Datasets for Statistical Analysis

Babyboom Data

Keywords:

Poisson, geometric, binomial, normal, births, hospital,

Categories:

one-sample data;
health

Description:

The data give the birth weight, gender, and time of birth of 44 babies born in the 24-hour period of 18 December 1997 at the Mater Mother's Hospital in Brisbane, Australia. (At the time, this was a record number of births in one 24-hour period.)

Variables:

Time	The time of the birth, as recorded on a 24-hour clock
Gender	The gender of the baby: 1 means a girl; 2 means a boy
Weight	The birth weight in grams
Minutes	The minutes since midnight when the birth occurred.

Data Quality:

There are no missing values.

Source:

The data appeared in the Sunday Mail newspaper (Brisbane Australia) on 21 December 1997. They were used in the following article:

• Dunn, Peter K. (1999). 'A Simple Dataset for Demonstrating Common Distributions' in the Journal of Statistics Education, 7(3).

Notes:

See the above article for further details. The number of births per hour can be modelled using a Poisson distribution. The exponenial distribution can then be used for the modelling the time between births. The binomial distribution can be used to determine if the data suggests a high proportion of boys. The geometric distribution can be used to to model the number of births untl a boy/girl is born. The normal distribution can be used to model birth weights. All the distributions fit reasonably well, apart from (predictably) the normal for birth weights.

The data could also be used for a runs-test for gender, equality of proportions, demonstrating skewed data (birth weights) and goodness-of-fit tests.

References:

• Dunn, Peter K. (1999). 'A Simple Dataset for Demonstrating Common Distributions' in the Journal of Statistics Education, 7(3).