Deidentified Frozen Seafood Dataset

About Dataset

This dataset contains de-identified data on testing for infection in consignments of frozen seafood where importation into Australia was attempted. The data is from an approximately four-year period between 2015 and 2022. Each row represents a batch, where a batch is either a consignment, or a subdivision of a consignment. A sample of 13 bags was selected from each consignment. Each bag contains on average 40 items. A pool of 5 items was selected from each bag. Pools were then tested for the presence of infection. The result is a number of positive tests out of 13 pools tested.

Deidentification steps included:

Tonnage was rounded to the nearest whole number and truncated above at 20.

One year with relatively few observations was grouped with an adjacent year.

Country of export, country of harvest, consignment and year were all randomly scrambled to arbitrary code numbers. The year numbers are not in chronological order.

Countries of harvest or export with less than 20 batches have been removed. Consignments with more than two batches have been removed.

Variable Definitions

Variable Name

Values

Description

batch.scrambled

1, …, 2850

Randomised batch identifier number. The data set contains one row per batch.

consignment.scrambled

1, …, 2680

Randomised consignment identifier number. Consignments contain either one or two batches. (Consignments with more batches have been excluded from this dataset.)

country.export.scrambled

1, …, 5

Randomised country of export identifier number. (Countries with relatively few batches have been excluded.)

country.harvest.scrambled

1, …, 8

Randomised country of export identifier number. (Countries with relatively few batches have been excluded.)

tonnage

0, …, 20

Weight of batch cargo in tonnes (rounded to the nearest whole tonne and truncated at 20).

year.scrambled

1, 2, 3, 4

Randomised year of importation (note: not in temporal order)

numPos

0, …, 10

Number of positive pooled tests (out of 13)

Author Field

I am the uploader of this de-identified dataset, which is published with the permission of the data owner.