Review 1 - Understanding Shortcomings in Research

When comparing groups of people in the way how they move, we first must make sure we do it in a way the acquired data can actually tell us something meaningful. Understanding and acknowledging some pitfalls when designing, conducting, and interpreting the data is crucial to NOT conclude something misleading or wrong. Here are two examples of how to do things better:

Confounders

When wanting to determine in what way one factor affects another (in this case gender affects movement), then we must ensure, we match the groups for every other meaningful confounding factor. We know and preach the necessity of muscular strength for athletes and understand that the lack of it, can have negative consequences for their performance and movement. Yet, very few studies actually match for strength or training experience when quantifying movement between males and females – those that do, tell a pretty convincing story that gender is not a “cause”…

Besides just the physical capacity of subjects (i.e., muscular strength), how do you think the quantity and quality of training impact an athlete’s movement pattern? When discussing the “self-regulation” approach, just the sheer lack of strength will delimit an athlete’s amount of movement solutions for a particular task…isn’t it then good enough to first get as strong as possible while learning how to utilize this newfound strength within the skill? How do you approach that?

Validity of data

We need to be confident about the accuracy of the reported data, considering we are interested in every single degree of motion. The location of our joint centers and joint axis determines the calculation of kinetics and kinematic later on, so we can’t allow any deviations here. Especially when trying to get some data for knee valgus and IR or non-sagittal plane movement, as seen in all the ACL research. With a static calibration (most papers refer to a plug-in gate calibration method) you estimate the joint center based on the location of the attached markers from one single frame. Marker attachment is however the number one source of error. A better method is to use is a dynamic or functional calibration, especially when quantifying fast movements like sprints, jumps, or CODs.

Do you consider the methods, when applying research results in your practice? Is it easy to understand what the authors did and what potential limitations are?