Thinking and Working Scientifically: Carrying out scientific enquiry

The first step in any scientific enquiry is planning. This starts with a hypothesis: a testable statement that proposes an explanation for an observation. For example, 'The rate of photosynthesis increases with light intensity'. From this, you can make a specific prediction for your experiment, like 'If I move the lamp closer to the pondweed, the plant will produce more bubbles of oxygen per minute'. To test this, you must identify your variables. The Independent Variable (IV) is the one thing you change (light intensity). The Dependent Variable (DV) is the thing you measure to see the effect of the change (number of bubbles per minute). All other factors that could affect the outcome, called Control Variables, must be kept the same to ensure a fair test (e.g., temperature, carbon dioxide concentration, type of plant).

A good method is a clear, numbered, step-by-step set of instructions that another person could follow to repeat your experiment exactly. It should include the specific apparatus you will use and the quantities of substances. Before starting, you must perform a risk assessment. This involves identifying hazards (e.g., sulfuric acid, a Bunsen burner), assessing the risk (e.g., acid splashing in eyes causing blindness, burns from the hot flame), and stating the control measure to minimise that risk (e.g., wear safety goggles, tie back long hair). Choosing the right equipment is also vital. For example, use a measuring cylinder for approximate volumes but a burette or pipette for precise, accurate measurements. A gas syringe is better for collecting and measuring a gas than counting bubbles.

When carrying out your experiment, your goal is to collect high-quality data. This means the data should be accurate, precise, and reliable. Accuracy is how close a measurement is to the true value. You can improve accuracy by using correctly calibrated instruments and avoiding errors like parallax error (reading a scale from an angle). Precision is how close repeated measurements are to each other, which depends on the instrument's scale (e.g., a ruler with mm markings is more precise than one with only cm markings). Reliability is about consistency. To make your results reliable, you should repeat your readings at least three times and then calculate a mean (average). This process also helps you to spot and deal with anomalous results – readings that don't fit the pattern and can be excluded from the mean.

Once you have collected your data, you need to present it clearly. The best way to do this is in a results table. A good table has clear headings at the top of each column, includes the correct units in the headings (not in the body of the table), and has a border drawn with a ruler. The independent variable goes in the first column, and the dependent variable in the subsequent columns. After tabulating your data, you often need to draw a graph to visualise the relationship between your variables. Remember the acronym SLAP: Scale (use a sensible scale that covers at least half the graph paper), Line (draw a single, smooth line or curve of best fit), Axes (label both axes with the quantity and units, with the IV on the x-axis and DV on the y-axis), and Points (plot your points accurately using a small 'x' or a point in a circle). A line of best fit shows the trend in the data and helps you spot anomalies, which are points far away from the line.

After analysing your data, you must draw a conclusion. A conclusion is a short summary that states what you have found. It should link the independent and dependent variables and describe the relationship or trend shown in your graph or table (e.g., 'As the temperature increases, the rate of reaction increases'). You must support your conclusion with evidence by quoting data. For example, '...for instance, at 20°C the rate was 10 cm³/s, but at 40°C the rate increased to 22 cm³/s'. Finally, you must evaluate your experiment. This means identifying limitations or sources of error in your method (e.g., 'It was difficult to judge the exact moment the reaction stopped') and suggesting specific, realistic improvements (e.g., 'Use a colorimeter to measure the colour change more objectively'). A good evaluation explains *how* the improvement would lead to more accurate or reliable data.