IB Physics Internal Assessment: Data Analysis Criteria

Overview and Weighting

The Data Analysis criterion is worth 6 marks, which accounts for 25% of your total Internal Assessment (IA) grade. This criterion evaluates your ability to record, process, and present data in ways that effectively address your research question.

Requirements for Top Marks (5-6 Level)

To achieve the highest marks, your report must demonstrate the following:

• Communication: The recording and processing of data must be both clear (easily understood) and precise (strictly following conventions).
• Uncertainties: The report must show evidence of an appropriate consideration of uncertainties.
• Processing: The processing of data relevant to the research question must be carried out appropriately and accurately.

1. Data Collection and Tables

• Sufficient Data: Collect enough relevant data to address your research question.
• Table Formatting: Provide appropriate titles and reference your tables in the text (e.g., "Figure 1").
• Column Headers: Include concise column headings with units and uncertainties. Standard formatting uses a solidus to separate the quantity and unit (e.g., d / cm) or parentheses.
• Qualitative Observations: Include labeled images, drawings, or written descriptions accompanying your raw data where applicable.

2. Units and Decimal Places

• SI/Metric Units: Use standard SI or metric units. Non-decimal units (like inches or degrees Fahrenheit) are not appropriate. Non-SI units (like eV or ly) are only acceptable if justified by the scientific context.
• Consistency: Maintain a correct and consistent number of decimal places based on your instrument's degree of precision. Match the decimal places of raw data, processed values, and uncertainties.

3. Uncertainties and Errors

• Measurement Uncertainty: Justify the size of your uncertainty based on the experiment. The minimum is the instrument's least count, but realistic uncertainties (such as human reaction time) or variations in repeated trials are often larger.
• Repeated Measurements: Determine the uncertainty of a mean by calculating one-half of the range between the maximum and minimum values.
• Significant Figures: Most uncertainties should be expressed to a single significant digit, though two are acceptable if the first digit is a "1".
• Propagation: Use the non-statistical rules in the DP Physics data booklet for propagating uncertainties through calculations.

4. Processing Data and Graphing

• Methodology: Show your processing clearly. Means, standard deviations, or percentages are sufficient for simple processing. For complex spreadsheets, include a screenshot with the formula used. Provide worked examples for unorthodox calculations.
• Axes and Labels: Label both axes with full quantity names or symbols, and include units. Ensure the scale and proportions are adequate.
• Trend Lines: Draw appropriate best-fit lines or curves guided by theory. Do not simply connect data points. Scatter should be roughly equal above and below the line. Do not force the line through the (0,0) origin unless theoretically sound. Extrapolating beyond your maximum value can provide additional evidence.
• Uncertainty Bars and Min/Max Lines: Include uncertainty bars. Estimate minimum and maximum trend lines by eye considering all uncertainty bars—do not merely connect the first and last data points.
• Linearization: Linearization is not mandatory, but if used, minimum and maximum lines can help determine uncertainty in the gradient and intercept.

5. Handling Outliers

• Do Not Delete Erroneously: Do not systematically remove data points to make results fit a theory. Outliers are real measurements.
• Justification Required: If you exclude data, you must provide a strong scientific justification. Remeasure if human error or methodological flaws are suspected.
• Limit Deletion: Never reject two or more data points as it compromises the dataset's integrity. Consider presenting the analysis with and without the outlier to reveal its impact.