Table 2. A comparison of the advantages and disadvantages of MARP, DP and PRP for estimating joint relations during movement tasks.
Measure Potential Uses Advantages Disadvantages

MARP
DP
• Continuous movements
• Discrete movements
• Serial movements?
MARP
• Quantifies (estimates) inter-joint relationships during movements
• Averages all relative curve data points
• Reflects the whole curve dynamics
• Accounts for both displacements and velocities of joint or segment angles
DP
• Quantifies variability across several trials or cycles of movements
• Examines the variability along the whole curve or variability of the shape of the curves
• Multiple step calculation
• Postural joint movements and positions may require normalization of angular displacements and velocities
• Non- stationary relationships more likely seen in serial task; resulting in phase wrapping, e.g. one joint movement and one joint not moving
• Phase distortions (phase shifts and phase wrapping) in relative phase curve may distort or bias values
• Requires visual inspection and manual corrections for longer movements
• Absolute value biases the value away from zero
• Redundant values within 360-degree data

PRP
SD of PRP
• Continuous movements
• Serial movements?
PRP
• Quantifies (estimates) inter-joint relationships during movements
• Simple calculation to determine relative phase relationship
• Can be used with joint moving through different amplitudes and positions
• Eliminates asymmetries and irregularities in the actual motion of a task (Zanone and Kelso, 1992)

Inter-trial SD of PRP
• Quantifies variability across a several periods with in a task.
• Measured stability of postural coordination pattern in continuous tasks
• May eliminate important joint relationship dynamics between maximal peaks.
• Must have at least two unambiguous maximum or minimum points within the reference angle; decreases utility with discrete movements or if examining movement intervals
• Shorter duration tasks may have a small number of reference periods available to calculate
   PRP
• Selecting peaks may be difficult within serial tasks: non-sinusoidal signals (Wheat and Glazer, 2006)
• Postural movements are often small and gradual resulting in a decrease number of unambiguous points.
• Requires visual inspection of point selection.
• Redundant values in 360 within degree data