Field Analysis:SASSA Interpretation
From SASSA
SASSA Home Page ⇒ Field Analysis Home Page ⇒ Field Interpretation ⇒ SASSA Interpretation tool
[edit] SASSA Field Interpretation Tool
The SASSA Field Tool contains an interpetation tool designed to help archaeological users unfamiliar with the interpretation of soils and sediments. Using soil properties described in the field users can obtain guidance in answering the following common geoarchaeological queries.
The interpretation tool is intended to support users who are less familiar with soil/sediment recording and interpretation by providing help in answering the common context-specific field queries listed below.
The aim is to introduce users to those soil properties that may be useful when thinking about the particular query. The interpretations will never be as reliable as those of an experienced archaeologist or geoarchaeologist making observations in the field.
[edit] Interpretation Recording Sheets
Printable recording sheets can be accessed by clicking on the attachments banner at the bottom of this page. These may be used in the absence of a mobile computing device to ensure that all data needed to answer a particular query is collected. This data can then be used in the SASSA Field Tool at a later time to answer that particular query. Choose a query from the list of attachments to open the PDF, which can be saved or printed.
Users are urged to familiarise themselves with the use and limitations of the SASSA Interpretation Tool before using these sheets in the field. Unless stated otherwise all the questions relate to the particular context of interest.
Follow the arrows answering the questions as you go. When a highlighted box is checked follow the arrows along the branch and down any sub-stems, before tracing your way back and carrying on down the main stem. A "X" after a box signifies the end of the tree and no more questions need be answered.
- Is this a buried soil?
- Is this deposit the 'natural'?
- Has this deposit been affected by in-situ burning
- Has this deposit been affected by bioturbation?
- Is this a colluvial deposit?
- Has this deposit been affected by podzolisation?
- Has this deposit been affected by waterlogging (gleying)?
- Is this a water-lain deposit?
[edit] Weightings and calculating the score
The final answer to each query in the interpretation tool will be a number between 0 and 100. The closer to 100 the more likely the query is to be true; the closer to 0 the more unlikely. A score of 50 or thereabouts indicates that either there is not enough evidence to answer the question or the evidence is conflicting and the system doesn't know.
The system uses weightings of between 0 and 1 attached to each question answer to calculate the score. These weightings are shown on the printable interpretation tool recording sheets, but are not shown in the digital version as they make little sense in isolation. The way these weightings work is as follows.
Each geoarchaeological query consists of a "tree" of questions. These questions are arranged both vertically (trunks) and horizontally (branches); see figure 1. The actual arrangement of questions is specific to each query. Weightings along the branches are multiplied together, whilst weightings in trunks are added together. The calculation always works back from the terminal ends of the branches and branches and sub-branches should be resolved before adding trunk and sub-trunk weightings together; see figure 2,
The weightings in the questions down the main trunk are divided such that the highest weighting for each question will add up to 1. The weightings along the branches act to modify the trunk weightings either to reduce the weighting given to the trunk answer or to leave it unchanged. The result is a score between 0 and 1 which is multiplied by 100 to give the final score. This is interpreted using the chart below.
| 0-15 | It is very unlikely the query proposition is correct |
| 16-30 | It is unlikely the query proposition is correct |
| 31-45 | It is moderately unlikely the query proposition is correct |
| 46-55 | There is insufficient data or the evidence is too ambiguous to answer this query |
| 56-70 | It is moderately likely the query proposition is correct |
| 71-85 | It is likely the query proposition is correct |
| 86-100 | It is very likely the query proposition is correct |
Nb. The weightings do not reflect any statistical significance

