Description Usage Format Details Source References See Also Examples
Results of chemical analyses of 48 specimens of Romano-British pottery published by Tubb et al. (1980).
The numbers are the percentage of various metal oxides found in each sample
for elements of concentrations greater than 0.01%.
This is the original data set from Tubb et al. (1980), in contrast to Pottery
.
1 |
A data frame with 48 observations on the following 12 variables.
Region
a factor with levels Gl
NF
Wales
Site
a factor with levels AshleyRails
Caldicot
Gloucester
IsleThorns
Llanedryn
Kiln
a factor with levels 1
2
3
4
5
Al
amount of aluminum oxide, Al_2O_3
Fe
amount of iron oxide, Fe_2O_3
Mg
amount of magnesium oxide, MgO
Ca
amount of calcium oxide, CaO
Na
amount of sodium oxide, Na_2O
K
amount of potassium oxide, K_2O
Ti
amount of titanium oxide, TiO_2
Mn
amount of manganese oxide, MnO
Ba
amount of BaO
The specimens are identified by their rownames
in the data frame.
Kiln
indicates at which kiln site the pottery was found;
Site
gives the location names of those sites.
The kiln sites come from three Region
s, ("Gl"=1, "Wales"=(2, 3), "NF"=(4, 5))
,
where the full names are "Gloucester", "Wales", and "New Forrest".
The variable Kiln
comes pre-supplied with contrasts to test interesting
hypotheses related to Site
and Region
.
Slightly modified from files found at: http://people.tamu.edu/~dcarlson/quant/data/RBPottery.html
Baxter, M. J. 2003. Statistics in Archaeology. Arnold, London.
Tubb, A., A. J. Parker, and G. Nickless. 1980. The Analysis of Romano-British Pottery by Atomic Absorption Spectrophotometry. Archaeometry, 22, 153-171.
Pottery
for the related (subset) data set
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 | data(Pottery2)
# contrasts for Kiln correspond to between Region [,1:2] and within Region [,3:4]
contrasts(Pottery2$Kiln)
pmod <-lm(cbind(Al,Fe,Mg,Ca,Na,K,Ti,Mn,Ba)~Kiln, data=Pottery2)
Anova(pmod)
# extract coefficient names for linearHypotheses
coefs <- rownames(coef(pmod))[-1]
# test differences among regions
linearHypothesis(pmod, coefs[1:2])
# test differences within regions B, C
linearHypothesis(pmod, coefs[3:4])
heplot(pmod, fill=c(TRUE,FALSE), hypotheses=list("Region" =coefs[1:2], "WithinBC"=coefs[3:4]))
# all pairwise views; note that Ba shows no effect
pairs(pmod, fill=c(TRUE,FALSE))
# canonical view, via candisc::heplot
if (require(candisc)) {
# canonical analysis: how many dimensions?
(pcan <- candisc(pmod))
heplot(pcan, scale=18, fill=c(TRUE,FALSE), var.col="darkgreen", var.lwd=2, var.cex=1.5)
## Not run:
heplot3d(pcan, scale=8)
## End(Not run)
}
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.