Running: Running and Oxygen Consumption

In byandell/pda: Practial Data Analysis for Designed Experiments

Description

This study examined how the metabolic cost of locomotion varied with speed, stride frequency and body mass. Cost was determined by measuring oxygen consumption (vo2), analyzing the oxygen content in air inhaled and exhaled by the subjects through a mask. The rate of oxgen consumption was measured for three subjects locomoting at all combinations of low and high levels of the three factors — running speed, stride frequency, and mass distribution in the leg. The first factor was set using a treadmill, the second by synchronizing the subjects' pace with a metronome, and the third by varying the positions of weights strapped onto the legs or waist. In addition to these three design factors, other variables were measured by filming each trial with a high-speed motion camera. Ignore these other measurements for this problem.

The order of the eight combinations was randomly assigned for each test session. For each test combination, subjects continued running until their rate of oxygen consumption levelled off, between eight and 15 minutes into the run, signalling that steady state had been reached. Subjects were given daily sessions for an initial training period, during which the rate of oxygen consumption at each test combination decreased to a relatively constant level. Subjects were considered trained when their oxygen consumption for all test combinations were consistent and repeatable between test sessions. All subjects achieved the trained state within two weeks. The data used in this study were collected in single sessions from trained subjects. However, there still might be some concern about incomplete training causing systematic differences in consumption during the session.

Usage

data(Running)

Format

Running data frame with 24 observations on 13 variables.

[,1]	subject	factor	subject identifier
[,2]	order	ordered	order of treatment
[,3]	speed	factor	running speed
[,4]	stride	numeric	stride frequency
[,5]	mass	numeric	attached mass in g
[,6]	vo2	numeric	oxygen consumption
[,7]	tm	numeric	time
[,8]	etot	numeric	etot
[,9]	cn	numeric	cn
[,10]	ar	numeric	ar
[,11]	ecn	numeric	ecn
[,12]	ahm	numeric	ahm
[,13]	vhm	numeric	vhm

Source

MJ Myers

References

Myers MJ, Steudel K and White SC (1993) 'Uncoupling the correlates of locomotor costs: a factorial approach', J Experimental Zoology 265, 211-223.

Examples

data( Running )

Running$code <- interaction( Running$subject, floor( Running$speed ),
  factor( Running$stride ), factor( Running$mass ))

Running.fit <- lm( vo2 ~ subject+speed+stride+mass, Running )
Running.intfit <- lm( vo2 ~ speed*stride*mass, Running )

print( xyplot( vo2 ~ order, Running, groups = subject,
  type = "p", pch = levels( Running$code ), cex = 2,
  xlab = "order of treatment", ylab = "VO2",
  main = "Running Problem" ),
  more = TRUE, split = c(1,1,2,1) )

Running$pred = predict( Running.fit )
print( xyplot( pred ~ order, Running, groups = subject,
  type = "p", pch = levels( Running$code ), cex = 2,
  xlab = "order of treatment", ylab = "VO2",
  main = "(sub.speed.stride.mass)" ),
  split = c(2,1,2,1) )

byandell/pda documentation built on May 13, 2019, 9:27 a.m.