function returning observed Cronbach alpha with bootstrap confidence interval around that value

function returning observed Cronbach alpha with bootstrap confidence interval around that value

Usage

getBootCIalpha(
  dat,
  verbose = TRUE,
  na.rm = TRUE,
  nLT20err = TRUE,
  nGT10kerr = TRUE,
  bootReps = 1000,
  conf = 0.95,
  bootCImethod = c("perc", "norm", "basic", "bca")
)

Arguments

dat

data, matrix, data frame or tibble and numeric item data as columns

verbose

logical, if FALSE switches off messages

na.rm

logical which causes function to throw error if there is missing data

nLT20err

logical, throws error if length(na.omit(dat)) < 20, otherwise returns NA for CI

nGT10kerr

logical, throws error if length(na.omit(dat)) > 10k to prevent very long rusn, override with FALSE

bootReps

integer for number of bootstrap replications

conf

numeric confidence interval desired as fraction, e.g. .95

bootCImethod

method of deriving bootstrap CI from "perc", "norm", "basic" or "bca"

Value

list of length 3: obsAlpha, LCLAlpha and UCLAlpha

History/development log

Started before 5.iv.21 Improved help 12.iv.26

See also

Other bootstrap CI functions: getBootCICSC(), getBootCICorr(), getBootCIgrpMeanDiff(), getBootCImean()

Other Chronbach alpha functions: getChronbachAlpha()

Author

Chris Evans

Examples

if (FALSE) { # \dontrun{
### I will create some examples of how the function can be used,
### in base R or tidyverse modes.
### create some data: 200 rows, 10 variables
### Gaussian distribution
set.seed(12345) # just to get same data every time
matrix(rnorm(2000), ncol = 10) -> matDat
### now pipe that matrix into a data frame
matDat %>%
### this next line saves going through the rather
### laborious way of constructing variable
### names that you have to follow if
### you go direct to as_tibble() !
as.data.frame() %>%
as_tibble() %>%
### so now we have a 200x10 tibble (i.e. a
### tidyverse version of a data frame)
### let's make some of these participants
### young and some old
### make the odd numbered ones "young"
mutate(age = if_else(row_number() %% 2 == 1,
                       "young",
                       "old")) %>%
### get variables in a more sensible order
select(age, everything()) -> tibDat
### OK, we have some data, let's use it!

### As with all bootstrap functions, it's wise
### to set the random number generator (RNG) seed
### before running the function to ensure you see
### exactly the same results every time.
### That's a different issue from the one above
### about setting the RNG seed to get the same data
### every time when generating data.  Same principle
### different effect.
###
### this is base R idiom and using the function's default arguments
### to analyse a matrix
getBootCIalpha(matDat)
###
### this is base R analysing a tibble (or data frame)
getBootCIalpha(tibDat[ , -1])
###
### now tidyverse, still entire tibble
tibDat %>%
select(-age) %>%
summarise(alphaCI = list(getBootCIalpha(pick(everything())))) %>%
unnest_wider(alphaCI) # unnest to get the results


### more useful tidyverse, analysing the age groups separately
tibDat %>%
group_by(age) %>%
  summarise(alphaCI = list(getBootCIalpha(pick(everything())))) %>%
  ungroup %>%
  unnest_wider(alphaCI) # unnest again
} # }