Experimental Design

In A-level Biology, good experimental design begins with planning, which helps ensure that an investigation is valid and yields meaningful results. Scientists should conduct preliminary research to identify effective experimental techniques, select suitable apparatus, and determine how to control variables that could affect the outcome. 

For instance, before investigating the effect of pH on enzyme activity, it is essential to understand how enzyme function is influenced by pH, temperature, and substrate concentration, and how to measure activity accurately. 
Preliminary studies can also be performed to identify unknown confounding variables, refine the methodology, and decide on appropriate volumes or concentrations needed during the investigation. 
Without proper experimental design, including preliminary work, the investigation risks being invalid due to uncontrolled factors.

Identification of Variables

Every experiment involves variables, which are factors that can influence the outcome. 

• The independent variable is the one that the experimenter deliberately changes, such as temperature or pH. 

• The dependent variable is the one that is measured to determine the effect of the independent variable, such as rate of reaction or volume of gas produced. 

• Control variables are all other factors that could affect the outcome and must be kept constant to ensure valid results. 

For example, when testing the effect of temperature on enzyme activity, it is crucial to keep pH, substrate concentration, and enzyme concentration the same in all tests. Failure to control these variables may introduce errors, reducing the validity of the data.

Control experiments are essential components of scientific investigations because they provide a baseline or reference point that allows scientists to determine whether the observed effects are truly due to the independent variable being tested. The results of the experiment where the independent variable is being changed will be compared to the results of the control experiment.

Evaluating Experimental Methods

Evaluating an experiment involves critically assessing the method to identify weaknesses or limitations that could affect the accuracy and reliability of results. A method limitation refers to any aspect of the procedure that could produce consistent errors, such as poor measurement techniques or unsuitable equipment. 

Scientists must distinguish between random errors, which vary unpredictably from one measurement to the next (e.g. due to environmental fluctuations), and systematic errors, which consistently affect measurements in the same way (e.g. a miscalibrated balance). 

Evaluations should also consider accuracy (how close results are to the true value), precision (how consistent repeat measurements are), reliability (whether the results can be reproduced), and validity (whether the method measures what it is intended to, have other variables been controlled and has a control experiment been used). 

Using Units and Significant Figures

Correct use of scientific units is vital when recording and presenting data. Units must be placed in column headers of tables and labelled clearly on graph axes.

Measurements should be recorded and reported using the appropriate number of significant figures, which are the digits that convey meaningful precision in a value. Rules for significant figures include: all non-zero digits are significant, zeros between non-zero digits are significant, and trailing zeros are significant only if a decimal point is present.

For example, the number 29.009 has five significant figures, whereas 0.00079 has two. Rounding should be based on the number of significant figures present in the raw data or specified in the question.

Presenting and Processing Experimental Data

Quantitative data must be tabulated with clear headings, units in the headers (not in data cells), and the independent variable in the first column. When processing quantitative data, scientists may calculate averages (means), rates, or other derived values. 

Where applicable, standard deviation can be calculated to represent variability in the data. 

Graphs should be used to visualise trends, with line graphs or scatter plots for continuous data and bar charts for categorical or discrete data. Axes must be correctly labelled with units, and the scale should allow all data to fit neatly within the graph area.

Precision, Accuracy and Uncertainty

Precision refers to how close repeat measurements are to each other, and can be improved by taking more repeats, minimising the effect of random errors.

Accuracy is how close a result is to the true value, which can be affected by systematic errors such as calibration faults. Accuracy can be improved by using apparatus with greater resolution (smaller intervals) or changing the method to reduce subjectivity, e.g using a colorimeter instead of looking for a colour change.

Uncertainty is the range within which the true value is expected to lie, based on the resolution of the instrument. For example, a measuring cylinder marked in 1 cm³ increments has a greater uncertainty than one marked in 0.1 cm³. 

The percentage error quantifies uncertainty and is calculated as:

Percentage error=(uncertainty / measured value) ×100

Choosing equipment with the appropriate resolution for the expected measurement range helps minimise error and improve data reliability.

Quantitative vs Qualitative Results

Biology experiments produce two types of data: quantitative, which involves numerical measurements such as temperature or mass, and qualitative, which involves descriptions such as colour or texture. 

• Quantitative data must be consistently recorded to the correct number of decimal places. For example, if measuring time in seconds using a stopwatch with millisecond resolution, values should be recorded to three decimal places. 

• Qualitative observations should be as objective as possible to reduce subjectivity. 

Both types of data can be useful in drawing conclusions, depending on the nature of the experiment.

Mathematical Analysis of Results

After collecting data, scientists often calculate mean values to summarise their findings and standard deviations to assess variability. 

A small standard deviation suggests that the values are tightly grouped around the mean and the data is precise. A large standard deviation indicates greater variability. 

When comparing groups, the overlap (or lack thereof) in standard deviation ranges can help determine whether differences are likely to be significant. If the standard deviations do not overlap, it may suggest a statistically significant difference between the groups being compared.

Graph Interpretation and Tangents

When plotting data on graphs, it is essential to choose the correct type of graph based on the data type. 

• Scatter graphs and line graphs are used for continuous data

• Bar charts are used for discrete data. 

The independent variable goes on the x-axis and the dependent variable on the y-axis. 

To find the initial rate of a reaction from a curved graph, draw a tangent at the starting point of the curve and calculate the gradient using the formula:

Rate=Δy / Δx

Drawing valid conclusions involves using processed data and scientific reasoning to answer the original hypothesis. 

A conclusion should be based on the evidence presented and must consider whether the experimental results are consistent with what was expected. 

It is important to evaluate the precision and accuracy of data, and take into account sources of error. 

The scope of a conclusion refers to the extent to which findings from an experiment can be confidently applied, generalised, or interpreted based on the evidence collected. A conclusion must always remain within the boundaries of the investigation and should only address what was directly tested and measured. For example if an drug has been tested on mice it would not be suitable to conclude its effect on humans.