Averaging Trials Versus Averaging Trial Peaks: Impact on Study Outcomes

Area of Science:

• Biomechanics
• Human movement analysis

Background:

• Gait data analysis often averages multiple trials or participants.
• Discrete data points (peaks, minima) are used for statistical analysis.
• Inconsistent methods for averaging gait data exist in literature.

Purpose of the Study:

• Compare statistical outcomes of two gait data averaging methods.
• Evaluate averaging peaks from multiple trials versus a single peak from an average profile.
• Determine the impact on dependent variables in biomechanical research.

Main Methods:

• Paired-samples t tests to compare dependent variables.
• Analysis of peak values derived from averaged individual trials.
• Analysis of peak values derived from a single average trial profile.
• MATLAB simulations using sine curves to model phase shifts.

Main Results:

• Averaging individual trial peaks yielded significantly larger dependent variable magnitudes.
• Identifying a peak from the average profile resulted in smaller magnitudes.
• Temporal differences in peak locations between trials caused discrepancies.
• Simulations confirmed misrepresentation of peaks in average profiles with phase shifts.

Conclusions:

• Averaging individual trial peaks is a more accurate representation of gait data.
• Using a single peak from an average profile can misrepresent data magnitude and temporal characteristics.
• Standardizing gait data analysis methods is crucial for reliable biomechanical research.