Skip to main content

Table 5 Bayesian regressions predicting migration to gambling and gambling spend at follow up

From: A longitudinal replication study testing migration from video game loot boxes to gambling in British Columbia, Canada

Logistic Regression with baseline microtransaction spending as predictors

Variables

B

98.75% CI

SE

OR

98.75% CI OR

BF10

 Constant

− 3.71

− 5.13, − 2.36

0.56

   

 DPM spending

0.02

− 0.15, 0.20

0.07

1.02

0.86, 1.22

0.72

 Loot Box spending

0.24

0.07, 0.40

0.07

1.27

1.07, 1.50

7.46

Logistic regression with baseline RLI scores as predictors

Variables

B

98.75% CI

SE

OR

98.75% CI OR

BF10

 Constant

− 2.06

− 2.50, − 1.63

0.17

   

 RLI

0.52

0.08, 0.97

0.18

1.70

1.09, 2.63

3.19

Linear regression with baseline microtransaction spending as predictors

Variables

B

98.75% CI

SE

  

BF10

 Constant

0.11

− 0.78, 0.99

0.35

   

 DPM spending

0.03

− 0.09, 0.15

0.05

  

1.54

 Loot Box spending

0.22

0.10, 0.33

0.05

  

40.11

Linear regression with baseline RLI scores as predictors

Variables

B

98.75% CI

SE

  

BF10

 Constant

1.44

1.10,1.78

0.14

   

 RLI

0.56

0.21, 0.90

0.14

  

5.79

  1. Model prior distributions were calculated based on data from Brooks & Clark [3]. Expense variables were log base 2 + 1 transformed to reduce positive skew. RLI scores were standardized