Determination of rational sample size in parasitological studies by bootstrap method
Sampling of field material in parasitological studies is a highly time consuming procedure, which successful implementation also requires substantial financial and human resources. Therefore it is important to determine the rational sample size as a balance between the informative of the data from the small samples and the excess costs for collecting the unnecessary data. Due to aggregated distribution of parasites in populations of hosts it is difficult to calculate the confidence intervals of the statistical characteristics using the methods of classical statistics. In such situations it is need to use the non-parametric methods and resampling. The study offers practical recommendations to determine the rational sample size for calculation the interval estimates of the statistical parameters and the precision using the Bag of Little Bootstraps. The precision is defined as a confidence interval such that the estimate of the mean should be within some value of the true mean. The approach is illustrated on the example of monogenean parasites of Ligophorus llewellyni and L. pilengas from the so-iuy mullet. The initial data set included 224 elements for each parasite species. The results showed that the level of precision and parameter of aggregation are strongly affected by the sample size. It was found that the width of the confidence interval was equal or less of the empirical mean for samples more than 45 elements. The mean abundance is systematically underestimated for samples with 20–35 individuals. The small samples (n=10) have led to the unreliable estimates. The similar results were obtained in studies of Marques and Cabral. The proposed here approach will relief the sampling plan design and will help researchers to define the rational sample size and the precision level for the estimated mean abundance in parasitological studies.
(1) Israel, G. D. Determining Sample Size; PEOD6; 1992.
(2) Курочкин, Ю. В. Методическое пособие по паразитологическому инспектированию морских рыб; ТИНРО: Владивосток, 1979.
(3) Karandinos, M. G. Optimum Sample Size and Comments on Some Published Formulae. Bull. Entomol. Soc. Am. 1976, 22 (4), 417–421 DOI: 10.1093/besa/22.4.417.
(4) Opit, G. P.; Throne, J. E.; Flinn, P. W. Sampling Plans for the Psocids Liposcelis entomophila and Liposcelis decolor (Psocoptera: Liposcelididae) in Steel Bins Containing Wheat. J. Econ. Entomol. 2009, 102 (4), 1714–1722 DOI: 10.1603/029.102.0440.
(5) Shvydka, S.; Sarabeev, V.; Estruch, V. D.; Cadarso-Suárez, C. Optimum sample size to estimate mean parasite abundance in fish parasite surveys. Helminthologia 2018, 55 (1),
52–59 DOI: 10.1515/helm-2017-0054.
(6) Anderson, R. M.; Gordon, D. M. Processes influencing the distribution of parasite numbers within host populations with special emphasis on parasite-induced host mortalities. Parasitology 1982, 85 (Pt 2), 373–398.
(7) Shaw, D. J.; Dobson, A. P. Patterns of macroparasite abundance and aggregation in wildlife populations: a quantitative review. Parasitology 1995, 111 Suppl, S111-27.
(8) Poulin, R. Explaining variability in parasite aggregation levels among host samples. Parasitology 2013, 140, 541–546 DOI: 10.1017/S0031182012002053.
(9) Rózsa, L.; Reiczigel, J.; Majoros, G. Quantifying parasites in samples of hosts. J. Parasitol. 2000, 86 (2), 228–232 DOI: 10.1645/0022-3395(2000)086[0228:QPISOH]2.0.CO;2.
(10) Marques, J. F.; Cabral, H. N. Effects of sample size on fish parasite prevalence, mean abundance and mean intensity estimates. J. Appl. Ichthyol. 2007, 23 (2), 158–162 DOI: 10.1111/j.1439-0426.2006.00823.x.
(11) Efron, B.; Tibshirani, R. J. An introduction to the bootstrap. Monographs on Statistics and Applied Probability, No. 57. Chapman and Hall, London, 436 p; Chapman and Hall: London, 1993; Vol. 57.
(12) Kleiner, A.; Talwalkar, A.; Sarkar, P.; Jordan, M. I. A scalable bootstrap for massive data. J. R. Stat. Soc. Ser. B (Statistical Methodol. 2014, 76 (4), 795–816 DOI: 10.1111/rssb.12050.