randfindR is a package for the programming language R in development for the analysis of randomness in sequences of distinct options (e.g., numbers or letters). randfindR can help you determine with several indices how systematic a sequence of options is according to different benchmarks of randomness. Its focus lies on the detection of patterns that are typically found in human generated sequences.

Once accepted (!), you can install the released version of randfindR from CRAN with (this is not yet possible):

#install.packages("randfindR")

And the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("TImA97/randfindR")

This quick guide is meant to show you the indices of randomness that can be computed with randfindR. As of now, randfindR offers 15 different indices of randomness that can be easily computed as can be seen in the following explanation.

All functions take as first argument a vector (your random sequence). In most cases you also have to specify the number of possible options a sequence can contain. The following index indicates how regular a sequence is with a value of 0 indicating no regularity and a value of 1 maximum regularity.

library(randfindR)

reg_index(c(1,2,1,2,1,1,2,2,2), options = 2)
#> [1] 0.4654518

The optionsargument is important because many indices will reflect the fact that one or several options were omitted in a sequence. Let’s take the example from above, but assume this time that there are three possible options, i.e., one option was completely omitted!

reg_index(c(1,2,1,2,1,1,2,2,2), options = 3)
#> [1] 0.5773503

As you can see, the resulting index is now higher, indicating increased regularity.

Some indices have additional arguments or do not have the optionsargument. More information and examples on how to use all indices can be found in the documentation.

If you want all available indices or a list of indices of your choice to be computed, you can use the all_rand() function. This function will by default return all indices of randomness for a given vector or data.frame (in the latter case indices are computed using a row-wise format).

Let’s take the example from above again, but this time call the all_rand() function:

round(all_rand(c(1,2,1,2,1,1,2,2,2), options = 2), 2)
#>       digram_rep      repetitions           series    cluster_ratio 
#>             5.00             3.00            12.00             0.69 
#>       null_score        reg_index       runs_index     coupon_score 
#>             0.00             0.47             0.25             2.33 
#>        gap_score      poker_score        rng_index       rng2_index 
#>             2.00             1.00             0.59             0.55 
#>         tp_index redundancy_index       var_digits 
#>            85.71             0.89             0.25

But you can also explicitely specify the indices that should be computed:

all_rand(c(1,2,1,2,1,1,2,2,2), options = 2, indices = c("reg_index", "rng_index"))
#> reg_index rng_index 
#> 0.4654518 0.5869447

You can just enter all relevant arguments into the all_rand() function and it will take care of everything else and insert your specified arguments in the right places!

The following list contains all indices that are implemented in randfindR. More indices may be added in the future.

• digram_rep
• repetitions
• series
• cluster_ratio
• null_score
• reg_index
• runs_index
• coupon_score
• gap_score
• poker_score
• rng_index
• rng2_index
• tp_index
• redundancy_index
• var_digits

Details on how to use these algorithms can be found in the documentation by entering ?index_nameinto your R-console.

The indices implemented in this package are based on the following publications:

Towse, J.N., Neil, D. Analyzing human random generation behavior: A review of methods used and a computer program for describing performance. Behavior Research Methods, Instruments, & Computers 30, 583–591 (1998). https://doi.org/10.3758/BF03209475

Ginsburg N, Karpiuk P. Random Generation: Analysis of the Responses. Perceptual and Motor Skills. 1994;79(3):1059-1067. https://doi.org/10.2466/pms.1994.79.3.1059

Skliar, Osvaldo, Ricardo E. Monge, Guillermo Oviedo, and Víctor Medina. 2009. Indices of regularity and indices of randomness for m-ary strings. https://doi.org/10.15517/rmta.v16i1.1418