Welcome to Scientific Evaluation
Welcome to one of the most rewarding parts of your CCEA A2 Life and Health Sciences journey! You have designed your experiment, safely carried out your practical work, and collected your raw data. Now comes the crucial step: Scientific Evaluation.
In your Unit A2 1 Portfolio, scientific evaluation is where you step back and look critically at your entire investigation. It is not just about saying whether your experiment "worked." Instead, it is about asking tough, thoughtful questions: How reliable are these results? How much error is hidden in the equipment? Did the method truly test the hypothesis? How do my findings compare with published scientific research?
Don't worry if this sounds intimidating at first! Evaluating an experiment is a step-by-step skill that anyone can master. Let's break it down together.
---Section 1: The Core Language of Scientific Quality
To write a high-scoring evaluation, you need to use precise scientific terms correctly. Examiners often spot students mixing up these key definitions, so getting them clear right now will instantly boost your grade.
1. Accuracy vs. Precision
These two words mean very different things in scientific analysis:
• Accuracy: How close a measured experimental value is to the true or accepted theoretical value.
• Precision: How close repeated measurements are to each other under identical conditions (the degree of agreement or clustering between repeats).
Analogy — The Dartboard:
Imagine throwing three darts at a bullseye. If all three hit the outer ring close together in the top-right corner, your throws are precise (tightly grouped) but inaccurate (far from the bullseye). If your darts land scattered around the bullseye and their average position is right in the center, they are accurate on average, but have low precision. If all three land right inside the tiny bullseye, they are both accurate and precise!
2. Repeatability vs. Reproducibility
Both terms describe how reliable your results are when tests are repeated, but the conditions are different:
• Repeatability: The closeness of agreement when the same investigator repeats the test using the same apparatus, reagents, and laboratory over a short time frame.
• Reproducibility: The closeness of agreement when the test is carried out by different experimenters, in different laboratories, or using different apparatus.
3. Scientific Validity
• Validity: The extent to which your experiment actually investigates the specific aim or hypothesis you set out to test, without bias or the influence of uncontrolled confounding variables.
If you set out to investigate how enzyme concentration affects reaction rate, but you allow the temperature of the room to rise by \(8^\circ\text{C}\) during the experiment, your results are invalid because temperature changes also alter enzyme kinetics.
Section Key Takeaway: Tight clustering means precision; closeness to the true value means accuracy; consistency by you means repeatability; consistency by others means reproducibility; and a fair test that answers the real aim means validity.
---Section 2: Quantifying Errors and Uncertainties
Every single measurement taken in a science laboratory carries some degree of uncertainty. High-level evaluations do not treat numbers as "perfect"; they calculate and explain those uncertainties.
Systematic Errors vs. Random Errors
Errors in data generally fall into two categories:
1. Systematic Errors:
These are predictable, consistent errors that shift all your data in one direction (always slightly too high, or always slightly too low).
• Examples: A balance that reads \(+0.05\text{ g}\) when empty (a zero error), an uncalibrated colorimeter, or reading a meniscus from above eye level every time (parallax error).
• How to fix: Systematic errors cannot be eliminated by simply repeating the test and calculating a mean. You must recalibrate equipment or correct the technique.
2. Random Errors:
These are unpredictable, temporary fluctuations that cause measurements to vary unpredictably above and below the true value.
• Examples: Slight room temperature drafts, electrical noise in digital sensors, or human reaction time variations when clicking a manual stopwatch.
• How to fix: Random errors are minimized by conducting multiple repeated trials and calculating a mean value.
Absolute Uncertainty
The absolute uncertainty (\(\Delta x\)) is the numerical margin of doubt associated with any measuring instrument:
• Analogue instruments (e.g., rulers, liquid-in-glass thermometers, burettes):
\(\text{Absolute Uncertainty} = \pm(\text{half the smallest scale division})\)
Example: A thermometer with marks every \(1.0^\circ\text{C}\) has an absolute uncertainty of \(\pm 0.5^\circ\text{C}\).
• Digital instruments (e.g., digital balances, digital pH probes):
\(\text{Absolute Uncertainty} = \pm(\text{the least significant digit})\)
Example: A two-decimal-place balance reading \(3.42\text{ g}\) has an uncertainty of \(\pm 0.01\text{ g}\).
Calculating Percentage Uncertainty
To understand which instrument caused the greatest limitation in your method, calculate the percentage uncertainty for each measurement:
\(\text{Percentage Uncertainty} = \left(\frac{\text{Absolute Uncertainty}}{\text{Measured Value}}\right) \times 100\%\)
Worked Example:
A student measures two different liquid volumes using a measuring cylinder with an absolute uncertainty of \(\pm 0.5\text{ cm}^3\):
• Volume A = \(5.0\text{ cm}^3\)
\(\text{Percentage Uncertainty} = \left(\frac{0.5}{5.0}\right) \times 100\% = 10.0\%\)
• Volume B = \(50.0\text{ cm}^3\)
\(\text{Percentage Uncertainty} = \left(\frac{0.5}{50.0}\right) \times 100\% = 1.0\%\)
Notice: Measuring smaller quantities with the same equipment leads to a much higher percentage uncertainty! This is an excellent point to raise in your portfolio evaluation.
Handling Anomalies (Outliers)
An anomaly is a data point that falls significantly outside the expected trend or range of repeated values.
• Identify: Spot the value that does not match the concordance of other repeats (e.g., values of \(12.1\text{ s}\), \(12.3\text{ s}\), and \(18.6\text{ s}\)).
• Action: Discard the anomalous reading (\(18.6\text{ s}\)) when calculating the mean.
• Justify: In your written evaluation, explain why it was excluded (e.g., delayed timing trigger or contamination) and, where possible, repeat the reading.
Section Key Takeaway: Systematic errors bias data in one direction and require recalibration; random errors cause scatter and are reduced by taking averages. Always find which measurement created the highest percentage uncertainty.
---Section 3: Structuring Your A2 1 Portfolio Evaluation
In Unit A2 1, your scientific evaluation report must be detailed, structured, and evidence-based. To achieve the highest marks, address each of the four core pillars below:
Pillar 1: Critical Review of Data Quality
• Examine the scatter of your repeat readings around the line of best fit or mean value.
• Discuss whether error bars overlap on graphs (overlapping error bars suggest differences between conditions may not be significant).
• State clearly whether experimental uncertainties were small enough to leave your final conclusions valid.
Pillar 2: Comparison with Secondary / Published Literature
• Connect your primary experimental results back to the secondary scientific sources you reviewed in the introductory literature review of your portfolio.
• Do your rate constants, optimum pH values, or energy values match published scientific consensus?
• If your results differ from published baseline values, discuss what methodological differences or systematic errors could explain the discrepancy.
Pillar 3: Specific Procedural Modifications
Examiners award high marks for specific, realistic improvements rather than vague statements.
• Weak evaluation: "I would be more careful next time and use better equipment."
• High-scoring evaluation: "To reduce heat loss to the surroundings during enthalpy determinations, the simple polystyrene cup should be upgraded to a vacuum-jacketed calorimeter, and the manual thermometer (uncertainty \(\pm 0.5^\circ\text{C}\)) replaced with a calibrated temperature data logger (uncertainty \(\pm 0.05^\circ\text{C}\))."
• Apparatus upgrade ideas to consider:
— Replacing manual stopclocks with automated light gates or colorimeter data loggers to remove human reaction time error.
— Using micropipettes or Grade A volumetric pipettes instead of standard measuring cylinders to minimize percentage volume error.
— Using a thermostatically controlled water bath with a PID controller to maintain a constant temperature for enzyme assays.
Pillar 4: Suggestions for Further Work
Extend the investigation logically to test the scientific theory further:
• Suggest testing intermediate values around a key point (e.g., testing pH values at \(0.2\) intervals between \(\text{pH } 6.0\) and \(\text{pH } 8.0\) to locate the precise optimum).
• Suggest testing a wider range of concentrations or exploring how a different cofactor/inhibitor impacts the system.
Section Key Takeaway: Build your evaluation on four pillars: evaluate data scatter, compare with published literature, suggest specific apparatus upgrades, and propose targeted extensions.
---Section 4: Common Pitfalls & How to Avoid Them
Make sure you avoid these common traps when writing your A2 1 evaluation:
• Trap 1: Confusing accuracy with precision.
Fix: Never describe tightly clustered repeat readings as "accurate" unless you know they match the true literature value. Call them precise or repeatable.
• Trap 2: Generic improvement comments.
Fix: Never write "repeat the experiment more times" as your main modification. Identify the exact source of error (e.g., visual judgment of a color change endpoint) and name the specific instrument that fixes it (e.g., a colorimeter set at \(\lambda = 540\text{ nm}\)).
• Trap 3: Averaging in anomalous data.
Fix: Check your raw tables carefully. If a value is an obvious outlier, exclude it from the mean and explain why in your evaluation text.
• Trap 4: Forgetting the literature review.
Fix: Always cite and compare your numerical conclusions with the secondary sources gathered in Section 1 of your portfolio.
Section 5: Final Revision Checklist
Before submitting your Unit A2 1 Scientific Evaluation, check off each of these questions:
• Did I distinguish between accuracy, precision, repeatability, and reproducibility?
• Have I calculated percentage uncertainties for all key measuring instruments?
• Did I identify which measurement contributed the largest percentage error?
• Have I classified my errors correctly as either systematic or random?
• Have I identified and justified the exclusion of any anomalies?
• Have I compared my findings directly against secondary literature values?
• Are my proposed modifications detailed, named, and realistic?