Analysis and Interpretation of Data
Uploaded by
Shania SmithAnalysis and Interpretation of Data
Uploaded by
Shania SmithCIE Biology International A-level
Analysis and Interpretation of Data
Practical Notes
www.pmt.education
Analyse and interpret data to reach conclusions
Processing results
Data can be processed in different ways depending on the aim of the practical. Simple processing
methods include calculating means, changes, and rates.
Means are calculated when the practical is repeated or multiple samples are taken. Taking a mean
value of many data points allows the variability of the results to be assessed. Using means,
standard deviation can also be calculated, which is a measure of the spread of data. Standard
error, which is a measure of the reliability of the mean i.e. how close the sample mean is to the
real mean, can be calculated from standard deviation. The formula for standard error is:
Standard deviation / √n
Standard error can be presented on graphs in error bars, where the range is + and - 2x standard
error. Generally, if the error bars on two values do not overlap, the values are significantly different.
Change may be calculated for change of mass, length or temperature eg. in osmosis-related
practicals. Percentage change may sometimes be required, as this allows comparison when the
starting point is different. The formula of percentage change is:
(final value - initial value) / initial value x 100%
Rates can be directly or inversely proportional to the results collected. If time to complete a
certain reaction is measured, often in the case of enzyme reactions, rate is calculated by 1/time.
However, rate can also be indicated by the release of a product, eg. volume of gas produced in a
fixed time, then rate is directly proportional to the volume and may not require further processing.
For certain practicals (often involving field investigations and genetic crosses), statistical tests will
have to be used to analyse the significance of the results. There are four statistical tests that
students are expected to use:
● Spearman’s rank coefficient
● Pearson’s linear correlation coefficient
● T-test
● Chi-squared test
The first step to performing a statistical test is to write a null hypothesis. A null hypothesis states
that there is no significant correlation or difference between the two data sets students have
collected. Next, whether the data sets are being compared or correlated, and the nature of the
data - whether it is continuous or discontinuous. Select the appropriate statistical test using the
table below.
www.pmt.education
Statistical test Purpose Type of data Degrees of Freedom
T-test Compare, find if Continuous, normally (n1 - 1) + (n2 - 1)
difference is distributed eg. means
significant
Chi-squared Compare observable Frequency, discrete No. of categories - 1
and expected values, and categorical
find if difference is
significant
Spearman’s rank Correlate Ordinal data No. of pairs of data - 2
Pearson’s linear Correlate Continuous, linear, No. of pairs of data - 2
correlation normally distributed
Use the formula provided to calculate a value. Find the critical value in a table provided at p=0.05
and the calculated degrees of freedom, where p=0.05 indicates a 5% probability that the
difference or relationship in the data is due to chance alone. If the calculated value is larger than
the critical value, the null hypothesis is rejected, and there is a smaller than 5% probability that
the difference or relationship is due to chance alone.
Another statistical test called Simpson’s Index of Biodiversity is specific to field investigations on
biodiversity. It gives a value between 0 and 1, where values closer to 1 have higher biodiversity.
The formula may or may not be provided in the exam, hence it should be memorised.
D = 1-(Σ (n/N)2)
Patterns and trends
To describe patterns and trends, first describe the overall trend, quoting data to support your
observations. Afterwards, move onto describing any particular features of the graph, eg. peaks
and troughs, or data points that do not fit the overall trend.
Students may be asked to identify anomalies. Anomalies are data that do not fit the trend.
When calculating a mean, the anomalies in the data set should be excluded.
Finding unknowns
Certain practicals require students to estimate an unknown value, eg. concentration of an
unknown solution. This often requires students to produce a set of standards to compare against,
which may be colour standards, time taken for the reaction of a known solution, or a graph plotted
using standard solutions.
To increase the accuracy of the estimate, the number of different standards used should be
increased at smaller intervals around the estimate.
www.pmt.education
