Practise › Questions › Variables and controls: what an experiment can show

Variables and controls: what an experiment can show questions

Independent, dependent and controlled variables and how to choose the levels of one; the difference between a control variable and a control experiment; what a negative control and a positive control each rule out; confounding variables; how a correlation is argued up to a causal claim; and validity, accuracy, precision, repeatability and reproducibility as five separate ideas.

5 original questions · 18 marks · the variables and controls: what an experiment can show notes · Scientific method and quantitative biology

Every question here is written for this library rather than taken from a past paper. Write your answer out before opening a mark scheme: the schemes award marks point by point, and the points are much easier to see when you have something of your own to compare against.

• 4 marks · Explain

  A student measures oxygen production by pondweed at five distances from a filament lamp, and the rate rises as the lamp is brought closer. Explain why repeating the whole investigation ten times would not make the conclusion valid.

  Mark scheme

  1. B1 temperature is a confounding variable here: a filament lamp is mostly a heater, so bringing it closer raises the temperature of the water as well as the light intensity
  2. B1 either the light or the warmth could have raised oxygen production, and the readings collected cannot separate the two effects
  3. B1 every repeat changes both quantities together in exactly the same way, so more repeats trace the same two routes again rather than telling them apart
  4. B1 repeats improve precision, not validity; the fix is a change of design, such as a water-filled flask acting as a heat filter or an LED source, with the temperature measured at each distance to show it stayed put

• 4 marks · Compare

  Compare accuracy with precision, and compare repeatability with reproducibility, as descriptions of a set of measurements.

  Mark scheme

  1. B1 accuracy is closeness to the true value, whereas precision is how closely repeat readings agree with one another
  2. B1 a random error such as misjudging a meniscus scatters readings either side of the true value and damages precision, whereas a systematic error such as an unzeroed balance shifts every reading the same way and damages accuracy
  3. B1 readings can therefore be precise and inaccurate at the same time, which is why agreement between repeats is no evidence at all of being right
  4. B1 repeatability is the same person getting the same results with the same method and apparatus, whereas reproducibility is somebody else, or different apparatus or a different method, getting them

• 4 marks · Suggest

  A student tests six extracts of seaweed for antibacterial activity by soaking paper discs in each and placing them on a lawn of bacteria. No clear zone appears around any of the six discs. Suggest two control experiments the student should have set up, and suggest what each would have shown.

  Mark scheme

  1. B1 a negative control: a disc soaked in the solvent the extracts were made up in, and nothing else
  2. B1 if that disc produces a zone, the solvent was killing the bacteria and any zone around an extract could not be credited to the seaweed
  3. B1 a positive control: a disc carrying an antibiotic the bacterium is known to be sensitive to
  4. B1 a zone around it shows the bacteria were alive, the plate was seeded properly and a zone would have been visible had there been one to see; with no zone anywhere the honest conclusion is that the investigation failed, not that the extracts have no effect

• 4 marks · Evaluate

  A survey of 5000 adults finds that those with a higher intake of oily fish have fewer heart attacks over the following ten years. A newspaper reports that eating oily fish prevents heart attacks. Evaluate that conclusion.

• 2 marks · State

  State what a negative control is designed to rule out, and state what a positive control is designed to rule out.

Practise variables and controls: what an experiment can show one question at a time

The player marks nothing for you. It shows one question, waits, then shows the scheme so you can mark yourself, and brings a question back sooner when it went badly.