For practical, lab, and experiment courses, Eclat Institute may issue an internal Certificate of Completion/Attendance based on participation and internal assessment.
This is an internal centre-issued certificate, not an MOE/SEAB qualification or accreditation.
Recognition (if any) is determined by the receiving school, institution, or employer.
For SEAB private candidates taking science practical papers, SEAB states you should either have taken the subject before or complete a practical course before the practical exam date.
Planning a revision session? Use our study places near me map to find libraries, community study rooms, and late-night spots.
TL;DR An anomalous result is a data point that deviates clearly from the pattern shown by the rest of your data. You must circle it on the graph and flag it in the table. You must NOT delete it from your write-up. You may exclude it from the best-fit line or mean calculation only if you state a specific reason. Every subject - Biology, Chemistry, Physics - expects you to handle anomalies this way, and every mark scheme rewards students who identify them and explain them. This guide shows you exactly how.
Why anomalous results matter in practicals
Anomalous results appear in every O-Level science paper. The ACE (Analysis, Conclusions and Evaluation) strand in Paper 3 explicitly asks you to identify anomalies and comment on them. Students who spot and address an anomaly earn the evaluation mark. Students who quietly skip over it - or worse, who delete the point from their table - lose marks and sometimes trigger examiner concern about data integrity.
The good news: the skill is entirely learnable. There is a consistent three-step rule across all three science subjects, and the phrasing for the ACE sentence follows the same logic whether you are measuring enzyme reaction rates in biology, calculating concordant titres in chemistry, or recording resistance readings in physics.
1 | What counts as anomalous
A result is anomalous when it does not fit the pattern established by the rest of your data. There are three practical situations where this shows up.
Outlier on a graph. When you plot your data points, one point lies noticeably far from the best-fit line or curve while all the others cluster around it. "Noticeably" means the point cannot be explained by normal measurement scatter - it stands out visually and you would need to force the best-fit line well away from the other points to include it.
Non-concordant titre in chemistry. In a titration, you aim for at least two concordant titres: readings that agree within 0.10 cm³ of each other. A titre of 23.15, 23.20, and 25.80 gives you three values. The 25.80 differs from the others by more than 2.00 cm³ - this is not normal variation and the 25.80 is anomalous. It must not be included in the mean.
Repeats that disagree beyond acceptable range. In enzyme rate or rate-of-reaction experiments where you take two or three repeat readings, an acceptable spread is typically within 5-10% of the mean. If readings are 12, 13, and 21 cm³ of gas collected, the 21 is anomalous. Including it would inflate the mean and misrepresent the trend.
The repeat count connection
Repeats exist partly to help you spot anomalies. With only one reading per condition, you cannot tell whether a surprising value is anomalous or simply what the experiment produces. With three repeats, a value that appears once while the other two agree gives you evidence to call it anomalous. This is why repeat counts matter for uncertainty: more repeats give you better evidence for your decision.
When you plot your data and one point sits well away from the trend, use a pencil circle around that point. Do not erase the point. The circle signals to the examiner that you have noticed the deviation - this itself earns you a mark in the ACE strand.
The circle goes around the plotted point, not around part of the best-fit line. A neat, small circle about 5-6 mm in diameter is appropriate.
Step 2: Flag in the table
In your results table, place an asterisk × or note next to the anomalous value. Some students write "anomalous" in a narrow column to the right, or add a footnote below the table. What matters is that the value is still recorded in full - you must never erase or cross out an anomalous reading. It happened; it is part of your data record.
Step 3: Note in the write-up
In the ACE section of your write-up, write one sentence that names the anomaly, states its value, and identifies a plausible cause. This is the sentence that earns the evaluation mark.
3 | The include/exclude decision
This is the question students get most confused about: should you include the anomalous point in your best-fit line or mean calculation?
The default answer is: exclude it, but only with justification.
You may exclude an anomalous value from:
The best-fit line on a graph
The mean titre calculation in titration
The mean rate calculation in biology or chemistry
You must NOT:
Delete the point from your data table
Omit it from your graph entirely (it must be plotted)
Average it in without comment (this loses marks because the examiner can see you have not addressed it)
The justification does not need to be long. One sentence is enough: name the value, state it deviates from the pattern, and suggest a specific cause. "Possibly" is acceptable - you are in an exam, not a peer-reviewed paper. What is not acceptable is saying "I excluded this result because it was wrong" without any physical or procedural reason.
When not to exclude
If your anomalous point falls on what looks like a clear curve (for example, an enzyme rate curve that reaches a plateau), reconsider whether the point is truly anomalous or whether your best-fit line was incorrect. Draw the line that fits the majority of points, circle anything that deviates, and comment. Do not redraw the curve to pass through every point just to avoid having an anomaly - that would make a worse graph overall.
4 | Worked ACE sentence examples by subject
Each example below gives you a realistic data scenario, the anomalous point, and a model ACE sentence. Use these as templates and adapt the numbers and causes to your own data.
Biology: enzyme reaction rate
Scenario. An experiment measures the volume of oxygen produced by catalase in hydrogen peroxide at different temperatures. Results for three repeats at 50 °C are 2.1, 2.3, and 6.8 cm³ per minute.
The 6.8 cm³ min⁻¹ value is anomalous. The other two repeats at 50 °C agree closely (2.1 and 2.3 cm³ min⁻¹), so the 6.8 cm³ min⁻¹ reading deviates by more than 4 cm³ min⁻¹ from the pair.
ACE sentence: "The 50 °C reading of 6.8 cm³ min⁻¹ was anomalous (circled on graph) and was excluded from the mean. A possible cause is that the gas syringe plunger may have stuck and then released suddenly, giving an artificially high reading; repeating this condition would confirm whether the pattern holds."
Chemistry: titration concordance
Scenario. A student carries out three titrations to find the volume of sodium hydroxide needed to neutralise a fixed volume of hydrochloric acid. Titre values are 23.15, 23.20, and 25.80 cm³.
The 25.80 cm³ titre differs from the concordant pair (23.15 and 23.20) by more than 2.50 cm³. This is outside the accepted concordance range of 0.10 cm³.
ACE sentence: "The 25.80 cm³ titre is anomalous and was not included in the mean. A likely cause is that the endpoint was overshot - the indicator changed from colourless to pink and then further to dark pink before the student stopped adding alkali. The mean was calculated from the two concordant titres (23.15 and 23.20 cm³) giving 23.18 cm³."
Physics: resistance measurements
Scenario. An experiment records the resistance of a nichrome wire at different lengths. Readings at 20, 40, 60, 80, and 100 cm are: 1.2, 2.3, 3.4, 5.9, and 5.1 Ω.
The 80 cm reading of 5.9 Ω lies above the trend while the 100 cm reading of 5.1 Ω is lower than expected. Both deviate from the linear pattern. The 5.9 Ω reading appears anomalously high relative to the best-fit line through the other four points.
ACE sentence: "The resistance reading of 5.9 Ω at 80 cm was anomalous (circled on graph) and excluded from the best-fit line. A possible cause is a poor contact between the crocodile clip and the wire at that position, introducing additional contact resistance. Repeating the measurement with cleaned clip contacts would verify this."
5 | Subject-by-subject quick reference
Subject
Where anomalies appear
Acceptable range before flagging
Common causes
Biology
Enzyme rate, osmosis mass changes, photosynthesis bubble counts
More than 10-15% from the other repeats
Timing error, leaking gas syringe, temperature fluctuation
Chemistry
Titration titres
Differs by more than 0.10 cm³ from the concordant pair
Endpoint overshoot, air bubble in burette, rinsing error
Physics
Resistance, extension, temperature
Clearly off the best-fit line by more than normal scatter
Poor clip contact, parallax on ruler, zero error not corrected
6 | Common mistakes and how to avoid them
Quietly deleting the point. Some students erase anomalous values from their table or simply do not plot them. Examiners notice when a data point that should appear on the graph is absent. Deleting data is treated as dishonest practice. Keep every value; circle the anomaly.
Averaging all data including anomalies. If you calculate the mean titre using 23.15, 23.20, and 25.80 cm³, you get 24.05 cm³ - over 0.8 cm³ higher than the correct value of 23.18 cm³. This incorrect mean flows through your calculation and gives a wrong final answer. Always identify and exclude anomalies before calculating a mean.
Blaming "human error." "The anomalous result was due to human error" earns zero marks for the evaluation sentence. Human error is not a specific cause. The examiner wants to know what you or the equipment did that introduced the deviation. Replace "human error" with a precise description: "the burette tip was not fully submerged in the conical flask during the run, allowing a small air bubble to form."
Inventing a cause with no physical basis. On the other hand, do not guess wildly. The cause you suggest must be mechanically possible given the apparatus you used. If you are measuring gas volume with a syringe, causes might include leaks, sticking plungers, or timing errors. Causes like "the temperature in the room changed by 20 °C" are implausible and will not be credited.
Forcing the best-fit line through the anomaly. Drawing a curve that passes through every plotted point, including the anomaly, is a graph error (equivalent to dot-to-dot drawing). Draw the best-fit through the majority of points and let the anomaly sit outside the line with a circle around it.
7 | Connecting anomalies to uncertainty
When you have three repeats and one is anomalous, you exclude it and compute the mean from two values. The uncertainty in your mean is therefore based on two readings rather than three. In chemistry, you would quote the concordant pair mean as your titre with the acknowledgment that a non-concordant reading was obtained.
At H2 level, you may be asked to calculate percentage uncertainty explicitly. If your two usable readings are 23.15 and 23.20 cm³, the range is 0.05 cm³. The percentage uncertainty = (range / mean) × 100 = (0.05 / 23.18) × 100 = 0.22%. This is the kind of quantified evaluation that earns ACE marks at both O-Level and H2 level.
