The main limitation relates to the fact that both 3D motion analysis and plantar pressure systems are known to be linked to variable data output [
21,
30]. The current study took steps to minimise this variation however, and the degree of variation is similar for both taped and untaped conditions. 3D motion analysis is a relatively new method of analysing the effect of LD taping on rearfoot motion. All surface marking systems carry a certain degree of error when estimating the motion of joints, however the CODA motion analysis system is sufficiently reliable if a number of gait cycles are used, similar to this study [
20,
21]. It is difficult to compare absolute values of plantar pressure systems across studies, and it is more appropriate to compare plantar pressure distributions under constant conditions, as in this study [
31]. Secondly, a strict protocol was followed when using the F-Scan in order to make the procedure reliable. The F-scan system is highly correlated with force platform measures [
16] and is sufficiently reliable [
15], particularly when a mean of 3 steps is taken as the representative value [
16], similar to recommendations for other pressure measurement systems [
32]. Thirdly, other factors which could affect validity e.g. walking speed and surface contact [
33], were consistent between taped and untaped conditions. The use of footwear was different for each measurement type, but once again this was consistent between taping conditions. Ideally, the measurement of 3D motion and plantar pressure would occur simultaneously to ensure the gait cycle analysed was identical, and the effect of taping could not have changed. The desire to examine in-shoe plantar pressures obviously would not allow visualisation of the skin markers. Therefore, simultaneous data collection was not possible and correlations between changes in kinematics and plantar pressure distribution were neither possible nor appropriate. This potential bias was minimised by gathering multiple cycles for each measurement system, in line with recommendations regarding a suitable number for adequate reliability for each system [
16,
21]. This resulted in a different number of gait cycles being performed for plantar pressure and motion analysis, however the number of gait cycles did not vary between the taped and untaped conditions. The absence of a force plate also meant the authors had to visually gauge where heel-strike and toe-off occur. This method has, however, been recommended by the manufacturers and been described in previous research [
20]. The need to reposition motion analysis skin markers after the removal of LD tape requires that the kinematic results be interpreted with some caution, as there is a small risk that this could have resulted in slight changes in kinematic angles. Similar to some previous LD taping trials [
5,
7], the reliability of the investigators was not established in the current study, however this is a potential source of error. This is particularly important given the small magnitude of change between conditions and the high variability of the data. The SEM values for kinematic data exceeded the statistically significant difference observed between taped and untaped conditions. Therefore the data should be interpreted with caution, as some of the difference observed between groups could be due to simply measurement error. A clearer indication of the reliability of the study protocol would require a test-retest reliability study to be performed in advance. The sample size is however in line with previous LD taping trials [
5,
7,
9,
10,
26]. This high level of data variability is commonly noted in studies of plantar pressure, LD taping and lower limb kinematics [
5,
15,
16]. The effect of taping was examined only during the stance phase, due to the fact that maximum pronation has been found to occur during the middle-to-late stance phase of the gait cycle [
1], and symptoms are usually related to weight bearing. The size of the change with LD taping was statistically significant, but we cannot say whether this would be clinically significant. We did not examine whether the taping was performed identically for each subject, however one person performed all taping to minimise error and the tape applied did not change between the two measurement techniques. The sample size was small, and a suitable power calculation was not performed due to the exploratory nature of the study, and this limits external validity. Subjects were not randomly selected, but were a sample of convenience. The amount of time subjects were given to become accustomed to the tape varied somewhat between 2 and 3 minutes, which is a potential source of error. Also, there is a very slight risk of a residual effect of taping even after its removal, which could potentially have affected the baseline 'untaped' kinematic data. Finally, neither subjects nor investigators were blinded to taping condition, as this was not feasible.