Presenting and Processing A-Level Practical Data

1. Data must preserve meaning

Processing is not decoration: every derived column and graph must connect the raw observations to the physical model.

2. Tables

A clear table places raw readings before derived quantities and gives the unit once in each heading.

• Use consistent decimal places for repeated readings from the same instrument.
• Give calculated values sensible significant figures; do not imply precision absent from the raw data.
• State the transformation in the heading, such as 1/x, T² or ln y.
• Keep units attached to the quantity, not repeated in every data cell.

3. Graph construction

• Put the independent or transformed independent quantity on the horizontal axis unless the required linear form implies otherwise.
• Label each axis with quantity and unit.
• Choose a simple scale that uses a substantial part of the plotting area; the origin need not be included unless it is physically or mathematically relevant.
• Plot points precisely and draw a best-fit line or curve representing the overall trend, not a dot-to-dot path.
• Investigate anomalous points before deciding whether they should influence the fit.

4. Gradient and intercept

Use two well-separated points on the best-fit line, not two raw data points, to calculate a manual gradient: m = (Δ y)/(Δ x).

The gradient unit is the vertical-axis unit divided by the horizontal-axis unit. The intercept has the vertical-axis unit.

If Y = mX + c, identify the experimental graph quantities with Y and X before interpreting m and c. A numerical gradient without its physical meaning is incomplete.

5. Linearisation

Transform uncertainties and units consistently. A straight-looking plot alone does not validate a model if the axes were chosen after seeing the data without a theoretical reason.

5A. Spreadsheet workflow

1. Enter raw data without overwriting the original observations.
2. Put each derived formula in a new column and fill it down.
3. Check one row manually, including units and powers of ten.
4. Use an x–y scatter chart, not a category line chart.
5. Apply an appropriate fit and record its equation without reporting unjustified digits.
6. Translate the fitted gradient and intercept back into the required physical constants.

5B. Common mistakes

• Putting units inside the data cells or omitting them from headings.
• Forcing a best-fit line through the origin without a model-based reason.
• Using adjacent plotted points for the gradient, making coordinate-reading uncertainty large.
• Treating a spreadsheet trendline equation as self-explanatory.
• Linearising the equation but plotting the untransformed uncertainty as though nothing changed.

6. Worked Examples

Next: Analysing and Evaluating Experiments