Mindfulness and Bias: Evidence and Gap Maps

1 Moderators:

I relied on the coding in the Google Sheet for all variables. The coding of the moderators is described below.

1.1 EGMs are completed for the following moderators:

• int_type: Mindfulness Intervention Type (1= Mindfulness, 2= Langer Mindfulness, 3= LKM, 4= Compassion, 5= ACT, 6= MBSR, 7= Hybrid, 8= “meditators” unspecified)

• duration (1= < 1 hour, 2= < 1 week of training, 3= multi-week)

• in_out_grp: Type of target group: 1 = ingroup (e.g., internalized oppression) 2= Outgroup (e.g., prejudice)

• target_identity (1=race/ethnicity, 2= gender, 3= social class/the poor, 4= age, 5= body image/ weight, 6= physical/ mental disability/ addiction, 7= multicultural, 8= sex orientation, 9= religion, 10= other)

• Country/race: (1=US White, 2= US diverse or unspecified, 3= UK White, 4= UK diverse or unspecified, 5= Other Europe, 6= Middle Eastern, 7= Asia/India, 8= Australia/NZ, 9= Multinational, 10= Canada

1.2 EGMs pending for the following moderators:

1.2.1 Categorial variables: Need help coding these

These need to be re-coded to correspond with the categories in Doris’s EGM prototype, as per below:

• Sample population: 1 = General population (students); 2 = Helping professionals (teachers/counsellors/medical providers)

• Mindfulness measures: 1=FFMQ, 2=Langer Mindfulness Scale, 3=MAAS, 4=Toronto Mindfulness Scale, 5=Freiburg Mindfulness Inventory, 6=KIMS, 7=Other

1.2.2 Continous variables: Need to calculate these

Need to calculate means and SDs for these variables across all studies for each of the outcomes, separately for intervention studies and correlational effects.

Gender: Percent female in the sample

Age of the sample

2 EGM of interventions effects

library(tidyverse)
d <- read_csv("Mindfulness and Prejudice_180520_for analysis.xlsx - Mindfulness and Prejudice_08042.csv")

d$sample_country <- d$`REC (1=US White, 2= US diverse or unspecified, 3= UK White, 4= UK diverse or unspecified, 5= Other Europe, 6= Middle Eastern, 7= Asia/India, 8= Australia/NZ, 9= Multinational, 10= Canada`
d$int_type <- as.numeric(d$int_type)

selected <- d %>% 
  select(`(-I) Implicit`:`(+O) Physical Helping`) 
categories <- d %>% 
  select(int_type, duration, in_out_grp, target_identity, sample_country) 

2.1 Number of studies

studies <- list()
for (i in names(categories)){
  for (var in names(selected)) {
    d %>% 
      filter(.data[[var]]==1 & .data$study_design==2) %>% 
      count(.data[[i]], .data$cluster) %>%
      group_by(.data[[i]]) %>% 
      tally() -> studies[[i]][[var]]
  }
}
boom <- list()
for (i in names(studies)){
  studies[[i]] %>% 
    reduce(full_join, by = i) %>% 
    arrange(.data[[i]]) %>% 
    rename_at(vars(names(.[-1])), ~ names(selected)) -> boom[[i]]
}
boom

$int_type

$duration

$in_out_grp

$target_identity

$sample_country

NA

2.2 Number of effect sizes

effectsizes <- list()
for (i in names(categories)){
  for (var in names(selected)) {
    d %>% 
      filter(.data[[var]]==1 & .data$study_design==2) %>% 
      count(.data[[i]]) -> effectsizes[[i]][[var]]
  } 
}
boom <- list()
for (i in names(effectsizes)){
  effectsizes[[i]] %>% 
    reduce(full_join, by = i) %>% 
    arrange(.data[[i]]) %>% 
    rename_at(vars(names(.[-1])), ~ names(selected)) -> boom[[i]]
}
boom

$int_type

$duration

$in_out_grp

$target_identity

$sample_country

NA

3 EGMs of correlations

Note: Some of the correlational effects are from intervention studies.

3.1 Number of studies

studies <- list()
for (i in names(categories)){
  for (var in names(selected)) {
    d %>% 
      filter(.data[[var]]==1 & .data$study_design==0) %>% 
      count(.data[[i]], .data$cluster) %>%
      group_by(.data[[i]]) %>% 
      tally() -> studies[[i]][[var]]
  }
}
boom <- list()
for (i in names(studies)){
  studies[[i]] %>% 
    reduce(full_join, by = i) %>% 
    arrange(.data[[i]]) %>% 
    rename_at(vars(names(.[-1])), ~ names(selected)) -> boom[[i]]
}
boom

$int_type

$duration

$in_out_grp

$target_identity

$sample_country

NA

3.2 Number of effect sizes

effectsizes <- list()
for (i in names(categories)){
  for (var in names(selected)) {
    d %>% 
      filter(.data[[var]]==1 & .data$study_design==0) %>% 
      count(.data[[i]]) -> effectsizes[[i]][[var]]
  } 
}
boom <- list()
for (i in names(effectsizes)){
  effectsizes[[i]] %>% 
    reduce(full_join, by = i) %>% 
    arrange(.data[[i]]) %>% 
    rename_at(vars(names(.[-1])), ~ names(selected)) -> boom[[i]]
}
boom

$int_type

$duration

$in_out_grp

$target_identity

$sample_country

NA