Coral-eating seastar Culcita novaeguineae data (volume loss version) — culcitalvolume • lme4

This is the same experiment outlined in culcitalogreg except this data set only contains values where predation occurred, and the original volume as well as volume lost was recorded.

Usage

data("culcitalvolume")

Format

A data frame with 50 observations on the following 7 variables.

ttt

represents the combinations of different symbionts, which the treatment condition is explicitly written in plain language.

volume

describes the size of the coral, measured in cm\(^3\).

predvolume

the amount of volume loss from the coral, measured in cm\(^3\).

crab

describes whether the crab was present in the experiment. n: not present, y: present.

shrimp

describes whether the shrimp was present in the experiment. n: not present, y: present.

block

a numeric variable indicating the experimental block. There are 10 blocks in total, each corresponding to a large, octagonal, flow-through seawater tank approximately 0.5m deep and 2m in diameter.

ttt2

a relabelled version of ttt.

1: no exosymbionts,
2: pair of Alpheus lottini only (Alpheus; shrimp),
3: pair of Trapezia serenei only (Trapezia; crab),
4: pair of Alpheus lottini and pair of Trapezia serenei (‘Alpheus and Trapezia’).

See also

The version which contains whether or not predation occurred, culcitalogreg.

Source

(McKeon et al. 2012)

References

McKeon CS, Stier AC, McIlroy SE, Bolker BM (2012). “Multiple defender effects: synergistic coral defense by mutualist crustaceans.” Oecologia, 169(4), 1095–1103. doi:10.1007/s00442-012-2275-2 .

Examples

## Modifying to create a new response variable
vdata <- transform(culcitalvolume, 
                   propeaten = predvolume/volume,
                   tvol = log(predvolume))
## One-way analysis
(cvm1 <- lmer(tvol ~ ttt2 + (1|block), data = vdata))
#> Linear mixed model fit by REML ['lmerMod']
#> Formula: tvol ~ ttt2 + (1 | block)
#>    Data: vdata
#> REML criterion at convergence: 76.182
#> Random effects:
#>  Groups   Name        Std.Dev.
#>  block    (Intercept) 0.4043  
#>  Residual             0.4263  
#> Number of obs: 50, groups:  block, 10
#> Fixed Effects:
#> (Intercept)         ttt2  
#>      7.0908      -0.3024  
(cvm2 <- lmer(propeaten ~ ttt2 + (1|block), data = vdata))
#> Linear mixed model fit by REML ['lmerMod']
#> Formula: propeaten ~ ttt2 + (1 | block)
#>    Data: vdata
#> REML criterion at convergence: -51.557
#> Random effects:
#>  Groups   Name        Std.Dev.
#>  block    (Intercept) 0.1342  
#>  Residual             0.1083  
#> Number of obs: 50, groups:  block, 10
#> Fixed Effects:
#> (Intercept)         ttt2  
#>     0.42136     -0.09591  
## Two-way analysis 
(cvm3 <- lmer(tvol ~ crab*shrimp + (1|block), data = vdata))
#> Linear mixed model fit by REML ['lmerMod']
#> Formula: tvol ~ crab * shrimp + (1 | block)
#>    Data: vdata
#> REML criterion at convergence: 42.2635
#> Random effects:
#>  Groups   Name        Std.Dev.
#>  block    (Intercept) 0.4650  
#>  Residual             0.2687  
#> Number of obs: 50, groups:  block, 10
#> Fixed Effects:
#>   (Intercept)          craby        shrimpy  craby:shrimpy  
#>       6.71636       -0.09575       -0.37645       -0.82619  
(cvm4 <- lmer(propeaten ~ crab*shrimp + (1|block), data = vdata))
#> Linear mixed model fit by REML ['lmerMod']
#> Formula: propeaten ~ crab * shrimp + (1 | block)
#>    Data: vdata
#> REML criterion at convergence: -46.6655
#> Random effects:
#>  Groups   Name        Std.Dev.
#>  block    (Intercept) 0.1342  
#>  Residual             0.1081  
#> Number of obs: 50, groups:  block, 10
#> Fixed Effects:
#>   (Intercept)          craby        shrimpy  craby:shrimpy  
#>       0.32225       -0.15098       -0.10358       -0.05804