These reflections were prompted by a recent performance of Odissi by the Nayikas Dance Theatre Company, the New York City Ballet and other sources.

I would like to discuss an issue that is near and dear to my heart: the methodology of measurement. Is it possible to develop a method to compare the skill level of a classical ballet dancer and a classical Odissi dancer? I choose these two as an initial comparison because both require long discipline to reach the highest levels and have long choreographic traditions, yet also have some fairly significant differences.

In classical ballet, one of the highest achievements a dancer can attain is the ability to dance en pointe, and hold a position without shaking for some extended period of time. One of the elements that makes classical Odissi so impressive is the dancers' ability to stand on one foot without shaking for extended periods of time. Ballet is normally danced in pointe shoes and Odissi is normally danced barefoot. Since we are interested in comparing peak performance to peak performance within existing disciplines, we can't measure the ballet dancer dancing barefoot, just as we can't measure the Odissi dancer using pointe shoes. Such tests would tell us something, but not about the authentic experience of dancing within each discipline. Our goal is to find a common, directly comparable measurement that can be derived from apparently different subject matter. (For an example of a similar challenge solved by my colleagues and myself, see Stoddart, Trish; Abrams, Robert; Gasper, Erika & Canaday, Dana. 2000. "Concept Maps as Assessment in Science Inquiry Learning - A Report of Methodology" in The International Journal of Science Education. Volume 22, Number 12, p. 1221-1246.)

I would start by measuring the time from the top of the arc of a one foot position to the point at which noticeable shaking begins. Eventually these start and stop times could be determined by the analysis of digitized video, but it should be possible to get reasonably accurate results with a team of three observers who would undergo an inter-rater agreement training phase.

The ability to hold still is partly a function of how tired one's body is when the attempt is made. This is an authentic question when applied to dance because a dancer may be called upon to hold still at any point in a work. Thus, to be fully useful, the measurements ought to be taken after a short warm up, and then again at several points after periods of controlled amounts of physical exertion. This would mimic the sport of the Bi-athalon, where people ski and then have to shoot at targets.

I would run these measurements with Odissi dancers standing on one flat foot, ballet dancers standing on one flat foot, and ballet dancers standing en pointe. My hypothesis is that with these three sets of measurements, one could establish a co-efficient to convert flat footed hold times to en pointe hold times. I would predict that the co-efficient is at least partly related to the ratio of the surface area of the flat foot to the surface area of the pointe part of the pointe shoe.

Once we had all of this data, we could analyse it to see what kinds of cross-dance conclusions can be drawn. While it is unlikely to provide exact comparisons between two specific dancers, it is likely to tell us something about the differences in dance ability across groups and the changes in ability over time.