16
ELECTRONIC FIELD SURVEY TABLES Version 1.0 The spreadsheets contained in this workbook represent direct ca by Cannon and Roe (Bureau of Rural Science, Department of Prima The motivation for producing this workbook arose for several re desire to generate specific answers not dependent on tables or Every attempt has been made to remain consistent with the appro functions, in a form that allows direct calculations, rather th In some cases, a table has been calculated as well, although th of academic interest. The tables are as intuitive as I can make them. In most cases, indicate where/what should be placed in relevant cells. In add specific inputs are required or desirable, drop-down boxes offe I wish to thank Tony Martin, and Mario D'Antuono for their kind in producing this workbook. DEDICATION Chris Hawkins This workbook is dedicated to the li Veterinary Epidemiologist animal health and veterinary epidemi WA Department of Agricultur Rob Cannon and Dick Roe, whose coope April 2003 burden from non-statistical field ve of the tables in Livestock Disease Surveys. A Field Manual for Veterinarian fact that Cannon and Roe is now out of print, and both authors ha and to put the functionality of Cannon and Roe into the hands of by Rob Cannon and Dick Roe, while using the flexibility of Excel production of Livestock Disease Surveys lif

Sample Sizes - Cannon & Roe

Embed Size (px)

DESCRIPTION

File berisi tentang cara menghitung jumlah sampel. Ditulis oleh Dr. Chris Hawkins berdasarkan buku yang ditulis oleh Canon & Roe

Citation preview

Page 1: Sample Sizes - Cannon & Roe

ELECTRONIC FIELD SURVEY TABLESVersion 1.0

The spreadsheets contained in this workbook represent direct calculations of each

by Cannon and Roe (Bureau of Rural Science, Department of Primary Industry 1982)

The motivation for producing this workbook arose for several reasons, in particular the

desire to generate specific answers not dependent on tables or reading off graphs,

Every attempt has been made to remain consistent with the approach taken

functions, in a form that allows direct calculations, rather than table lookup.In some cases, a table has been calculated as well, although this is moreof academic interest.

The tables are as intuitive as I can make them. In most cases, pop-up commentsindicate where/what should be placed in relevant cells. In addition, wherespecific inputs are required or desirable, drop-down boxes offer a range of choices.

I wish to thank Tony Martin, and Mario D'Antuono for their kind assistancein producing this workbook.

DEDICATIONChris Hawkins This workbook is dedicated to the lifetime of service toVeterinary Epidemiologist animal health and veterinary epidemiology provided by WA Department of Agriculture Rob Cannon and Dick Roe, whose cooperation in theApril 2003

burden from non-statistical field veterinarians!

of the tables in Livestock Disease Surveys. A Field Manual for Veterinarians.

fact that Cannon and Roe is now out of print, and both authors have 'retired', the

and to put the functionality of Cannon and Roe into the hands of all field veterinarians.

by Rob Cannon and Dick Roe, while using the flexibility of Excel statistical

production of Livestock Disease Surveys lifted a great

Page 2: Sample Sizes - Cannon & Roe

Sample size required to detect diseaseConfidence limits for number positive

Confidence 0.99 Direct calculation

1.00%1600 399

Cannon and Roe Table 1Table format: desired confidence: 0.9

Percentage of diseased animals in the population, OR percentage sampled and found clean

50% 40% 30% 25% 20% 10% 5%10 3 4 5 6 7 9 1020 4 5 6 7 8 13 1830 4 5 6 7 9 16 2340 4 5 6 8 10 17 2750 4 5 7 8 10 18 3060 4 5 7 8 10 19 3270 4 5 7 8 10 19 3380 4 5 7 8 10 20 3590 4 5 7 8 10 20 36

100 4 5 7 8 10 20 36120 4 5 7 8 10 20 38140 4 5 7 8 10 21 39160 4 5 7 8 10 21 40180 4 5 7 8 11 21 40200 4 5 7 8 11 21 41250 4 5 7 8 11 21 42300 4 5 7 8 11 22 42350 4 5 7 8 11 22 43400 4 5 7 8 11 22 43450 4 5 7 8 11 22 43500 4 5 7 8 11 22 43600 4 5 7 8 11 22 44700 4 5 7 9 11 22 44800 4 5 7 9 11 22 44900 4 5 7 9 11 22 44

1000 4 5 7 9 11 22 441200 4 5 7 9 11 22 451400 4 5 7 9 11 22 451600 4 5 7 9 11 22 451800 4 5 7 9 11 22 452000 4 5 7 9 11 22 453000 4 5 7 9 11 22 454000 4 5 7 9 11 22 455000 4 5 7 9 11 22 456000 4 5 7 9 11 22 457000 4 5 7 9 11 22 458000 4 5 7 9 11 22 45

Population size (N)

Percentage of diseased animals in the population, OR Percentage sampled and found clean

Population Size (N)

B8
Enter population size from which the sample will be taken.
H11
Enter desired confidence level.
Page 3: Sample Sizes - Cannon & Roe

9000 4 5 7 9 11 22 4510000 4 5 7 9 11 22 45

"Infinite" 4 5 7 9 11 22 45

Page 4: Sample Sizes - Cannon & Roe

Percentage of diseased animals in the population, OR percentage sampled and found clean

2% 1% 0.5% 0.1%10 10 10 1020 20 20 2030 30 30 3038 40 40 4045 50 50 5051 59 60 6056 68 70 7061 76 80 8065 83 90 9068 90 99 10074 102 118 12078 113 135 14082 122 151 16085 130 166 18087 137 180 20092 150 210 25095 160 235 30098 168 256 350

100 175 273 399101 180 288 448102 184 301 495104 191 321 587106 196 337 674107 200 350 755107 203 360 830108 205 369 900109 209 382 1024110 212 392 1130110 214 400 1220111 216 406 1299111 217 411 1367112 221 426 1607113 223 434 1750113 224 439 1845113 225 443 1912114 226 445 1962114 226 447 2000

Page 5: Sample Sizes - Cannon & Roe

114 227 448 2031114 227 449 2056114 230 460 2302

Page 6: Sample Sizes - Cannon & Roe

0.90.95

0.9750.99

0.9950.999

Page 7: Sample Sizes - Cannon & Roe

Chance of detecting positives with various intensities of monitoring

Sampling proportion: 0.2 Cannon & Roe formula, p 30Number of positives in the population.

Population 1 2 3 4 5 6 7 810 2 0.199 0.377 0.532 0.665 0.776 0.864 0.931 0.97520 4 0.199 0.367 0.507 0.623 0.716 0.792 0.851 0.89630 6 0.199 0.364 0.500 0.611 0.700 0.772 0.828 0.87340 8 0.199 0.363 0.496 0.605 0.692 0.762 0.818 0.86250 10 0.199 0.362 0.494 0.602 0.688 0.757 0.812 0.85560 12 0.199 0.362 0.493 0.599 0.685 0.753 0.808 0.85170 14 0.199 0.361 0.492 0.598 0.683 0.751 0.805 0.84880 16 0.199 0.361 0.491 0.597 0.681 0.749 0.803 0.84690 18 0.199 0.361 0.491 0.596 0.680 0.748 0.801 0.844

100 20 0.199 0.360 0.490 0.595 0.679 0.746 0.800 0.8431000 200 0.199 0.359 0.487 0.589 0.672 0.737 0.790 0.832

Using Excel function HYPGEOMDIST Note limitations to the use of HYPGEOMDISTNumber of positives in the population

Population 1 2 3 4 5 6 7 810 2 0.200 0.378 0.533 0.667 0.778 0.867 0.933 0.97820 4 0.200 0.368 0.509 0.624 0.718 0.793 0.852 0.89830 6 0.200 0.366 0.501 0.612 0.702 0.773 0.830 0.87440 8 0.200 0.364 0.498 0.607 0.694 0.764 0.819 0.86350 10 0.200 0.363 0.496 0.603 0.689 0.758 0.813 0.85760 12 0.200 0.363 0.495 0.601 0.686 0.755 0.809 0.85370 14 0.200 0.362 0.494 0.599 0.684 0.752 0.807 0.85080 16 0.200 0.362 0.493 0.598 0.683 0.751 0.804 0.84790 18 0.200 0.362 0.492 0.597 0.682 0.749 0.803 0.846

100 20 0.200 0.362 0.492 0.597 0.681 0.748 0.802 0.8441000 200 0.200 0.360 0.488 0.591 0.673 0.739 0.791 0.833

"N/A" is inserted when the requirements of the HYPGEOMDIST function are not met.With discretion, in such cases assume the probability of detecting at least one positive is close to 1.

No. Sampled

No. Sampled

B18
Enter whatever number you choose, but watch out that it's not too much for the formula to handle.
Page 8: Sample Sizes - Cannon & Roe

Chance of detecting positives with various intensities of monitoring

Number of positives in the population. 0.05

9 10 0.10.997 0.997 0.150.930 0.955 0.20.907 0.934 0.250.896 0.923 0.30.890 0.916 0.350.885 0.912 0.40.882 0.909 0.450.880 0.907 0.50.878 0.905 0.550.877 0.904 0.60.866 0.893 0.65

0.70.75

Note limitations to the use of HYPGEOMDIST 0.8Number of positives in the population 0.85

9 10 0.91.000 1.000 0.950.932 0.9570.909 0.9350.897 0.9240.891 0.9170.887 0.9130.884 0.9100.881 0.9080.879 0.9060.878 0.9050.867 0.894

is inserted when the requirements of the HYPGEOMDIST function are not met.With discretion, in such cases assume the probability of detecting at least one positive is close to 1.

Page 9: Sample Sizes - Cannon & Roe

Assumes sampling from a very large population.

Number of animals in the sample testedPrevalence 5 10 15 20 25 50 75 100 200

0.01 0.951 0.904 0.860 0.818 0.778 0.605 0.471 0.366 0.134

Hint: vary the prevalence as you wish; alter the number of animals as required.

Probability of Failure to Detect Disease at a Given Prevalence

Page 10: Sample Sizes - Cannon & Roe

Number of animals in the sample tested250 500 1000

0.081 0.007 0.000

at a Given Prevalence

Page 11: Sample Sizes - Cannon & Roe

Sample size for estimation of disease prevalence

Level of confidence 0.99

0.1 0.05 0.01 0.0010.01 7 26 657 65685

Finite Population CorrectionPopulaton size: 1600

7 26 466 1562

Expected Prevalence

Desired accuracy

B8
Enter the approximate expected prevalence.
D13
Enter the population size.
Page 12: Sample Sizes - Cannon & Roe

BINOMIAL CONFIDENCE LIMITSEstimating disease prevalence Upper Lower Two sided

Limits: 0.99 0.01 98%

Sample size: 134 Number positive: 5

Proportion positive: 0.0373134 Proportion of the pop'n sampled: 0.5

Limits* Corrected CritbinomUpper 0.095 0.067 11Lower 0.010 0.007 1

N/C Not calculated

*See Agresti A, and Coull BA (1998) Approximate is Better than "Exact" for Interval Estimation of Binomial Proportions. The American Statistician 52:(2) 119-126

D7
Insert the number sampled here
H7
How many were "positive" in your sample?
I9
What proportion of the total population was sampled?