Download presentation

1
**Computing in Archaeology**

Session 9. Sampling Assemblages © Richard Haddlesey

2
Aims To become familiar with sampling practices in an archaeological context

3
**Introduction to Sampling**

An area of excavation is a sample of the complete site which in itself is a sample of all sites of that type. The same goes for artefact assemblages. The essence of all sampling is to gain the maximum amount of information by measuring or testing just a part of the available material Fletcher & Lock 2005, 66

4
**Archaeological sample**

Sampled population Target population

5
Formal definitions Population: the whole group or set of objects about which inference is to be made Sampling fame: a list of the items, units or objects that could be sampled Variable: a characteristic which is to be measured for the units, such as weight of spearheads Fletcher & Lock 2005, 66

6
Formal definitions Sample: the subset or part of the population that is selected Sample size: the number in the sample. A sample size of 5 is considered small, while, formally, a sample size of 50 is large. The sample size maybe stated as a percentage of the sampling frame, e.g. a 10% sample Fletcher & Lock 2005, 67

7
Sampling strategies a simple random sample (probability sample USA) a systematic sample a stratified sample a cluster sample

8
population – 100 units . . . etc 100 obsidian spearheads

9
population – 100 units 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 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100

10
**A simple random number sample**

11
Random sampling If we have a sample of 100 spearheads, we simply pick 10 random numbers (i.e. 10%) Computers can help generate random sequences, but are not necessary You must avoid bias in your selection as this can result in scrutiny from others

12
**a simple random number sample**

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 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100

13
A systematic sample

14
Systematic sampling To take a systematic approach, we could choose every number ending in 4. Once again this would give us our 10% This method has the advantage of being easy to design unless the units have inherent patterning in their order

15
a systematic sample 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 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100

16
A stratified sample

17
Stratified sampling Here we take a random sample 5 from the top and five from the bottom Or 5 from the left, 5 right etc

18
a stratified sample 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 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100

19
a stratified sample 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 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100

20
A cluster sample

21
Cluster sampling Rather than select individual items, select clusters or groups of items that are close together This may result in bias values

22
a cluster sample 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 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100

23
a cluster sample 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 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100

25
**Downside to systematic**

Totally miss this context

28
**Common sample statistics:**

x – the sample mean s – the sample standard deviation p – the sample proportion (i.e. the proportion of the sample having a particular characteristic)

29
Stats The true population values for these statistics are usually unknown, and formally denoted by Greek letters

30
**Common sample statistics:**

known value estimate for x – the sample mean s – the sample standard deviation p – the sample proportion μ – the population mean

31
**Common sample statistics:**

known value estimate for x – the sample mean s – the sample standard deviation p – the sample proportion μ – the population mean σ – the population standard deviation

32
**Common sample statistics:**

known value estimate for x – the sample mean s – the sample standard deviation p – the sample proportion μ – the population mean σ – the population standard deviation π – the population proportion

33
**The central-limit theorem (the law of averages)**

In order to comment on how good an estimate the sample statistics are, the nature of their distribution needs to be known See Fletcher & Lock (2nd ED) 2005, Digging Numbers Oxbow 70-9

Similar presentations

Presentation is loading. Please wait....

OK

Chapter 4 Selecting a Sample Gay, Mills, and Airasian

Chapter 4 Selecting a Sample Gay, Mills, and Airasian

© 2018 SlidePlayer.com Inc.

All rights reserved.

To make this website work, we log user data and share it with processors. To use this website, you must agree to our Privacy Policy, including cookie policy.

Ads by Google