# Format and Combine

### How to use fapply2()

The fapply2() function applies two formats to two different vectors, and combines them when complete. This function can be used to collapse two columns into one. The fapply2() function is convenient to use when preparing your data for reporting. Here is an example:

# Create sample vectors
v1 <- c(27, 43, 22, 56)
v2 <- c(18.24324, 29.05405, 14.86486, 37.83784)

# Create data frame
dat <- data.frame("Counts" = v1, "Percents" = v2)

# Format and Combine
dat$CntPct <- fapply2(dat$Counts, dat$Percents, "%d", "(%.1f%%)") # View results dat # Counts Percents CntPct # 1 27 18.24324 27 (18.2%) # 2 43 29.05405 43 (29.1%) # 3 22 14.86486 22 (14.9%) # 4 56 37.83784 56 (37.8%) ### Use with dplyr The fapply2() function is suitable for use with the dplyr package. Here is the same example as above, but using dplyr mutate() instead of Base R: library(dplyr) # Create sample vectors v1 <- c(27, 43, 22, 56) v2 <- c(18.24324, 29.05405, 14.86486, 37.83784) # Create data frame dat <- data.frame("Counts" = v1, "Percents" = v2) # Format and Combine dat <- dat |> mutate(CntPct = fapply2(dat$Counts, dat$Percents, "%d", "(%.1f%%)")) # View results dat # Counts Percents CntPct # 1 27 18.24324 27 (18.2%) # 2 43 29.05405 43 (29.1%) # 3 22 14.86486 22 (14.9%) # 4 56 37.83784 56 (37.8%) ### Use with datastep The fapply2() function is also compatible with the datastep() function from the libr package. Here is the example again with the datastep(): library(libr) # Create sample vectors v1 <- c(27, 43, 22, 56) v2 <- c(18.24324, 29.05405, 14.86486, 37.83784) # Create data frame dat <- data.frame("Counts" = v1, "Percents" = v2) # Format and Combine dat <- datastep(dat, { CntPct <- fapply2(Counts, Percents, "%d", "(%.1f%%)") }) # View results dat # Counts Percents CntPct # 1 27 18.24324 27 (18.2%) # 2 43 29.05405 43 (29.1%) # 3 22 14.86486 22 (14.9%) # 4 56 37.83784 56 (37.8%) ## Using Assigned Formats Note that fapply2() will use formats assigned to the data frame columns if they are available. Assigning the formats to the columns first can simplify use of the function and promote format reuse. To assign the formats to the columns, use the formats() function, like so: # Create sample vectors v1 <- c(27, 43, 22, 56) v2 <- c(18.24324, 29.05405, 14.86486, 37.83784) # Create data frame dat <- data.frame("Counts" = v1, "Percents" = v2) formats(dat) <- list(Counts = "%d", Percents = "(%.1f%%)") # Format and Combine - Formats already assigned dat$CntPct <- fapply2(dat$Counts, dat$Percents)

# View results
dat
#   Counts Percents     CntPct
# 1     27 18.24324 27 (18.2%)
# 2     43 29.05405 43 (29.1%)
# 3     22 14.86486 22 (14.9%)
# 4     56 37.83784 56 (37.8%)

## Format Catalog with datastep()

The ability to use any formats assigned to the columns makes the fapply() function very useful when combined with format catalogs and the datastep() function. When the format catalog is assigned to the datastep(), it will automatically assign the formats in the catalog to any corresponding columns on the input data frame. This feature allows you to quickly assign saved formats to a new dataset, and use those formats to combine columns in the desired way. Observe:

library(libr)

# Create sample vectors
grp <- c("Group1", "Group2", "Group3", "Group4")
v1 <- c(27, 43, 22, 56)
v2 <- c(18.24324, 29.05405, 14.86486, 37.83784)
v3 <- c(5.24883, 8.83724, 2.39483, 9.12542)
v4 <- c(2.97632, 3.32845, 0.29784, 4.22156)

# Create data frame
dat <- data.frame("Group" = grp, "Counts" = v1, "Percents" = v2,
"Mean" = v3, "SD" = v4)

# View original data
dat
#    Group Counts Percents    Mean      SD
# 1 Group1     27 18.24324 5.24883 2.97632
# 2 Group2     43 29.05405 8.83724 3.32845
# 3 Group3     22 14.86486 2.39483 0.29784
# 4 Group4     56 37.83784 9.12542 4.22156

# Create format catalog
fc <- fcat(Counts = "%d", Percents = "(%03.1f%%)",
Mean = "%.1f", SD = "(%04.2f)")

# Format and Combine columns using Format catalog
dat2 <- datastep(dat, format = fc,
keep = v(Group, CntPct, MeanSD),
{

CntPct <- fapply2(Counts, Percents)
MeanSD <- fapply2(Mean, SD)

})
# View results
dat2
#    Group     CntPct     MeanSD
# 1 Group1 27 (18.2%) 5.2 (2.98)
# 2 Group2 43 (29.1%) 8.8 (3.33)
# 3 Group3 22 (14.9%) 2.4 (0.30)
# 4 Group4 56 (37.8%) 9.1 (4.22)

The above technique points to a method for sharing formats between programs and ensuring that statistical results are formatted consistently across programs.

## Use of Other Format Types

Note that the fapply2() function will accept any type of format supported by the fmtr package. That means you can use numeric formats, date formats, vector lookups, user-defined formats, and vectorized functions. The combination of these format types allows you to format and combine data in a powerful way that will enhance the impact of your analysis.