Introduction: Being a Scientific Detective

In Biology, performing an experiment is only half the battle. The real skill lies in looking back at what you did and asking: "Can I actually trust these results?"

Evaluating methods and results is like being a detective. You are looking for "clues" that might suggest the experiment wasn't a fair test, or that the data contains hidden errors. Whether you are investigating the vitamin C content in fruit juice or the tensile strength of plant fibres, the ability to criticise and improve a method is a vital skill for your Unit 3 exam.

1. Evaluating the Method: Was it a Fair Test?

When you "evaluate a method," you are looking at the procedure. You need to decide if the experiment was designed well enough to produce valid results (results that actually measure what they are supposed to).

Control Variables: The "Stay the Same" Rule

To make an experiment a fair test, every variable except the independent variable must be controlled (kept constant). If a student investigates the effect of temperature on enzyme activity (Core Practical 4) but forgets to control the pH, the results are unreliable because the change in activity might be due to the pH changing, not the temperature.

Quick Review Tip: In exam questions, if you are asked to "criticise" a method, look for variables that the researcher forgot to mention. Did they control the volume? The concentration? The age of the biological material?

Range and Intervals

A good method needs a wide range of values for the independent variable and small intervals between them.
Example: If you only test enzyme activity at \(10^{\circ}C\), \(20^{\circ}C\), and \(30^{\circ}C\), you might miss the "optimum" temperature if it happens to be at \(25^{\circ}C\).
Improvement: Suggest a wider range (e.g., \(0^{\circ}C\) to \(60^{\circ}C\)) and smaller intervals (e.g., every \(5^{\circ}C\)).

Repeatability: The Power of Three

A single result could be a fluke. We use repeat readings (usually at least three) to:

  1. Identify anomalies (inconsistent readings).
  2. Calculate a mean, which is more likely to be close to the true value.

Key Takeaway: A strong method controls all variables, uses a wide range of measurements, and includes repeats to ensure reliability.

2. Evaluating Results: Spotting the Errors

Once the data is in the table, you need to look for uncertainties and errors. Don't worry if this seems tricky; it’s all about looking for things that don't "fit."

Anomalies: The Odd Ones Out

An anomaly is a result that does not fit the trend of the rest of the data.
What to do: If you spot one, you should check the reading (if possible) or discard it before calculating your mean. If you include an anomaly in your mean, your final result will be inaccurate!

Significant Figures: Keep it Consistent

A common mistake is providing results with too much precision. Your results should be reported to the limits of the least accurate measurement.
Analogy: If your ruler only measures to the nearest millimetre, you can't claim a plant fibre is \(10.2345 mm\) long. That's just guessing!

Uncertainty and Systematic Error

Every piece of equipment has a limit.

  • Uncertainty: This is the "plus or minus" range of a measurement (e.g., a thermometer might be accurate to \(\pm 0.5^{\circ}C\)).
  • Systematic Error: This happens when the equipment is consistently "off" (e.g., a weighing scale that isn't set to zero properly). This shifts all your results in one direction.
Note: For more on how to calculate percentage error, see the "Measurement, Uncertainty and Systematic Error" chapter.

3. Evaluating Conclusions: Correlation vs. Causation

Just because two things happen at the same time doesn't mean one caused the other. This is a huge part of Topic 1 (Cardiovascular Disease) and frequently appears in Unit 3.

Correlation: A relationship between two variables (e.g., as \(X\) increases, \(Y\) increases).
Causation: One variable directly causes the change in the other.

Example: There is a correlation between ice cream sales and shark attacks. Does ice cream cause shark attacks? No! The confounding variable is warm weather, which makes people buy ice cream and go swimming in the sea.

Did you know?

In exams, you are often asked to "Comment on the conclusion." You should look for:

  • Sample Size: Was the study done on 5 people or 5,000? Small samples are less representative.
  • Standard Deviation: If the "error bars" on a graph overlap, the difference between two groups might not be significant.
  • Uncontrolled variables: Could something else have caused the result?

4. Identifying Improvements

If an exam question asks you to suggest improvements to a procedure, think about these "Big Three":

  1. Precision: Use a more precise piece of equipment (e.g., a colorimeter instead of just looking at colour standards with your eyes in Core Practical 1).
  2. Reliability: Increase the number of repeats to better identify anomalies.
  3. Validity: Use a water bath to keep temperature constant or a buffer solution to keep pH constant.

Summary Checklist for the Exam

When evaluating, always ask yourself:
\( \bullet \) Are the units correct (e.g., \(mol \cdot dm^{-3}\))?
\( \bullet \) Are there enough repeat readings?
\( \bullet \) Was the temperature/pH/concentration controlled?
\( \bullet \) Does the data actually support the conclusion, or is it just a coincidence?
\( \bullet \) Are there any "odd" results (anomalies) that should be ignored?

Key Takeaway: Evaluation isn't about saying an experiment was "bad"—it's about identifying the limitations of the data and suggesting how to make the evidence stronger.