Table 3 Summary of statistical comparison of mixed-effects linear models with and without lionfish biomass as an additional fixed effect predictor for each of four fish groups.

From: No change in key reef fish herbivores or reef fisher yields in Barbados a decade after the lionfish invasion

Fish group

Fixed effects component

Df

AIC

logLik

Deviance

Chisq

p

Parrotfishes

Without lionfish

8

 − 5.68

10.84

 − 21.68

  
 

With lionfish

9

 − 4.22

11.11

 − 22.22

0.54

0.462

Surgeonfishes

Without lionfish

8

19.19

 − 1.59

3.19

  
 

With lionfish

9

20.50

 − 1.25

2.50

0.69

0.406

Wrasses

Without lionfish

8

 − 35.32

25.66

 − 51.32

  
 

With lionfish

9

 − 33.32

25.66

 − 51.32

0.01

0.934

Damselfishes

Without lionfish

8

 − 21.86

18.93

 − 37.86

  
 

With lionfish

9

 − 28.93

23.47

 − 46.93

9.08

0.003

  1. All models used average fish biomass per transect during a given reef survey for a given fish group as response variable and period (pre- and post-invasion) and season (winter, summer and fall) (and their two-way interaction) as fixed effect factors. Site was treated as a random effect. Df- degrees of freedom; AIC- Akaike Information Criteria value; logLik- log likelihood ratio value; Chisq- Chi square statistic; p-p-value (based on one degree of freedom). Bold font indicates that adding lionfish biomass to the model significantly (p < 0.05) improved the model. All biomass values were square root transformed prior to the analyses.
Back to article page