Abbreviations: FoMO = trait FoMO; FCP = FoMO Cue Presence; ACP = Alcohol Cue Presence

library(InteractionPoweR)

1. Alcohol Craving Model

1a. 2-way Interaction between trait FoMO and FCP

#ACQ: FoMOxFCP 2-way interaction
power_test = power_interaction_r2(
  alpha = 0.05,                  # p-value
  N = seq(200,600,by = 50),      # sample size
  r.x1.y = .28,                   # correlation between FoMO and ACQ
  r.x2.y =  .06,                  # correlation between FCP and ACQ
  r.x1.x2 = .03,                  # correlation between FoMO and FCP
  r.x1x2.y = .17)                # correlation between FoMOxFCP and ACQ
## [1] "Checking for errors in inputs..."

Suggests N=259 for FoMOxFCP Interaction to be Power >= .80

power_estimate(power_data = power_test, # output from power_interaction()
               x = "N",        # the variable we want a precise number for
               power_target = 0.8         # the power we want to achieve 
)
## [1] 258.3721
plot_power_curve(power_data = power_test, # output from power_interaction()
                 power_target = .8,       # the power we want to achieve 
                 x = "N",                 # x-axis
                 group = "r.x1x2.y"       # grouping variable
)

1b. 2-way Interaction between trait FoMO and ACP

power_test = power_interaction_r2(
  alpha = 0.05,                  # p-value
  N = seq(200,700,by = 50),      # sample size
  r.x1.y = .28,                   # correlation between FoMO and ACQ
  r.x2.y =  .12,                  # correlation between ACP and ACQ
  r.x1.x2 = .06,                  # correlation between FoMO and ACP
  r.x1x2.y = .20)                # correlation between FoMOxACP and ACQ
## [1] "Checking for errors in inputs..."

Suggests N=250 for FoMOxFCP Interaction to be Power >= .80

power_estimate(power_data = power_test, # output from power_interaction()
               x = "N",        # the variable we want a precise number for
               power_target = 0.8         # the power we want to achieve 
)
## Warning in power_estimate(power_data = power_test, x = "N", power_target = 0.8):
## Parameter value is out of data range
## [1] NA
plot_power_curve(power_data = power_test, # output from power_interaction()
                 power_target = .8,       # the power we want to achieve 
                 x = "N",                 # x-axis
                 group = "r.x1x2.y"       # grouping variable
)

1c. 2-way Interaction between trait FCP and ACP

power_test = power_interaction_r2(
  alpha = 0.05,                  # p-value
  N = seq(200,1000,by = 50),      # sample size
  r.x1.y = .06,                   # correlation between FCP and ACQ
  r.x2.y =  .12,                  # correlation between ACP and ACQ
  r.x1.x2 = .00,                  # correlation between FoMO and ACP
  r.x1x2.y = .11)                # correlation between FoMOxACP and ACQ
## [1] "Checking for errors in inputs..."

Suggests N=640 for FoMOxFCP Interaction to be Power >= .80

power_estimate(power_data = power_test, # output from power_interaction()
               x = "N",        # the variable we want a precise number for
               power_target = 0.8         # the power we want to achieve 
)
## [1] 639.2942
plot_power_curve(power_data = power_test, # output from power_interaction()
                 power_target = .8,       # the power we want to achieve 
                 x = "N",                 # x-axis
                 group = "r.x1x2.y"       # grouping variable
)

1. Drinking Likelihood Model

2a. 2-way Interaction between trait FoMO and FCP

power_test = power_interaction_r2(
  alpha = 0.05,                  # p-value
  N = seq(10,600,by = 100),      # sample size
  r.x1.y = .02,                   # correlation between FoMO and DL
  r.x2.y =  .44,                  # correlation between FCP and DL
  r.x1.x2 = .03,                  # correlation between FoMO and FCP
  r.x1x2.y = .43)                # correlation between FoMOxFCP and ACQ
## [1] "Checking for errors in inputs..."

Suggests N=70 for FoMOxFCP Interaction to be Power >= .80

power_estimate(power_data = power_test, # output from power_interaction()
               x = "N",        # the variable we want a precise number for
               power_target = 0.8         # the power we want to achieve 
)
## [1] 78.17723
plot_power_curve(power_data = power_test, # output from power_interaction()
                 power_target = .8,       # the power we want to achieve 
                 x = "N",                 # x-axis
                 group = "r.x1x2.y"       # grouping variable
)
## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : Chernobyl! trL>n 6

## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## parametric, : Chernobyl! trL>n 6
## Warning in sqrt(sum.squares/one.delta): NaNs produced
## Warning in stats::qt(level/2 + 0.5, pred$df): NaNs produced
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf

2b. 2-way Interaction between trait FoMO and ACP

power_test = power_interaction_r2(
  alpha = 0.05,                  # p-value
  N = seq(10,700,by = 100),      # sample size
  r.x1.y = .02,                   # correlation between FoMO and DL
  r.x2.y =  .35,                  # correlation between ACP and DL
  r.x1.x2 = .00,                  # correlation between FoMO and ACP
  r.x1x2.y = .34)                # correlation between FoMOxACP and DL
## [1] "Checking for errors in inputs..."

Suggests N=87 for FoMOxFCP Interaction to be Power >= .80

power_estimate(power_data = power_test, # output from power_interaction()
               x = "N",        # the variable we want a precise number for
               power_target = 0.8         # the power we want to achieve 
)
## [1] 86.17656
plot_power_curve(power_data = power_test, # output from power_interaction()
                 power_target = .8,       # the power we want to achieve 
                 x = "N",                 # x-axis
                 group = "r.x1x2.y"       # grouping variable
)

2c. 2-way Interaction between trait FoMO and ACP

power_test = power_interaction_r2(
  alpha = 0.05,                  # p-value
  N = seq(10,800,by = 50),      # sample size
  r.x1.y = .44,                   # correlation between FCP and DL
  r.x2.y =  .35,                  # correlation between ACP and DL
  r.x1.x2 = .00,                  # correlation between FCP and ACP
  r.x1x2.y = .51)                # correlation between FoMOxACP and ACQ
## [1] "Checking for errors in inputs..."

Suggests N=45 for FoMOxFCP Interaction to be Power >= .80

power_estimate(power_data = power_test, # output from power_interaction()
               x = "N",        # the variable we want a precise number for
               power_target = 0.8         # the power we want to achieve 
)
## [1] 45.46613
plot_power_curve(power_data = power_test, # output from power_interaction()
                 power_target = .8,       # the power we want to achieve 
                 x = "N",                 # x-axis
                 group = "r.x1x2.y"       # grouping variable
)