Introduction to Evaluating Procedures and Results

Welcome! In Chemistry, doing an experiment is only half the job. The other half is asking: "How good was my experiment?" and "Can I trust my results?"

Evaluation is about looking at your method and your data to find weaknesses. It’s not about admitting you did a "bad" job; it’s about being a great scientist by identifying where errors might have crept in. In your OxfordAQA International AS exams, you will often be asked to look at a set of results or a procedure and suggest why they might be slightly off and how to fix them.

1. Accuracy, Precision, and Errors

Before we evaluate, we need to know what we are looking for. These three terms are the foundation of any evaluation:

  • Accuracy: How close your result is to the "true" or accepted value. If you calculate the boiling point of water as \(100 ^\circ\text{C}\), you are very accurate!
  • Precision: How close your repeated measurements are to each other. If you do a titration three times and get \(24.50 \text{ cm}^3\), \(24.55 \text{ cm}^3\), and \(24.50 \text{ cm}^3\), your results are precise.
  • Errors: The difference between your measurement and the true value.

Random vs. Systematic Errors

There are two main "troublemakers" in any experiment:

Random Errors: These are unpredictable and vary each time you repeat the experiment. Example: A slight breeze affecting a balance, or your eyes being at a slightly different angle when reading a burette. Solution: Repeat the experiment and calculate a mean (average).

Systematic Errors: These are "built-in" errors that happen every single time you perform the procedure. Example: A thermometer that always reads \(1 ^\circ\text{C}\) too high, or heat escaping from a calorimeter. Solution: These are harder to fix by averaging. You usually need to change the equipment or the method.

Quick Review: If your results are all very close together but far from the real answer, you probably have a systematic error. If your results are scattered all over the place, you have random error issues!

2. Evaluating Equipment: Uncertainties

Every piece of measuring equipment has a limit to how precise it can be. This is called the uncertainty. For example, a ruler might be accurate to \(\pm 1 \text{ mm}\).

Calculating Percentage Uncertainty

To evaluate if your equipment was appropriate, we calculate the percentage uncertainty using this formula:

\( \text{Percentage Uncertainty} = \frac{\text{Uncertainty of the equipment}}{\text{Reading taken}} \times 100 \)

The "Double Reading" Rule: For equipment where you take two readings to get one value (like a burette or a thermometer), you must multiply the uncertainty by 2.

Example: If a burette has an uncertainty of \(\pm 0.05 \text{ cm}^3\) per reading, and you calculate a titre of \(25.00 \text{ cm}^3\):

\( \text{Uncertainty} = 0.05 \times 2 = 0.10 \text{ cm}^3 \)

\( \% \text{ Uncertainty} = \frac{0.10}{25.00} \times 100 = 0.4\% \)

How to improve: To reduce percentage uncertainty, you can either use a more precise piece of equipment (e.g., a volumetric pipette instead of a measuring cylinder) or use larger volumes/masses so the "error" is a smaller fraction of the total.

3. Evaluating Procedures in Required Practicals

Let’s look at common procedural flaws in the AS required practicals. Identifying these is a major part of the AO4 assessment objective.

Practical 1: Titrations

  • The Problem: Overshooting the end-point (the solution changes color too deeply). Evaluation: This leads to a larger-than-expected titre volume. Fix: Add the solution drop-wise near the end-point.
  • The Problem: Leaving the funnel in the burette. Evaluation: Drops of liquid could fall in during the titration, changing the volume. Fix: Remove the funnel before starting.

Practical 2: Measuring Enthalpy Changes

This is the most common "evaluation" topic in exams because the results are rarely perfect!

  • The Problem: Heat Loss to the surroundings. Evaluation: The measured temperature change (\( \Delta T \)) will be lower than it should be, making the calculated \( \Delta H \) less exothermic. Fix: Use a lid, insulate the beaker (polystyrene cup), or use a bomb calorimeter for combustion.
  • The Problem: Incomplete Combustion (in spirit burner experiments). Evaluation: Less energy is released than expected. Fix: Ensure a steady supply of oxygen.

Practical 4: Distillation

  • The Problem: Loss of volatile product (it escapes as gas). Evaluation: This lowers the percentage yield. Fix: Ensure the condenser is cold enough and all joints are sealed with grease or PTFE tape.

4. Evaluating Data: Anomalies and Averages

When you look at a set of results, look for the anomaly (the "weird" one). An anomaly is a result that does not fit the pattern of the others.

What to do with anomalies:

  1. Identify it (e.g., "The titre of \(26.50 \text{ cm}^3\) is an anomaly because the others are \(24.10\) and \(24.15\)").
  2. Exclude it from your mean calculation.
  3. Investigate why it happened (did you overshoot the end-point? Was the temperature different?).

Don’t worry if this seems tricky at first! Just remember: a mean is only reliable if you use concordant results (results within \(0.10 \text{ cm}^3\) of each other in titrations).

5. Comparing Results: Percentage Yield and Atom Economy

Evaluating a process also involves looking at how "efficient" it was. (See the chapter on Amount of Substance for calculation steps).

  • Percentage Yield: Tells you how much product you actually made compared to the maximum possible. Low yield suggests product was lost during transfer, filtration, or evaporation.
  • Atom Economy: Tells you how much of the starting mass ended up in the desired product. A low atom economy means you are making a lot of waste, which is bad for the environment and for profit!
Key Takeaway: Evaluation isn't just about finding mistakes; it's about suggesting improvements. If the heat loss is high, suggest insulation. If the uncertainty is high, suggest a larger sample size or more precise equipment.

Common Exam Mistakes to Avoid

  • Vague language: Don't just say "human error." This gets zero marks! Be specific: "Delayed reaction time when stopping the stopwatch" or "Difficulty identifying the exact color change."
  • Forgetting the "2x" rule: When calculating uncertainty for a change in temperature or volume, always remember that you took two readings.
  • Including anomalies: If you include a wildly different result in your average, your final answer will be skewed. Always check for concordancy first.

For more details on how to handle the data itself, see the chapters on "Data handling, graphs and uncertainties" and "Experiment planning and control of variables."