MODELING OVERDISPERSED COUNT DATA: A COMPARATIVE STATISTICAL ANALYSIS OF PERCEIVED CARNIVORE ABUNDANCE
DOI:
https://doi.org/10.69980/kwpx3275Keywords:
overdispersion, count regression, Negative Binomial regression, perceived abundance, carnivore modelingAbstract
In ecological and perception-based research, counts often do not follow the Poisson distribution of equidispersion, which may impact the fit of the model and statistical inference. This study aimed to characterize dispersion patterns in perceived bear, leopard, and wolf abundance, estimate associated predictors, and compare Poisson and Negative Binomial regression performance. A common predictor structure was used to analyze the species-specific count outcomes. Descriptive statistics, dispersion indices, and Poisson and Negative Binomial regression were analyzed. Model performance was assessed by AIC, BIC, Pearson dispersion, and residual diagnostics, and adjusted effects were presented as incidence rate ratios (IRRs) with 95% confidence intervals (CIs). There was strong overdispersion for all species, with dispersion indices of 268.949 for bear, 269.515 for leopard, and 529.372 for wolf. The fit of the models was significantly enhanced using Negative Binomial regression, with the Pearson dispersion values of 1.585, 2.016, and 0.960 obtained for the three models, respectively. The perceived bear abundance was positively correlated with danger score. For leopards, age and family kill-belief were significantly positively correlated, and happiness/pride was significantly negatively correlated, while none of these factors was statistically significant for wolves. Negative Binomial regression was a better fit for the observed count structure, highlighting the need to consider the dispersion of counts and species-specific inference when modeling heterogeneous count outcomes.
References
1.Bektashi, X., Rexhepi, S., & Limani–Bektashi, N. (2022). Dispersion of count data: a case study of Poisson distribution and its limitations. Asian Journal of Probability and Statistics, 19(2), 18-28.
2.Broekhuis, F., Elliot, N. B., Keiwua, K., Koinet, K., Macdonald, D. W., Mogensen, N., ... & Gopalaswamy, A. M. (2021). Resource pulses influence the spatio‐temporal dynamics of a large carnivore population. Ecography, 44(3), 358-369.
3.Campbell, H. (2021). The consequences of checking for zero‐inflation and overdispersion in the analysis of count data. Methods in Ecology and Evolution, 12(4), 665-680.
4.Carvalho, F. J., de Santana, D. G., & Sampaio, M. V. (2020). Modeling Overdispersion, Autocorrelation, and Zero-Inflated Count Data Via Generalized Additive Models and Bayesian Statistics in an Aphid Population Study: Carvalho et al. Neotropical entomology, 49(1), 40-51.
5.Chen, D., & Anderson, C. (2024). Categorical data analysis. arXiv preprint arXiv:2409.02942.
6.Clark, J. S. (2020). Models for ecological data: an introduction. Princeton university press.
7.Cummings, P. (2019). Negative Binomial Regression. In Analysis of Incidence Rates (pp. 271-292). Chapman and Hall/CRC.
8.Ebrahimi Monfared, Z., Mirkarimi, S. H., Mohammadi Kangarani, H., & Soofi, M. (2025). Fear and belief predict perceived carnivore abundance in Golestan National Park, Iran. https://doi.org/10.5281/ZENODO.17469318
9.Elliot, N. B., & Gopalaswamy, A. M. (2017). Toward accurate and precise estimates of lion density. Conservation Biology, 31(4), 934-943.
10.Forsyth, D. M., Ramsey, D. S., & Woodford, L. P. (2019). Estimating abundances, densities, and interspecific associations in a carnivore community. The Journal of Wildlife Management, 83(5), 1090-1102.
11.Harrison, X. A., Donaldson, L., Correa-Cano, M. E., Evans, J., Fisher, D. N., Goodwin, C. E., ... & Inger, R. (2018). A brief introduction to mixed effects modelling and multi-model inference in ecology. PeerJ, 6, e4794.
12.Jiménez, J., Nuñez-Arjona, J. C., Rueda, C., González, L. M., García-Domínguez, F., Muñoz-Igualada, J., & López-Bao, J. V. (2017). Estimating carnivore community structures. Scientific Reports, 7(1), 41036.
13.Kruppa, J., & Hothorn, L. (2021). A comparison study on modeling of clustered and overdispersed count data for multiple comparisons. Journal of Applied Statistics, 48(16), 3220-3232.
14.Li, H., Luo, W., & Baek, E. (2024). Multilevel modeling in single-case studies with zero-inflated and overdispersed count data. Behavior Research Methods, 56(4), 2765-2781.
15.Matthiopoulos, J., Fieberg, J. R., & Aarts, G. (2023). Species-Habitat Associations: Spatial data, predictive models, and ecological insights.
16.Payne, E. H., Gebregziabher, M., Hardin, J. W., Ramakrishnan, V., & Egede, L. E. (2018). An empirical approach to determine a threshold for assessing overdispersion in Poisson and negative binomial models for count data. Communications in Statistics-Simulation and Computation, 47(6), 1722-1738.
17.Payne, E. H., Hardin, J. W., Egede, L. E., Ramakrishnan, V., Selassie, A., & Gebregziabher, M. (2017). Approaches for dealing with various sources of overdispersion in modeling count data: Scale adjustment versus modeling. Statistical methods in medical research, 26(4), 1802-1823.
18.Reangsephet, O., Lisawadi, S., & Ahmed, S. E. (2018, June). Improving estimation of regression parameters in negative binomial regression model. In International Conference on Management Science and Engineering Management (pp. 265-275). Cham: Springer International Publishing.
19.Saputro, D. R. S., Susanti, A., & Pratiwi, N. B. I. (2021, February). The handling of overdispersion on Poisson regression model with the generalized Poisson regression model. In AIP Conference Proceedings (Vol. 2326, No. 1, p. 020026). AIP Publishing LLC.
20.Sidumo, B., Sonono, E., & Takaidza, I. (2024). Count regression and machine learning techniques for zero-inflated overdispersed count data: application to ecological data. Annals of Data Science, 11(3), 803-817.
21.Tang, W., He, H., & Tu, X. M. (2023). Applied categorical and count data analysis. CRC Press.
22.Tinker, M. T., Yee, J. L., Laidre, K. L., Hatfield, B. B., Harris, M. D., Tomoleoni, J. A., ... & Miles, A. K. (2021). Habitat features predict carrying capacity of a recovering marine carnivore. The Journal of Wildlife Management, 85(2), 303-323.
