Long-held beliefs about territoriality in animals, including primates, holds that for a population to be territorial, it must be competing for limited resources and those resources must be defendable (after Brown, 1964). Research into primate territoriality has suggested that in addition to limited, defendable resources, to be territorial a group of primates must be able to reach the edge of their homerange at least once a day (Mitani & Rodman, 1979). To explain exceptions to this rule, namely primate populations that can reach the edge of their homeranges daily and yet are not territorial, researchers have examined additional life history variables such as feeding regime and group feeding weight in an attempt to explain the existence of or lack of territoriality.
For this homework, I used a dataset that I prepared a few years ago for an analysis of life history variables and male care. I chose to look at the predictor variables social organization and group size to see how they related to the outcome variable of territoriality (a binary yes/no variable) (N = 271 populations).
After looking at group size and realizing that for this analysis, I would need to divide the group sized into discreet bins and there isn’t any reasearch advising how to divide them, I decided to look at social organization and diet as the two predictor variables. Research has found diet or level of folivory to predict territoriality. I added a diet dataset to the life history dataset.
Brown, J.L. (1964). The evolution of diversity in avian territorial systems. Wilson Bull. 76, 160-169. ; Mitani, J. C., & Rodman, P. S. (1979). Territoriality: the relation of ranging pattern and home range size to defendability, with an analysis of territoriality among primate species. Behavioral ecology and sociobiology, 5(3), 241-251.
#x2=15.699, df=5, p=.00776
chisq.test(Lifehistory$socialorg, Lifehistory$territorial)## Warning in chisq.test(Lifehistory$socialorg, Lifehistory$territorial): Chi-
## squared approximation may be incorrect##
## Pearson's Chi-squared test
##
## data: Lifehistory$socialorg and Lifehistory$territorial
## X-squared = 15.699, df = 5, p-value = 0.00776#x2=70.257, df=10, p=3.955e-11
chisq.test(LHDiet$socialorg,LHDiet$diet)## Warning in chisq.test(LHDiet$socialorg, LHDiet$diet): Chi-squared approximation
## may be incorrect##
## Pearson's Chi-squared test
##
## data: LHDiet$socialorg and LHDiet$diet
## X-squared = 70.257, df = 10, p-value = 3.955e-11#x2=36.777, df=3, p=5.13e-08
chisq.test(LHDiet$territorial,LHDiet$diet)## Warning in chisq.test(LHDiet$territorial, LHDiet$diet): Chi-squared
## approximation may be incorrect##
## Pearson's Chi-squared test
##
## data: LHDiet$territorial and LHDiet$diet
## X-squared = 36.777, df = 3, p-value = 5.13e-08LHD<-glm(territorial~socialorg+diet,data=LHDiet,family=binomial)
summary(LHD)##
## Call:
## glm(formula = territorial ~ socialorg + diet, family = binomial,
## data = LHDiet)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.7172 -1.2245 0.7584 0.8171 1.1553
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.06330 0.30906 -0.205 0.837723
## socialorg 0.05781 0.06356 0.910 0.363084
## dietfrugivore 0.87318 0.23238 3.758 0.000172 ***
## dietinsectivore 17.43366 576.63335 0.030 0.975881
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 502.32 on 416 degrees of freedom
## Residual deviance: 452.80 on 413 degrees of freedom
## (210 observations deleted due to missingness)
## AIC: 460.8
##
## Number of Fisher Scoring iterations: 16library(gtsummary)
LHD%>%tbl_regression(exponentiate=TRUE)## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred| Characteristic | OR1 | 95% CI1 | p-value |
|---|---|---|---|
| socialorg | 1.06 | 0.94, 1.20 | 0.4 |
| diet | |||
| folivore | — | — | |
| frugivore | 2.39 | 1.52, 3.78 | <0.001 |
| insectivore | 37,268,610 | 0.00, 30,507,192,011,480,819,624,749,228,604,351,022,297,082,999,450,882,490,095,164,545,823,074,146,006,578,407,336,614,821,888 | >0.9 |
|
1
OR = Odds Ratio, CI = Confidence Interval
|
|||
The interesting case I wanted to examine for this homework was pair-living frugivorous groups since gibbons, the primates I study, are most commonly pair living and are frugivorous. According to the model, pair living frugivore groups are 70% more likely to be territorial. Fruit is a resource that primates have the potential to defend against conspecifics and pair living individuals may have a higher incentive to defend their resources against conspecifics than larger groups where not all individuals can or will participate in defense (Willems et al., 2015). Frugivores of any social grouping are more likely to be territorial than folivores. For instance, pair living folivores are only 46% likely to be territorial. It is not possible to defend leaves against conspecifics since they are so widely distributed, and is probably not worth the cost of defense.
Willems, E. P., Arseneau, T. J. M., Schleuning, X., & van Schaik, C. P. (2015). Communal range defence in primates as a public goods dilemma. Philosophical Transactions of the Royal Society B: Biological Sciences, 370(1683), 20150003.
library(emmeans)
rg<-ref_grid(LHD)
marg_logit<-emmeans(object=rg,specs=c("socialorg","diet"),type="response")
marg_logit## socialorg diet prob SE df asymp.LCL asymp.UCL
## 3.88489 folivore 0.540233 0.04517533 Inf 0.451356 0.626623
## 3.88489 frugivore 0.737780 0.02797243 Inf 0.679412 0.788825
## 3.88489 insectivore 1.000000 0.00001317 Inf 0.000000 1.000000
##
## Confidence level used: 0.95
## Intervals are back-transformed from the logit scale