CONFOUNDED STUDY PRODUCES DANGEROUS IMPLICATIONS
Hohmann, A., Fehr, U., Reuss, A., Kieser, S., & Straub, S. (2010). Specific strength training and start performance in swimming. A paper presented at the XIth International Symposium for Biomechanics and Medicine in Swimming, Oslo, June 16–19, 2010.
[Editor's note: This study does not conform to acceptable group or intra-subject experimental designs and therefore, do not allow one to infer causality (which the authors do).]
In this study it was hypothesized that improving maximum strength and power in squat, counter movement, and drop jumps lead to better start times in elite swimmers (M = 5; F = 2). Ss from different German national swim teams performed a four-week training intervention to optimize maximum strength and jumping power, as well as the start technique. In single case experiments, each S performed maximum swim start trials as well as strength and jumping power tests before and after the four-week training period. Four weeks later a retention test was administered. The overall start time over a distance of 7.5 m was recorded by video analysis. Kinematic parameters (block time, flight time, and overall start time, angles at take-off and water entry, take-off velocity) were calculated by motion analysis.
Ss improved strength and power, as well as start time. The best female S (WC 2006 qualifier) improved maximum strength of both legs, and drop jump index (136 to 166 pts) during the training period. In the final qualification competition, the best female and male S reached personal best start times in both crawl and backstroke.
Speculation. The single case experiments showed that a four-week training intervention with three weekly training sessions aimed at the maximum and explosive strength of the legs, as well as the swim start technique might be helpful for improving in most of the strength parameters and the overall start performance in elite swimmers. It is not known if both factors (strength training and start practice) affected start performance or whether only one factor was causal or to that matter some uncontrolled factor was causal. Only acceptable experimental designs in a research protocol would answer the causal possibilities.
This is an excellent example of an unacceptable confounded study. Since no control group was used and the study was directly open to experimenter bias, the logic of the selective conclusions is spurious. Any observed changes could have been caused by, among other things, pure biomechanical improvements in starts, better leg power, an interaction between the skill and physical variables, or the summation of both variables. When using two general forms of training (strength and technique), minimal requirements would be a control group that does nothing special, a strength group, a technique group, and a double-treatment group. Consequently, the conclusions of this study are not warranted except for the fact that improvements were noted over the period of the investigation. The causes for those improvements are not clear.
Return to Table of Contents for Training for Swimming.