Laboratory and Experimental Skills
The Laboratory and Experimental Skills domain tests measurement, uncertainty, significant figures, graphing, experimental design, and data interpretation. Questions often present a data table or describe a setup and ask for the slope's meaning, a source of error, or an improved procedure.
Measurement and uncertainty
Report a measurement to the precision of the instrument, usually half the smallest division for analog scales. Significant figures follow the operation: addition keeps decimal places, multiplication keeps the fewest significant figures. Percent error = |experimental − accepted|/accepted × 100%.
Accuracy, precision, error
Accuracy is closeness to the true value; precision is repeatability. Random error scatters results and is reduced by averaging repeated trials; systematic error shifts all results the same way and is not fixed by averaging. Identify specific sources-friction, reaction time, parallax, heat loss-rather than "human error".
Graphing
Plot the independent variable on the x-axis and the dependent variable on the y-axis. Draw a best-fit line, not a connect-the-dots path. The slope carries units (y-units per x-unit) and often equals a physical constant. Linearize nonlinear relationships-for a pendulum, plot T² versus L so the slope is 4π²/g. The R² value describes how well the line fits.
Design
Identify the independent, dependent, and controlled variables. Use repeated trials, a control group, and a sample size large enough to average out random effects. State how you would measure, not just what you would measure.
Worked example
A student plots T² (s²) versus length L (m) for a pendulum and gets a best-fit slope of 4.0 s²/m. Find g. For a pendulum, T = 2π√(L/g), so T² = (4π²/g)L. The slope equals 4π²/g, hence g = 4π²/4.0 ≈ 9.9 m/s². The calculation shows the slope's meaning and checks the result against the accepted value.
Traps
- Connecting data points instead of drawing a best-fit line.
- Omitting units from the slope.
- Calling a consistent offset "random error".
- Ignoring significant figures in the final answer.
- Changing more than one variable at a time.