Description:
Counts of the number of yeast cells were made each of 400 regions in a 20 x 20 grid on a microscope slide, comprising a 1 sq. mm. area. This experiment was repeated four times, giving samples A, B, C and D.
Student (1906) used these data to investigate the errors in random sampling. He says "there are two sources of error: (a) the drop taken may not be representative of the bulk of the liquid; (b) the distribution of the cells over the area which is examined is never exactly uniform, so that there is an 'error of random sampling.'"
The data in the paper are provided in the form of discrete frequency distributions for the four samples. Each shows the frequency distribution squares containing a count of 0, 1, 2, ... yeast cells. These are combined here in Yeast. In addition, he gives a table (Table I) showing the actual number of yeast cells counted in the 20 x 20 grid for sample D, given here as YeastD.mat.