poteries {SARP.compo}

Composition of Roman poteries

Description

This data set gives the oxide composition of several potteries found in five different archaelogic sites of the United Kingdom. Composition was obtained using atomic absorption spectrometry.

Usage

data(poteries)

Format

A data frame with 14 columns and 48 rows. Each row gives the composition of a pottery (columns 2 to 10), the archaelogical site where it was found (columns 6 and 7):

ID	factor	Pottery sample identifier (see original paper appendix)
Al	numeric	Percentage of aluminium oxide
Fe	numeric	Percentage of iron oxide
Mg	numeric	Percentage of magnesium oxide
Ca	numeric	Percentage of calcium oxide
Na	numeric	Percentage of natrium oxide
K	numeric	Percentage of kalium oxide
Ti	numeric	Percentage of titanim oxide
Mn	numeric	Percentage of manganese oxide
Ba	numeric	Percentage of baryum oxide
Site	factor	Kiln site
Pays	factor	Location of the kiln site
Couleur	factor	External color of the pottery
Date	factor	Approximate date of the pottery

Note

The DASL version of the dataset, as presented in the "Pottery stoty", does not include data on the poteries from the Gloucester site, neither the data on K, Ti, Mn and Ba oxides. It neither includes the color and date informations, and codes sites as their first letter only.

The DASL version of the dataset exists in the car package, as the Pottery dataset (with two locations differently spelled).

Source

Downloaded from the DASL (Data and Story Library) website, and completed from the original paper of Tubb et al.

References

A. Tubb, A. J. Parker, and G. Nickless (1980). The analysis of Romano-British pottery by atomic absorption spectrophotometry. Archaeometry, 22, 153-171.

Hand, D. J., Daly, F., Lunn, A. D., McConway, K. J., and E., O. (1994) A Handbook of Small Data Sets. Chapman and Hall – for the short version of the dataset.

Examples

  data( poteries )
  # Reconstruct the car version of this dataset
  dcar <- poteries[ , c( 'Al', 'Fe', 'Mg', 'Ca', 'Na', 'Site' ) ]
  dcar <- droplevels( dcar[ -which( dcar$Site == "College of Art" ), c( 6, 1:5 ) ] )
  levels( dcar$Site )[ c( 1, 3, 4 ) ] <- c( "AshleyRails", "Islethorns", "Llanedyrn" )

  # Reconstruct the DASL version of this dataset
  ddasl <- dcar[ , c( 2:6, 1 ) ]
  levels( ddasl$Site ) <- c( 'A', 'C', 'I', 'L' )