Welcome to Analysing Experimental Data and Drawing Conclusions

Welcome! In science, carrying out an experiment is only half the adventure. The real magic happens when you look at your results, spot patterns, and work out what they actually mean. This chapter is part of Unit 4: Practical Skills for your CCEA GCSE Single Award Science course. The skills you learn here will help you earn marks in both Booklet A (your practical tasks) and Booklet B (your written practical exam paper).

Don't worry if working with data and graphs seems tricky at first. We will break down every skill into simple, step-by-step pieces so that you feel confident and ready to tackle any exam question.


1. Experimental Variables and Types of Data

Before you collect or analyse results, you need to understand the variables involved in any scientific test.

The Three Key Variables

1. Independent Variable: This is the factor that you deliberately change or manipulate in the experiment.
Memory Trick: I change the Independent variable!

2. Dependent Variable: This is the factor that you measure to see the effect of changing the independent variable.
Memory Trick: The Data you collect is the Dependent variable!

3. Controlled Variables: These are all the other factors that you must keep strictly the same throughout the experiment. Keeping these constant ensures a fair test and makes your comparison valid.

Types of Data: Continuous vs. Categorical

The type of data you collect determines how you present it:

Continuous Data: Numerical measurements that can take any value within a given range (for example: time in seconds, temperature in \(^\circ\text{C}\), mass in grams, or volume in \(\text{cm}^3\)).
How to display it: Always plot continuous data using a line graph or a scatter graph with a line of best fit.

Discontinuous / Categorical Data: Data that falls into distinct groups, categories, or descriptive words (for example: eye colour, blood type, type of metal, or shoe size).
How to display it: Always plot categorical data using a bar chart with clear spaces between the bars.

Key Takeaway: Keep control variables constant for a fair test. Put continuous data on a line graph and categorical data on a bar chart with spaces.


2. Designing Tables, Processing Data, and Calculating Means

When you record your data in an exam table, CCEA examiners look for specific rules and conventions.

Table Design Rules

Column Positions: Always put the independent variable in the very first (left-hand) column. Place your dependent variable (including repeat trials and the mean) in the subsequent columns to the right.

Column Headers: Every column heading must contain both the quantity name and the correct unit. You can separate them with a solidus or brackets (for example: \(\text{Time } / \text{ s}\) or \(\text{Temperature } (^\circ\text{C})\)).

Clean Data Cells: Never write units inside the data cells themselves. Only write numbers in the body of the table because the unit is already stated at the top of the column.

How to Calculate a Mean (Average)

Repeating measurements helps make your data more reliable. To calculate the mean value:

\(\text{Mean} = \frac{\text{Sum of concordant / valid repeat trials}}{\text{Number of valid repeat trials}}\)

Dealing with Anomalous Results (Outliers)

An anomaly is a rogue result that does not fit the general pattern of the other repeats.

Important Exam Rule: You must identify and exclude (discard) anomalous results before you calculate the mean! Do not add an anomaly into your total.

Worked Example:
A student records three time trials for a reaction: Trial 1 = \(24\text{ s}\), Trial 2 = \(25\text{ s}\), Trial 3 = \(41\text{ s}\).
Trial 3 (\(41\text{ s}\)) is clearly an anomaly. Leave it out!
\(\text{Mean} = \frac{24 + 25}{2} = \frac{49}{2} = 24.5\text{ s}\)

Significant Figures and Decimal Places

Always record your calculated mean to the same number of decimal places or significant figures as the original raw measurements.

Key Takeaway: Headers must have units; table cells should only contain numbers. Always spot and remove anomalies before calculating your mean average!


3. Graph Construction Standards (CCEA Marking Conventions)

Drawing graphs accurately is worth many marks in your Unit 4 exam papers. Follow this step-by-step checklist to get full marks every time.

Step 1: Choose Your Axes

Place the independent variable on the horizontal \(x\)-axis (bottom).
Place the dependent variable on the vertical \(y\)-axis (side).

Step 2: Add Labels and Units

Copy the full column heading from your table directly onto the axis, including the unit (for example: \(Extension\text{ }/\text{ cm}\) or \(Current\text{ }/\text{ A}\)).

Step 3: Choose a Sensible, Linear Scale

Your scale must go up in regular, easy-to-use intervals such as \(1\text{s}\), \(2\text{s}\), \(5\text{s}\), or \(10\text{s}\). Avoid awkward multiples like \(3\text{s}\) or \(7\text{s}\).
The 50% Rule: Your plotted points must fill at least half (50%) of the available graph grid in both the horizontal and vertical directions.

Step 4: Plot Points Accurately

Mark each point precisely with a small, sharp 'x' or a neat encircled dot (\(\odot\)). Large blobs will lose marks.

Step 5: Draw the Line of Best Fit

Use a ruler for a straight-line trend, or draw a single, smooth, continuous curve for a non-linear trend.
Ensure an even balance of points on either side of the line.
Ignore any obvious anomalous points when deciding where to place your line.
Common Pitfall: Never connect your points dot-to-dot (tramlining) unless specifically instructed to do so!

Key Takeaway: Labels need units, scales must be regular and cover over half the grid, and lines of best fit must be smooth or ruled straight—never dot-to-dot.


Once your graph is drawn, you need to describe what it tells you and calculate key values from it.

Types of Relationships

Directly Proportional: A straight line that passes directly through the origin \((0,0)\). As the independent variable doubles, the dependent variable also doubles (\(y = kx\)).

Linear / Positive Correlation: A straight line where an increase in \(x\) causes a steady increase in \(y\). (Note: if it does not pass through \((0,0)\), it is linear, but not directly proportional).

Negative Correlation / Inversely Related: As \(x\) increases, \(y\) decreases.

Plateau / Levelling Off: The graph flattens out horizontally. This happens when a limiting factor is reached (for example, when a chemical reaction has fully finished, or an enzyme has reached maximum working speed).

Describing Non-Linear Trends Fully

If a graph is curved or changes direction, do not just write "it goes up". Describe each distinct section with values from the axes!
Example: "The mass of gas increases rapidly from \(0\) to \(4\text{ minutes}\), increases more slowly between \(4\) and \(8\text{ minutes}\), and levels off completely after \(8\text{ minutes}\)."

Calculating the Gradient of a Straight Line

The gradient measures the steepness of a straight line of best fit:

\(\text{Gradient } (m) = \frac{\Delta y}{\Delta x} = \frac{y_2 - y_1}{x_2 - x_1}\)

How to calculate it correctly:
1. Draw a large right-angled triangle on your line of best fit that covers more than half (50%) of the line.
2. Read the coordinates \((x_1, y_1)\) and \((x_2, y_2)\) from the points on your line (not from raw data points).
3. Divide the change in \(y\) (\(\Delta y\)) by the change in \(x\) (\(\Delta x\)).

Key Takeaway: A directly proportional graph is a straight line through \((0,0)\). Always use a large triangle on the line of best fit to calculate gradient.


5. Scientific Evaluation: The Key Terminology

In Unit 4 Booklet B, examiners often ask you to evaluate an experiment. To get full marks, you must use the correct scientific terms instead of vague everyday words.

The Evaluation Terms Explained

Repeatability: The closeness of agreement when the same experimenter repeats the investigation using the same method and equipment in the same lab.

Reproducibility: The closeness of agreement when a different experimenter repeats the investigation, or when it is carried out using different equipment or methods.

Reliability: How trustworthy your results are. You improve reliability by repeating the experiment, identifying and excluding anomalies, and calculating a mean.

Accuracy: How close a measured value or calculated mean is to the true or accepted value.

Precision: How close repeated measurements are to each other. Precision also relates to the resolution (smallest division) of your measuring instruments (for example, a burette measuring to \(0.1\text{ cm}^3\) is more precise than a measuring cylinder measuring to \(1\text{ cm}^3\)).

Validity: Whether the experiment actually investigates the question it was designed to answer. An experiment is valid if it is a fair test with all extraneous control variables kept constant.

Common Pitfalls to Avoid in the Exam

Do not use the words "reliable", "accurate", and "fair" as if they mean the same thing!
Saying "we repeat it to make it accurate" is incorrect. Repeating an experiment improves reliability, not accuracy.
Saying "it is reliable because we kept the volume constant" is incorrect. Controlling variables ensures validity (a fair test).

Key Takeaway: Repeating improves reliability. Controlling variables ensures validity. Better measuring tools increase precision.


Quick Review: Unit 4 Practical Skills Checklist

Before you sit your Booklet A or Booklet B exam, check that you can:

Identify independent, dependent, and controlled variables.
Set up a results table with proper headers (\(\text{Quantity } / \text{ unit}\)) and no units inside data cells.
Exclude anomalous results before calculating the mean.
Plot data on a linear scale covering over 50% of the grid.
Draw a smooth curve or ruled straight line of best fit (no dot-to-dot).
Use a large triangle on the line to calculate the gradient (\(\frac{\Delta y}{\Delta x}\)).
Correctly use the terms repeatable, reproducible, reliable, accurate, precise, and valid.