As a general principle, undefined information is excluded from the charts and uncertain data are averaged out. The way this has been done is outlined below for (a) individual hoards and (b) groups of hoards.
The visualisations we have developed (see here for a detailed description) are intended to facilitate the analysis of our dataset to meet general needs. They provide set queries to show trends. We do not aim to cater for all the specific requirements of individual researchers. For these users are encouraged to download data and use suitable visualisation software themselves.
Visualisations for individual hoards
Circulation visualisation
- The Map does not display uncertain data. These are listed in the section ‘Data not added to the map’.
Content visualisation
- Where there are multiple entries for any of the fields (Reign, Person, Mint, Denomination, Material), numbers have been averaged out between entries.
For example, 20 coins for which the Person is given as ‘Commodus (Caesar)’ OR ‘Commodus (Augustus)’ will have 10 coins counted as ‘Commodus (Caesar)’ and 10 coins as ‘Commodus (Augustus)’.
- Undefined data are included and listed in a separate column.
Visualisations for groups of hoards
Hoards found by decade
- Decades are displayed on the charts by their starting year (1900, 1910, etc.).
- In cases where the period of discovery of a hoard extends over two or more decades, values are averaged out between them. For example, one hoard found between 1800 and 1899 would have 0.1 counted for each of the ten decades between 1800 and 1899.
- Undefined data are not included in the charts. This applies to cases where no discovery date is given or to finds for which only the date they were first reported is known.
By closing date
- If a hoard has a closing date comprising a range of years, results are averaged out. For example, one hoard ending in AD 235–238 will have 0.25 counted for each of the four years (AD 235, 236, 237, and 238).
- Closing dates of hoards are often derived from the dates of reigns. In most cases, the reign dates of an emperor and his successor will overlap. Charts mapping data involving consecutive reigns will therefore show an artificial peak for such years, e.g. in AD 244 for hoards ending under Gordian III (AD 238–244) and Philip I (AD 244–249). This is more noticeable when charts include a large number of finds for both reigns.
- Undefined or uncertain data are not included in the charts. This applies to poorly recorded or partially preserved finds, where the documentation available does not give any definite information as to the end date of a hoard.
By closing reign
- In the rare eventuality that two values have been entered for the closing reign, values have been averaged out between both reigns.
Histograms: Coins by date of minting
- Dates are derived from data entered for each coin.
- Where no dates have been entered for a coin, data is derived from the Reign under which a given coin was issued, e.g. for a coin minted under Tiberius, this would then be AD 14 – 37.
