Skip to main content

Table 2. Predictors of offspring performance, including sucrose responsiveness, sucrose concentration, color training, and starvation resistance

From: Care-giver identity impacts offspring development and performance in an annually social bumble bee

Assay

Analysis

Model

Estimate

Standard error

Test statistic

p-value

Direction

Sucrose responsiveness

Binomial GLM

Yes/No (1/0) ~

Development time

+ random factor “source colony”

+ random factor “nest ID”

0.198

0.105

z = 1.875

0.061

Longer > Shorter

durations

Sucrose concentration

Gamma GLM

Concentration response ~

+ random factor “source colony”

+ random factor “nest ID”

na

na

na

na

na

Color

training

Binomial GLM

Yes/No (1/0) ~

Body size

+ random factor “source colony”

+ random factor “nest ID

2.200

1.289

z = 1.706

0.088

Large > Small

Color

learning

Binomial GLM

Yes/No (1/0) ~

+ random factor “source colony”

+ random factor “nest ID”

na

na

na

na

na

Starvation resistance

Gamma GLM

Response (hours) ~

Group (care-giver)

+ Body size

+ random factor “source colony”

+ random factor “nest ID”

+ assays

− 0.185

0.149

0.071

0.082

z = − 2.602

z = 1.815

0.009 **

0.069

QR > WR

Large > Small

  1. The Generalized Linear Model (GLM) for Binominal distribution was used in modeling data that had two possible outcomes (success or failure); and GLM for Gamma distribution was used in modeling non-integer data. Asterisks indicate statistical significance; significance code: ‘**’ 0.01