Datasets for Statistical Analysis

Butterfat and dairy cattle

Keywords:

butterfat, cattle, production

Categories:

regression, designed experiments, generalized linear models;
biology

Description:

The data give the average butterfat content (percentages) for random samples of twenty cows (ten two-year old and ten mature (greater than four years old)) from each of five breeds. The data are from Canadian records of pure-bred dairy cattle.

There are 100 observations on two age groups (two years and mature) and five breeds.

Variables:

Butterfat	The average butterfat percentages
Breed	The cattle breed: One of Ayrshire, Canadian, Guernsey, Holstein-Fresian or Jersey
Age	The age of the cow: Either Mature or 2years

Data Quality:

There are no missing values.

Source:

Hand et al. dataset 23.

Notes:

The data can be used in analysis of variance. In a regression analysis, a reciprocal transformation is useful. The data also are useful for fitting a Tweedie generalized linear model with p about 5.

References:

Originally from Sokal, R. R. and Rohlf F. J. (1981). Biometry, 2nd edition, San Fransisco: WH Freeman.