions. For example, a hypothesis might be made about the way that springs behave when they are loaded. From this a prediction can be made about what will happen to a spring when force is added.
Often an experiment involves things that can change, known as variables. Variables need to be identified, so they can then either be changed or controlled. There are three kinds of variable:
Scientists often want to find out if changing one variable makes a difference to other variables. In many (though not all) investigations the variables are kept constant - the control variables, apart from one which is varied - the independent variable. The effects of the independent variable is then determined by monitoring the dependent variable.
An example would be investigating whether increasing the temperature of the reactants might alter the rate of the reaction. As it is the temperature which is changing, that would be the independent variable. The changing temperature alters the rate of reaction, therefore the reaction rate is the dependent variable. When carrying out the experiment, care has to be taken that other variables that affect the rate of reaction, such as concentration of reactants, are kept constant. These are control variables.
The values are the measurements used for the independent variable. If, for example, one of the variables in an experiment was length, it would be important to decide the maximum and minimum values, and also the intervals between values. If enzyme activity at different pH values was being investigated, a decision would have to be made on what values of pH to use. This decision would take into account elements such as available equipment, time constraints, and safety.
When measurements are being taken, it is usually appropriate to repeat them. Sometimes, there are lots of possible readings that could be taken. For example if the distribution of daisies on a playing field was being explored, it wouldn’t be necessary to count every one; however, it wouldn’t be a good idea to just look at the ones close to the fence. A sampling technique should be used to decide which ones to look at. It might, for example, involve the method of randomly placing quadrats. A mean is then calculated.