221k views
3 votes
Suppose that you decide to use the capture - recapture method to estimate the size of a network. The size of the first sample (the "marked sample") is 100 nodes, the size of the second sample is 150 nodes, the intersection of the two samples is 100 nodes. Which of the following is a valid reason to be skeptical of this estimate? You can assume that the size of the network is much larger than 150 nodes.

User Perrako
by
7.8k points

1 Answer

1 vote

Final answer:

The skepticism in the capture-recapture result is due to a discrepancy in the data given: the intersection should not be equal to the entire marked sample if the assumption of random mixing is met.

Step-by-step explanation:

To understand why the given scenario from the capture-recapture method might be viewed skeptically, one must consider the methodology used. According to the capture-recapture principle, the first sample size is crucial in estimating the total population. In our example, there seems to be an error as the intersection between the first and second samples cannot logically be equal to the entire first sample. With a first sample size of 100 nodes and a second sample size of 150 nodes, it isn't feasible for all 100 nodes of the first sample to be found in the second sample without any additional unmarked nodes. This either suggests a mistake in data collection, or that the assumption of random mixing has not been met.

Moreover, the mark and recapture method works under the assumption that marked individuals mix evenly with the unmarked population and that they have no greater or lesser chance of recapture than the unmarked individuals. However, if a significant number of individuals learn to avoid or seek capture, this could bias the results and provide an inaccurate population estimate. The example provided improperly estimates the population size as 400, which based on the described intersection is more likely a simple mathematical error, or an indication that the sample sizes are not representative, or that other assumptions of the model are violated.

User Geraud Mathe
by
7.9k points