Mom Always Liked You Best:
Part II--The Authors Find a Method to Attack the Problem
During the breeding season of 1992, Bruce Lyon, John Eadie, and Linda Hamilton studied 90 pairs of coots nesting in the marshes of British Columbia. They decided to try and alter the plumage of the chicks by dyeing them. Unfortunately, the dye made the chicks sick and it removed the oils from their feathers. They next tried to alter the appearance of the chicks by cutting the orange tips off their body feathers. This produced black chicks that seemed to act the same as the normal orange chicks. The scientists now began a test of their hypothesis using this technique. What do you expect they might do?
Part III--What Should Be Measured?
The biologists decided to set up three types of nest conditions. In the first group of nests (let's call this the experimental group), they trimmed half of the chicks and made them black, and they left half of the brood with orange feathers. In a second group (call this a control group), all of the chicks were trimmed so they appeared black. In a third group of nests (call this another control group), all the chicks were left their natural color, orange. In all three groups, the chicks were captured within a day of hatching and were generally handled the same way even though some were trimmed. The chicks were kept in captivity for 30 minutes before being replaced in their nest. To control for the effects of hatching order in the experimental groups, the first chick hatched was randomly assigned to be trimmed or left orange. Thereafter, treatments were alternated with hatching order. The biologists worried about what kinds of data to collect, how to collect the data, and what kinds of results to expect. What would you suggest they do?
Part IV--At Last, Here Are Some Data!
The biologists decided to compare feeding rates, relative growth rates, and survival rates of the chicks in the different nests. Since the chicks had been individually color-marked, they could be easily observed and identified from floating blinds. To estimate growth rates, swimming chicks were photographed at known distances and their body length at waterline was estimated from projected slides. This measure of size is strongly correlated with body mass (r=0.97, n=43). Part of the data have been reproduced in the handout showing Figure 1. Predict the results of the other values by plotting the values on the graph. Why have you made these predictions?
Part V--The Rest of the Story
In the handout showing Figure 2 you will see the real results that the authors collected. How do the data compare with your predictions?
The authors ran a series of statistical tests and noted that there were no significant differences between the two control groups in any of the measures--that is, when the orange and black chicks were in separate nests, they had similar feeding, growth, and survival rates.
But in the experimental group where both black and orange chicks occurred together and the parents had a true choice of whom to feed, statistical analysis showed that the orange chicks fared better. Orange chicks were fed at a higher rate, had a higher growth rate, and enjoyed a higher survival rate than the black chicks in the same brood. What conclusions might the biologists make about their original hypothesis?