Introduction to Evaluation
In Biology, performing an experiment is only half the battle. The real science happens when you look back at what you did and ask: "How much can I actually trust these results?"
Evaluation isn't about admitting you did a "bad job." Instead, it is a sophisticated process of identifying the limitations of your equipment, the precision of your measurements, and the validity of your conclusions. This chapter will help you master the skills needed to critique any biological investigation—a key requirement for your OCR exams (AO3 skills).
1. Precision vs. Accuracy
These two terms are often used interchangeably in daily life, but in A Level Biology, they have very specific meanings. Understanding the difference is crucial for evaluating your data.
Accuracy
Accuracy refers to how close a measurement is to the true value. If you are measuring the pH of a solution that is known to be exactly \(pH 7.0\), and your probe reads \(pH 7.0\), your measurement is highly accurate.
Precision
Precision refers to how close repeated measurements are to each other. It relates to the "consistency" of your results. If you measure the same solution three times and get \(pH 6.1\), \(pH 6.1\), and \(pH 6.2\), your results are precise (because they are close together), even if they aren't accurate (because they are far from the true value of \(7.0\)).
Analogy: Think of a dartboard. If all your darts hit the bullseye, you are both accurate and precise. If all your darts land in a tight cluster in the top-left corner, you are precise but not accurate!
2. Identifying Anomalies
An anomaly (or outlier) is a result that does not fit the overall trend or pattern of the data.
How to spot them: When you look at your repeats for a specific condition, look for a value that is significantly different from the others. For example, if your time measurements are \(32s\), \(34s\), and \(58s\), the \(58s\) is likely an anomaly.
What to do with them:
1. Identify: Circle them on your results table.
2. Investigate: Try to figure out why it happened (e.g., did the temperature drop during that specific test?).
3. Exclude: Do not include anomalies when calculating your mean (average), as they will skew your results and make them less representative.
3. Understanding Uncertainties and Margins of Error
No piece of equipment is perfect. Every time you take a reading, there is a small "margin of error" known as uncertainty. This is usually half of the smallest scale division on your instrument.
Calculating Percentage Error
To evaluate how much an uncertainty actually matters, we calculate the percentage error. This helps us see the relative impact of the equipment's limitations on our final result.
The formula you need to remember is:
\(\text{Percentage Error} = \frac{\text{Uncertainty}}{\text{Reading}} \times 100\)
Example: If you use a thermometer with an uncertainty of \(\pm 0.5^{\circ}C\) to measure a temperature change of \(10^{\circ}C\):
\(\text{Percentage Error} = \frac{0.5}{10} \times 100 = 5\%\)
Quick Tip: If you are measuring a change in something (like using a burette or a ruler where you take a start reading and an end reading), the uncertainty is usually doubled because you have a margin of error at both the start and the end!
4. Limitations in Experimental Procedures
A limitation is a weakness in the method that might affect the validity or reliability of your results. Don't confuse these with "human errors" (like spilling a solution)—limitations are inherent to the way the experiment was designed.
Common biological limitations include:
• Subjective End-points: In an enzyme experiment (PAG4), deciding exactly when a color changes can be difficult and varies from person to person.
• Fluctuating Variables: Room temperature might change during the day, affecting the rate of reaction.
• Sample Size: Investigating only three plants might not be enough to represent the whole population (linked to Sampling Techniques in PAG3).
• Heat Loss: In experiments involving heat, energy might escape to the surroundings rather than staying in the reaction mixture.
5. Refining Experimental Design
Once you have identified the limitations, you must suggest refinements (improvements). This is a favorite topic for exam questions!
How to suggest refinements:
• To improve accuracy: Use more sophisticated equipment. Instead of judging a color change by eye, use a colorimeter (PAG5) to get a quantitative (numerical) value.
• To improve precision: Use equipment with smaller scale divisions (e.g., a micrometer instead of a ruler).
• To reduce uncertainty: Increase the quantity being measured. Measuring \(50cm^3\) of gas has a lower percentage error than measuring \(5cm^3\) using the same syringe.
• To improve control: Use a thermostatically controlled water bath to keep the temperature constant rather than a beaker of warm water.
Key Takeaway: Refining your experiment is about reducing the impact of variables you aren't testing and making your measurements as "true" as possible.
6. Drawing Valid Conclusions
A conclusion is only valid if it is supported by the data and if the experiment was a "fair test."
When evaluating a conclusion, ask yourself:
1. Does the data actually show a trend?
2. Were all the control variables kept the same?
3. Was the range of the independent variable wide enough? (e.g., testing enzymes at \(10^{\circ}C\), \(20^{\circ}C\), and \(30^{\circ}C\) doesn't tell you what happens at \(40^{\circ}C\)).
4. Is there a statistical test (like Chi-squared or T-test) that proves the results aren't just down to chance?
Note: For more details on calculating means and plotting graphs, refer to the chapter on "Analysis of experimental results."
Quick Review: The Evaluation Checklist
When you are asked to evaluate a procedure in an exam, run through this mental checklist:
• Are there any anomalies?
• What was the percentage error of the equipment?
• Were there any uncontrolled variables?
• Was the measurement subjective or objective?
• How could I refine the method to make it more accurate or precise?