A Proposal to estimate the roaming–dog Total in an urban area through a PPSWOR spatial sampling with sample size greater than two

Ornella Giambalvo; Roverto Sichera

Ornella Giambalvo
Roverto Sichera

Keywords: PPS sampling, design–based variance estimation, inclusion probability, MonteCarlo estimation, roaming dogs

Abstract

Dogs roaming in urban areas constitute an issue for public order, hygiene and health. Proper planning of actions for health and security control and allocation of financial funds require the knowledge of the roaming–dog–population size in a given urban area. Unfortunately, a reliable statistical procedure aimed to measure such population is not available yet in literature. This paper presents a simple, reproducible survey sampling procedure to estimate the
number of roaming dogs in an urban area through the description of a real study carried out on a restricted area of the city of Palermo, in southern Italy. A sample of areas is drawn by means of a drawn–by–drawn spatial sampling with probabilities proportional to the size and without replacement (PPSWOR). As inclusion probabilities are not available in closed form, they are estimated by Monte Carlo approach, which is of simple implementation and permits design-based variance estimation even when first-order inclusion probabilities are unknown.

References

Barabesi, L., L. Fattorini, and G. Ridolfi (1997). “Two-phase surveys of elusive populations”.
In: Proceedings of the Statistic Canada Symposium. Vol. 97, pp. 285–287.
Bennett, G. (1962). “Probability Inequalities for the Sum of Independent Random Variables”. In:
Journal of the American Statistical Association 57.297, pp. 33–45. doi: 10.1080/01621459.
1962.10482149.
Fattorini, L. (2006). “Applying the Horvitz: Thompson Criterion in Complex Designs: A Computer-
Intensive Perspective for Estimating Inclusion Probabilities”. In: Biometrika 93.2, pp. 269–
278.
Fattorini, L. (2009). “An adaptive algorithm for estimating inclusion probabilities and perform-
ing the Horvitz–Thompson criterion in complex designs”. In: Computational Statistics 24.4,
p. 623. doi: 10.1007/s00180-009-0149-9.
Grafström, A. and Y. Tillé (2012). “Doubly balanced spatial sampling with spreading and
restitution of auxiliary totals”. In: Environmetrics 24.2, pp. 120–131. doi: 10.1002/env.
2194.
Horvitz, D. G. and D. J. Thompson (1952). “A generalization of sampling without replacement
from a finite universe”. In: Journal of the American statistical Association 47.260, pp. 663–
685.
Italian Ministry of Health (2006). Rendicontazione annuale sul randagismo.
Jackman, J. and A. N. Rowan (2007). “Free–roaming dogs in developing countries: The benefits
of capture, neuter, and return programs”. In: Washington, DC: Humane Society Press.,
pp. 55–78.
Thompson, M. E. and C. Wu (2008). “Simulation-based randomized systematic PPS sampling
under substitution of units”. In: Survey Methodology 34.1, pp. 3–10.
World Organization for Animal Health (2010). Stray dog population control.
World Society for the Protection of Animals (2010). Surveying roaming dog populations: guide-
lines on methodology.