Peer-reviewed veterinary case report
The use of Geographic Information System (GIS) and non-GIS methods to assess the external validity of samples postcollection.
- Journal:
- Journal of veterinary diagnostic investigation : official publication of the American Association of Veterinary Laboratory Diagnosticians, Inc
- Year:
- 2009
- Authors:
- Richardson, Esther et al.
- Affiliation:
- Moorepark Dairy Production Research Centre
Plain-English summary
This study looked at how well we can trust the results from animal health samples taken from different populations. Researchers used both traditional methods, which are designed to ensure a good representation of the population, and other methods that might not be as reliable. They found that even with some limitations, they could still get a reasonable idea of how common certain diseases are in most areas of the country, except for one region where they didn't have enough samples. The study offers useful ways to check the reliability of sample results after they have been collected.
Abstract
External validity is fundamental to veterinary diagnostic investigation, reflecting the accuracy with which sample results can be extrapolated to a broader population of interest. Probability sampling methods are routinely used during the collection of samples from populations, specifically to maximize external validity. Nonprobability sampling (e.g., of blood samples collected as part of routine surveillance programs or laboratory submissions) may provide useful data for further posthoc epidemiological analysis, adding value to the collection and submission of samples. As the sample has already been submitted, the analyst or investigator does not have any control over the sampling methodology, and hence external validity as routine probability sampling methods may not have been employed. The current study describes several Geographic Information System (GIS) and non-GIS methods, applied posthoc, to assess the external validity of samples collected using both probability and nonprobability sampling methods. These methods could equally be employed for inspecting other datasets. Mapping was conducted using ArcView 9.1. Based on this posthoc assessment, results from the random field sample could provide an externally valid, albeit relatively imprecise, estimate of national disease prevalence, of disease prevalence in 3 of the 4 provinces (all but Ulster, in the north and northwest, where sample size was small), and in beef and dairy herds. This study provides practical methods for examining the external validity of samples postcollection.
Find similar cases for your pet
PetCaseFinder finds other peer-reviewed reports of pets with the same symptoms, plus a plain-English summary of what was tried across them.
Search related cases →Original publication: https://pubmed.ncbi.nlm.nih.gov/19737758/