cluster.datasets (version 1.0-1)

indian.caste.measures: Hartigan (1975) Indian Caste Measurements

Description

The table contains the correlations multiplied by 10000 for 22 caste groups each with 67 to 196 individuals. This is Table 17.6 in Chapter 17 of Hartigan (1975) on page 324.

Usage

data(indian.caste.measures)

Format

A data frame with 9 observations on the following 9 variables.

st

a numeric vector for the correlations with stature

sh

a numeric vector for the correlations with sitting height

nd

a numeric vector for the correlations with basal depth

nh

a numeric vector for the correlations with nasal height

hl

a numeric vector for the correlations with head length

fb

a numeric vector for the correlations with frontal breadth

bb

a numeric vector for the correlations with bizygometic breadth

hb

a numeric vector for the correlations with head breadth

nb

a numeric vector for the correlations with nasal breadth

Source

Rao, C. R. (1948). The utilization of multiple measurements in problems of biological classification, J. Royal Stat. Soc. B, 10, 159 - 193. SPAETH2 Cluster Analysis Datasets http://people.sc.fsu.edu/~jburkardt/datasets/spaeth2/spaeth2.html

Details

The data frame has as row names the variable names. The actual correlations are recovered by dividing the data frame by 10000. Hartigan suggests performing a factor analysis on the data set as well as performing a joining algorithm.

References

Hartigan, J. A. (1975). Clustering Algorithms, John Wiley, New York.

Examples

data(indian.caste.measures)