library(vazul)
library(dplyr)
library(stats)In certain studies, variable names should be masked to prevent researcher bias. Examples can include exploratory factor analysis, network analysis, etc. The vazul package provides means to mask variable names in a dataset, ensuring that analyses can be conducted without preconceived notions about the variables.
library(vazul)
library(dplyr)
library(stats)In this example, we will use the williams dataset from the {vazul} package.
data("williams", package = "vazul")
head(williams)
#> # A tibble: 6 × 25
#> subject SexUnres_1 SexUnres_2 SexUnres_3 SexUnres_4_r SexUnres_5_r Impuls_1
#> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 A30MP4LXV… 5 3 2 3 2 3
#> 2 A16X5FB3H… 7 7 7 4 2 6
#> 3 A1E9D1OT9… 2 4 6 3 3 3
#> 4 A16FPOYD7… 5 4 5 3 3 4
#> 5 A11NOTVHW… 5 5 6 3 3 5
#> 6 A3TDR6MXS… 5 6 6 3 2 5
#> # ℹ 18 more variables: Impuls_2_r <dbl>, Impul_3_r <dbl>, Opport_1 <dbl>,
#> # Opport_2 <dbl>, Opport_3 <dbl>, Opport_4 <dbl>, Opport_5 <dbl>,
#> # Opport_6_r <dbl>, InvEdu_1_r <dbl>, InvEdu_2_r <dbl>, InvChild_1 <dbl>,
#> # InvChild_2_r <dbl>, age <dbl>, gender <dbl>, ecology <chr>,
#> # duration_in_seconds <dbl>, attention_1 <dbl>, attention_2 <dbl>
glimpse(williams)
#> Rows: 112
#> Columns: 25
#> $ subject <chr> "A30MP4LXV4MIFD", "A16X5FB3HAFCKN", "A1E9D1OT9VJYD…
#> $ SexUnres_1 <dbl> 5, 7, 2, 5, 5, 5, 5, 4, 6, 4, 5, 3, 5, 6, 3, 6, 2,…
#> $ SexUnres_2 <dbl> 3, 7, 4, 4, 5, 6, 6, 6, 5, 4, 5, 2, 5, 3, 2, 3, 1,…
#> $ SexUnres_3 <dbl> 2, 7, 6, 5, 6, 6, 5, 5, 6, 4, 7, 3, 5, 3, 3, 6, 3,…
#> $ SexUnres_4_r <dbl> 3, 4, 3, 3, 3, 3, 3, 2, 3, 1, 3, 3, 3, 2, 3, 3, 4,…
#> $ SexUnres_5_r <dbl> 2, 2, 3, 3, 3, 2, 4, 3, 2, 1, 2, 2, 4, 6, 3, 1, 3,…
#> $ Impuls_1 <dbl> 3, 6, 3, 4, 5, 5, 5, 6, 6, 1, 6, 2, 4, 7, 5, 4, 5,…
#> $ Impuls_2_r <dbl> 3, 2, 3, 4, 2, 3, 3, 2, 3, 1, 3, 2, 3, 6, 3, 3, 7,…
#> $ Impul_3_r <dbl> 2, 1, 3, 4, 4, 3, 2, 3, 2, 1, 4, 1, 4, 5, 3, 3, 5,…
#> $ Opport_1 <dbl> 1, 5, 3, 4, 4, 4, 5, 5, 5, 1, 7, 3, 5, 7, 5, 2, 6,…
#> $ Opport_2 <dbl> 2, 7, 5, 4, 6, 3, 5, 4, 4, 4, 6, 4, 6, 5, 2, 3, 4,…
#> $ Opport_3 <dbl> 2, 7, 5, 4, 4, 4, 6, 5, 5, 1, 7, 3, 5, 5, 5, 3, 1,…
#> $ Opport_4 <dbl> 3, 6, 3, 5, 4, 6, 5, 5, 4, 4, 6, 3, 6, 4, 4, 3, 3,…
#> $ Opport_5 <dbl> 1, 6, 3, 4, 5, 4, 6, 6, 4, 1, 6, 3, 6, 5, 5, 4, 6,…
#> $ Opport_6_r <dbl> 3, 2, 3, 4, 3, 3, 2, 2, 3, 2, 1, 3, 1, 6, 4, 3, 4,…
#> $ InvEdu_1_r <dbl> 2, 3, 2, 3, 4, 3, 3, 3, 3, 2, 3, 4, 4, 4, 3, 1, 6,…
#> $ InvEdu_2_r <dbl> 3, 2, 3, 4, 3, 4, 2, 3, 4, 1, 2, 2, 3, 7, 4, 1, 3,…
#> $ InvChild_1 <dbl> 2, 5, 6, 5, 5, 5, 6, 5, 6, 1, 5, 2, 4, 4, 5, 2, 6,…
#> $ InvChild_2_r <dbl> 3, 2, 3, 4, 3, 4, 2, 4, 3, 2, 4, 2, 3, 7, 3, 2, 6,…
#> $ age <dbl> 34, 30, 40, 35, 26, 33, 33, 30, 48, 33, 40, 39, 25…
#> $ gender <dbl> 1, 1, 1, 2, 2, 2, 2, 1, 2, 2, 1, 1, 2, 1, 1, 2, 1,…
#> $ ecology <chr> "Hopeful", "Desperate", "Desperate", "Hopeful", "D…
#> $ duration_in_seconds <dbl> 164, 100, 47, 31, 40, 32, 98, 34, 32, 87, 71, 103,…
#> $ attention_1 <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 0,…
#> $ attention_2 <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1,…We will apply masking to the variables related to life history strategy, which are prefixed with SexUnres, Impuls, Opport, InvEdu, and InvChild. We’ll mask each variable group separately with randomized letter prefixes (e.g., C_01, A_01, E_01, etc.). This way, variables within the same original scale keep a common prefix for the analysis, but analysts won’t know which prefix corresponds to which original scale due to the randomization.
set.seed(84)
# Sample 5 random letters for the 5 variable groups
random_prefixes <- paste0(sample(LETTERS, 5), "_")
masked_williams <-
williams |>
mask_names(starts_with("SexUnres"), prefix = random_prefixes[1]) |>
mask_names(starts_with("Impul"), prefix = random_prefixes[2]) |>
mask_names(starts_with("Opport"), prefix = random_prefixes[3]) |>
mask_names(starts_with("InvEdu"), prefix = random_prefixes[4]) |>
mask_names(starts_with("InvChild"), prefix = random_prefixes[5])
# Show the randomized prefixes used (but not which corresponds to which)
sort(unique(sub("_.*", "_", grep("^[A-Z]_", names(masked_williams), value = TRUE))))
#> [1] "L_" "P_" "S_" "V_" "Y_"We can now perform an exploratory factor analysis (EFA) on the masked variables. Since the variable names are masked with randomized prefixes, we won’t know which original variables correspond to which factor, thus preventing bias in interpreting the results.
set.seed(123)
efa_blind <-
masked_williams |>
select(matches("^[A-Z]_")) |>
factanal(factors = 5, rotation = "varimax")
# Get the loadings of the EFA on the masked data
efa_blind |>
loadings() |>
print(cutoff = 0.3, sort = TRUE)
#>
#> Loadings:
#> Factor1 Factor2 Factor3 Factor4 Factor5
#> S_01 0.749 0.331
#> S_03 0.701
#> S_05 0.666 0.472
#> V_01 0.762
#> P_01 0.870
#> P_03 0.893
#> P_04 0.758
#> P_05 0.853
#> P_06 0.834
#> L_01 0.766
#> S_02 0.567 0.425
#> S_04 0.735 0.335
#> V_02 0.867
#> V_03 0.746
#> P_02 0.715
#> Y_01 0.839
#> Y_02 0.699
#> L_02 0.852
#>
#> Factor1 Factor2 Factor3 Factor4 Factor5
#> SS loadings 6.398 4.815 0.717 0.345 0.284
#> Proportion Var 0.355 0.267 0.040 0.019 0.016
#> Cumulative Var 0.355 0.623 0.663 0.682 0.698The loading table shows that factors are not necessarily loading to their original category. By using masking, researchers may be able to make decisions without being biased on the variable names.
Applying the same analysis on the original dataset reveals the names of the variables. Please note that the loadings may differ slightly due to the randomness in the factor analysis process.
keep_suffixesSome of the williams columns end in _r, marking items that are reverse-scored (e.g. SexUnres_4_r, Impuls_2_r). That suffix is not itself sensitive information to hide — it is a fixed analysis-relevant tag that a researcher needs to see in order to reverse-code the item correctly before scoring, even while the rest of the name stays masked. By default, mask_names() masks the suffix away along with everything else, so SexUnres_4_r becomes an opaque C_02 with no indication that it needs to be reverse-coded.
The keep_suffixes argument preserves one or more literal suffixes verbatim in the masked name, so the reverse-coding information survives masking:
set.seed(84)
masked_with_suffix <-
williams |>
mask_names(starts_with("SexUnres"), prefix = "C_", keep_suffixes = "_r")
masked_with_suffix |>
select(matches("^C_")) |>
names()
#> [1] "C_01_r" "C_02" "C_03" "C_04" "C_05_r"Notice that the two reverse-scored columns keep their _r suffix (e.g. C_01_r) while the rest of the name is masked as usual, and that masked columns are also sorted alphabetically by their masked name. If more than one supplied suffix could match the same column, the longest match is kept and mask_names() issues a warning naming the affected column(s).
keep_suffixes covers the common case of a single, fixed suffix that should always be preserved literally (not remapped to a randomized label). When both the prefix and the suffix carry meaning that must be masked - but consistently, so the same value always maps to the same masked label - the long-format approach shown next is the right tool instead.
Sometimes both the prefix and the suffix of a set of column names carry meaning that should stay consistent across columns. For example, exp_pre, exp_post, ctl_pre, and ctl_post encode both an experimental condition (exp/ctl) and a measurement occasion (pre/post).
Masking these names directly with repeated mask_names() calls, as above, does not preserve that consistency: each call assigns new labels independently, so there is no guarantee that, say, pre is masked to the same label in the exp_ columns as in the ctl_ columns. In this situation we recommend reshaping the data to long format first, so that the condition and occasion information become values in their own columns rather than fragments of the column names. mask_variables() can then mask those columns directly, which guarantees that a given value (e.g. "exp" or "pre") is always mapped to the same masked label everywhere it occurs. If a wide format is required for the confirmatory analysis, the masked data can be reshaped back afterward.
library(tidyr)
wide_demo <- data.frame(
id = 1:4,
exp_pre = c(10, 12, 9, 11),
exp_post = c(15, 14, 13, 16),
ctl_pre = c(8, 9, 10, 7),
ctl_post = c(9, 10, 11, 8)
)
wide_demo
#> id exp_pre exp_post ctl_pre ctl_post
#> 1 1 10 15 8 9
#> 2 2 12 14 9 10
#> 3 3 9 13 10 11
#> 4 4 11 16 7 8# 1) Reshape to long format, splitting the column names into
# a `condition` and a `time` variable
long_demo <- wide_demo |>
pivot_longer(
cols = -id,
names_to = c("condition", "time"),
names_sep = "_"
)
long_demo
#> # A tibble: 16 × 4
#> id condition time value
#> <int> <chr> <chr> <dbl>
#> 1 1 exp pre 10
#> 2 1 exp post 15
#> 3 1 ctl pre 8
#> 4 1 ctl post 9
#> 5 2 exp pre 12
#> 6 2 exp post 14
#> 7 2 ctl pre 9
#> 8 2 ctl post 10
#> 9 3 exp pre 9
#> 10 3 exp post 13
#> 11 3 ctl pre 10
#> 12 3 ctl post 11
#> 13 4 exp pre 11
#> 14 4 exp post 16
#> 15 4 ctl pre 7
#> 16 4 ctl post 8# 2) Mask the condition and time variables. Each variable keeps its own,
# internally consistent mapping (every "exp" becomes the same masked
# label, every "pre" becomes the same masked label, and so on).
set.seed(2024)
long_masked <- long_demo |>
mask_variables(condition, time)
long_masked |>
count(condition, time)
#> # A tibble: 4 × 3
#> condition time n
#> <chr> <chr> <int>
#> 1 condition_group_01 time_group_01 4
#> 2 condition_group_01 time_group_02 4
#> 3 condition_group_02 time_group_01 4
#> 4 condition_group_02 time_group_02 4# 3) Reshape back to wide format if a wide layout is needed for analysis
long_masked |>
unite(masked_name, condition, time) |>
pivot_wider(names_from = masked_name, values_from = value)
#> # A tibble: 4 × 5
#> id condition_group_02_time_…¹ condition_group_02_t…² condition_group_01_t…³
#> <int> <dbl> <dbl> <dbl>
#> 1 1 10 15 8
#> 2 2 12 14 9
#> 3 3 9 13 10
#> 4 4 11 16 7
#> # ℹ abbreviated names: ¹condition_group_02_time_group_01,
#> # ²condition_group_02_time_group_02, ³condition_group_01_time_group_01
#> # ℹ 1 more variable: condition_group_01_time_group_02 <dbl>Because masking is applied to condition and time as data values rather than as fragments of the column names, the masked labels for "exp"/"ctl" and "pre"/"post" are consistent across every column, both before and after reshaping. We recommend this long-format workflow whenever meaningful prefixes and suffixes both need to be masked; mask_names() remains the simpler choice when only a single, non-overlapping piece of the name needs masking, and mask_names(..., keep_suffixes = ...) covers the case of a fixed literal suffix that should be preserved as-is rather than masked (see above).
set.seed(123)
efa_orig <-
williams |>
select(SexUnres_1:InvChild_2_r) |>
factanal(factors = 5, rotation = "varimax")
# Get the loadings of the EFA on the original data
efa_orig |>
loadings() |>
print(cutoff = 0.3, sort = TRUE)
#>
#> Loadings:
#> Factor1 Factor2 Factor3 Factor4 Factor5
#> SexUnres_1 0.666 0.472
#> SexUnres_2 0.701
#> SexUnres_3 0.749 0.331
#> Impuls_1 0.762
#> Opport_1 0.893
#> Opport_2 0.834
#> Opport_3 0.870
#> Opport_4 0.758
#> Opport_5 0.853
#> InvChild_1 0.766
#> SexUnres_4_r 0.735 0.335
#> SexUnres_5_r 0.567 0.425
#> Impuls_2_r 0.867
#> Impul_3_r 0.746
#> Opport_6_r 0.715
#> InvEdu_1_r 0.699
#> InvEdu_2_r 0.839
#> InvChild_2_r 0.852
#>
#> Factor1 Factor2 Factor3 Factor4 Factor5
#> SS loadings 6.398 4.815 0.717 0.345 0.284
#> Proportion Var 0.355 0.267 0.040 0.019 0.016
#> Cumulative Var 0.355 0.623 0.663 0.682 0.698