2 Sampling Strategies in Archaeological Survey Free Online FlashCards

What does an archaeological sampling design specify?

A sampling design links the research question to the area or material examined, the units selected, how they are chosen, and the conclusions the observations can support.

How do a sampling unit and sampling frame differ?

A sampling unit is the spatial element selected for observation; the sampling frame is the defined set of units from which the sample is selected.

What does full-coverage survey guarantee—and not guarantee?

Full coverage means surveyors examine the entire defined area at a specified intensity; it does not guarantee detection of every buried site or artifact.

Name factors that can change archaeological detection.

Detection varies with vegetation, ground visibility, erosion, sediment cover, terrain, artifact size, surveyor experience, and field methods.

What limits the conclusions of a probability sample?

Inference must stay within the population and evidence the design can represent. A probability sample does not justify claims about omitted areas or unrecorded material classes, and imperfect detection still matters.

Must the selected unit match the recorded unit?

The selected unit need not be the recorded unit: a team may select a grid cell, survey transects within it, and record finds at smaller scales.

When can simple random sampling support population estimates?

Simple random sampling selects units randomly from the frame. If each unit has a known, nonzero selection probability, it can support statistical estimates for that defined population.

What is a key risk of systematic sampling?

Systematic sampling selects units or walks transects at regular intervals, usually from a randomized starting point. Repeating patterns in the archaeological record may interact with the regular layout.

Why use stratified sampling in an archaeological survey?

Stratified sampling divides the frame into meaningful groups, such as landforms, then samples within each so important but less extensive zones are represented.

How should unequal sampling fractions across strata affect overall estimates?

When strata are sampled at unequal fractions, overall estimates require appropriate weighting. Raw combined find counts should not be treated as an unbiased estimate for the whole area.

What is a trade-off of cluster or multistage sampling?

Cluster or multistage sampling selects groups first and then samples units within them. It can reduce travel and field costs, but nearby observations may be similar and affect precision.

What limits regional inference from purposive survey?

Purposive survey prioritizes locations expected to contain archaeological materials, using records, landscape knowledge, or predictive models. Unknown selection probabilities generally prevent treating it as a representative regional sample.