TESTING FOR TESTING SAKE
Rushall Thoughts, (1996).
In response to a released report from Australian Swimming Inc. about "scientific" testing conducted at a sprint/butterfly camp held in Singapore in June, 1996.
[Maw, G. (1996, June). Sprint freestyle/butterfly national event camp - Singapore Swimming Club, Singapore. Report prepared for Australian Swimming Incorporated, Canberra, Australia.]
- No set of swims were repeated and so it was not possible to gauge any changes with a degree of acceptable reliability. One can only guess at what happened.
- I am not aware that any of the sets has established validity or reliability and so cannot be deemed to be a scientific measure which can be compared to anything else.
- Each set has unique demands that have not been determined. They are labeled and justified on face validity which is not acceptable for protocols intended to yield measures. Since each measured set assesses something different and therefore, requires different response patterns and magnitudes, it is invalid to compare results (the "apples and oranges dilemma").
- The patterns of HRs and LAs in the sets were not consistent across Ss. This means that the standardization of the sets is inconsistent and cannot be generalized across the group with averages, etc. At best, it is only feasible to do case-study reporting. There is no scientific justification for pooling any data for trend descriptions.
- The single testings only allow descriptions of what was observed. One cannot infer from an N of one as in this case. All that there is are some measures taken on regimens which have no proven validity or reliability. That translates into TESTING FOR TESTING SAKE.
- A jaundiced viewpoint would be that this is an instance of trying to make science work and to justify its presence when really the primary concern should be benefits that can be provided athletes NOT programs.
- The testing is very "western." It is work-oriented and pays no heed to recovery between repetitions or daily sets. This does not lead to individually appropriate training stimuli which accommodates individual differences in recovery and work tolerances.
- The inter-individual variations in response patterns and levels are very noticeable which further precludes any generalization of results across subjects.
- The testing of swimmers who are on antibiotics or ill raises an ethical question. I am sure you are aware of the danger of antibiotics to the kidneys and liver when coupled with moderate to hard exercise. Ill swimmers should not have been further stressed by being placed in a "serious" assessment situation.
- So what happens with these results? They have no reliability, no comparative value to other sets of data, and are confounded by heat acclimatization. Although Dr. Maw rightfully recognizes some compromise because of heat/humidity adaptation and the exposure to altitude (in the environment of the plane flying to Singapore), it would have been scientifically appropriate to wait until these transitory factors had stabilized before doing any other form of stress/test assessment.
- The training programs seem to be mixed, that is, not enough repetitions of any particular swim were performed to stimulate meaningful changes. Mixed training produces mixed results so it is no wonder that Australian sprinting is poor.
- For a sprint/fly/anaerobic events much of the swimming speeds were not specific, all being too slow for intended competitive paces. If what was trained is transferred to competitive situations, then competitive times will be slow for that is the pace that was practiced. Even in lactate sets the level of lactate of the last repetition was high while the pace that was experienced was usually much slower than one would want in a race. What is the connection between these two phenomena for training benefits that can be transferred to competitions?
- Preliminary to considering what can be learned from each test is testing in a manner so that something might be learned. I cannot see where that was done. You cannot tell anything from one test because there is no reliability of data. It is like trying to say VO2max is one data point when in reality you have to have a minimum of three points, all within in a narrow range to show stability (reliability) of the phenomenon.
On the other hand, there are some established protocols in physiology testing that allow some inferences about phenomena (e.g., OBLA and LA threshold). But they have to be demonstrated in acceptable protocols with adequate controls over variables that might affect the reliability of data. Doing coach-determined sets, even though they have been labeled "heart rate set" or "lactate threshold set," does not fit this criterion. It does not matter how many times the sets have been done before, the numbers and objective verification and validation have to be provided for the scientific community to evaluate and they never have. Remember, Australian Swimming Inc. has always kept these things "secret" which, in science, means there are no such numbers or data or consistently demonstrated phenomena across subjects.
So I conclude: Nothing was learned from these measures because of the standard of exercise testing. Even if good testing had been done, one-off tests are still unreliable unless criteria for demonstrated phenomena are displayed.
The pattern of meaningless sport science testing continues in Australian swimming. I am not willing to take the opinions of any Australian "scientist" who works in this manner and gives "credence" to this being science.
Return to Table of Contents for this issue.