English      Chinese      French      German      Indonesian      Spanish
Search
Pooled Prevalence

Pooled prevalence for variable pool size and perfect tests


Paste data in the space below. Data columns can be in any order but must include columns as specified above, depending on the data format you have selected (2 columns for raw data and 3 columns for summary data). A header row specifying column names must also be included.

Download example data


This method estimates prevalence and confidence limits for variable pool sizes and assumed 100% test sensitivity and specificity. See the User Guide or Williams & Moffitt (2001) for more details. See demonstration analysis.

Input values

Required inputs are the desired level of confidence in the prevalence estimate and the format for data submission and 2 or 3 columns of testing data:

  • raw data - submit 2 columns of data, one labeled "PoolSize" containing the pool size for each pool tested and the other labeled "Result" containing the test result for each pool, coded as 1 = positive and 0 = negative;
  • summary data - submit 3 columns of data, one labeled "PoolSize" containing a list of pool sizes tested, one labeled "Pools" containing the number of pools tested for each pool size and one labeled "PosPools" containing the number of pools that were positive for each pool size.

Note: Columns may be in any order but must be labeled as above, including capitalisation.

The desired confidence level must be a decimal number >0 and <1 (for example, 0.99 = 99% or 0.95 = 95%).

Outputs

Outputs from this program are presented as a point estimate of prevalence plus upper and lower confidence limits for the specified level of confidence.