Anomalous Results, Conclusions & Evaluation (Cambridge (CIE) IGCSE Biology): Revision Note

What is an anomalous result?

• An anomalous result is a value that appears to be outside the general pattern of the results

  • It is usually an extreme value that is either very high or very low compared with the others

• Describe an anomalous result by saying the way in which it does not fit the pattern

  • Saying that it is "wrong", "a mistake" or "an error" is not enough to describe an anomalous result

• When you have anomalous results you should:

  • leave them out when you calculate a mean

  • ignore them when you draw a line of best fit

    • If you ignore a point on a graph, show this clearly, e.g. by drawing a circle around it

• Repeating a reading does not stop anomalous results happening, but it helps you to spot anomalies

• Examiner reports show that students lose marks by not showing which point on a graph they have treated as anomalous

How to write a conclusion

• A conclusion says what your results show, and should relate to the aim of the investigation

  • Before you write it, reread the aim in the question

• In your conclusion you should:

  • link the independent variable to the dependent variable

    • E.g. "as the temperature increased, the rate of water loss also increased"

  • support your conclusion with evidence from your results by stating values from the table or graph

• Only conclude what your results actually show, e.g.:

  • be careful not to state a causal link when one hasn't been proven; sometimes a conclusion may just state that two variables are linked, but cannot state that a change in one variable caused a change in another (i.e. correlation, not causation)

  • don't apply the results of a conclusion too broadly; results gained from studying woodlice may not be applicable to humans, for example

• Examiner reports show that students lose marks by writing conclusions that are not backed up by evidence from their results

Sources of error & improvements

• Error, in the context of scientific investigation, refers to specific errors in measurement, not just general mistakes like spilling a solution, or adding the wrong volume

• A measurement error is the difference between a measured value and the true value; measurement errors can be:

  • random errors: measurement errors that vary unpredictably, e.g. when starting and stopping a stopwatch by hand

    • You can reduce the effect of random errors by repeating readings and calculating a mean

  • systematic errors: these make every reading too high or too low by the same amount, e.g. a balance that does not read zero before you start

    • Repeating readings does not remove a systematic error, so you should calibrate the equipment (i.e. ensure that zero equals zero) before using

• Parallax error happens when you read a scale from an angle, so be sure to read scales at eye level, e.g. when reading a volume of liquid in a measuring cylinder, or reading a thermometer

• For every source of error you identify, suggest a specific, matching improvement that would reduce it

• Only suggest improving something that the given method does not already control