Article Text

Download PDFPDF
Interrater reliability assessment of a qualitative video analysis protocol to assess injury mechanisms and risks
  1. T Savage,
  2. A McIntosh,
  3. B Fréchède
  1. University of New South Wales, Sydney, Australia


Background Understanding injury aetiology is vital to the development of preventative programs. However, evaluating causation in multifactorial events is difficult. Qualitative analysis may present an effective method to understanding injury situations.

Objective To develop a reliable qualitative analysis method to assess injury risk, using tackling in rugby union as a target activity.

Design An analysis protocol was developed following consultation of epidemiological, coaching, and biomechanics literature. An inter-rater reliability (IRR) study was conducted to assess the protocols reliability. The results of the IRR study were reviewed and the results used to inform modifications which were tested in a second IRR study.

Setting Video analysis of sports skills.

Participants 10 volunteers with a knowledge of the target sport.

Interventions A standard instruction manual and video clips were provided to each participant. Participants were asked to code all events in the clips and a 5 min passage of play to test the protocol and the event definition used in the study.

Main outcome measurements IRR was assessed with percent agreement and Kappa.

Results Good or high agreement (>70%) was obtained in eight fields. Agreement among participants for ‘potential injury events’ was high (91%). Agreement in identifying events according to the definition, and the overall agreement in the selected fields was good (61%). In general, agreement in fields using scientific or rugby specific terms were low. Higher agreement was identified for dichotomous fields. The results of the second IRR study showed that a more specific definition of the point in time during a tackle at which observations were made improved the agreement in some fields.

Conclusion The results suggest general agreement between raters using the protocol was good, but care needs to be taken in explaining and defining the variables used. Following modifications, the protocol was used in a broader study examining technique and injury risk.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.