Table 4 Possible sources of heterogeneity

Methodology

Publication Bias

Convenience rather than random sampling may not represent true population

Lack of assessor blinding to group allocation

Small sample size

Population characteristics

Intensity of AKP

Duration of AKP

Recreational pursuits

Previous or concomitant physiotherapy

Age range

Previous knowledge pertaining to research question (i.e. population of physiotherapy students knowledge on VMO/VL debate and techniques to facilitate VMO

Procedural details

Differences between task

Differences between similar tasks- e.g stair height, timing of task

No of tests prior to data collection (learning effect/fatigue)

Results recorded as single test or mean of several

Electrode placement

Data sampling rates

EMG onset determination

Levels of smoothing or filtering