I am looking at the effects of seawater salinity levels on fertilization rates. Salinity levels are set so they are treated as factors.
Call:
glm(formula = cbind(fertilized, unfertilized) ~ salinity,
family = "quasibinomial", data = fert.sal)
Deviance Residuals:
Min 1Q Median 3Q Max
-5.2459 -0.3600 -0.0002 1.0971 4.9201
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -22.31 1929.46 -0.012 0.991
salinity20 17.96 1929.46 0.009 0.993
salinity25 23.26 1929.46 0.012 0.990
salinity30 24.36 1929.46 0.013 0.990
salinity33 27.42 1929.46 0.014 0.989
(Dispersion parameter for quasibinomial family taken to be 3.102026)
Null deviance: 7450.93 on 74 degrees of freedom
Residual deviance: 222.59 on 70 degrees of freedom
AIC: NA
Number of Fisher Scoring iterations: 17
When viewed graphically, differences in fertilization rates between salinity levels are quite obvious (as also shown by estimates); however, standard errors are strangely all similar and P-values non-significant. Everyone is welcome to point out where errors might be. Thanks
