Compare with Benchmarks

The compare_benchmark_* functions provide ways to compare benchmark with data like events and times.

Compare an Event with a Benchmark

Note: This function is not yet released. You need to install the development version to use this function. To install the developmental version enter this in your console: remotes::install_github("joe-chelladurai/uxr").

An event is “a thing that happens” (Oxford Learner’s Dictionary).

Example:

10 out of 14 people successfully completed a task on an interface. What is the probability of success above a benchmark of 80%?

You can use the compare_benchmark_event() function to answer this question. This function requires a benchmark value, number of events (in this case successful events), and the total number of events. To pass a benchmark value of 80%, you’ll need to specify it as 0.8. The event and total arguments each take numeric values 1 or above.

The function calculates the success rates and returns a cumulative probability value against the benchmark. This probability value indicates the success rate if we were to use this interface with potential users, given a representative sample.

library(uxr)

compare_benchmark_event(benchmark = 0.80,
                        event = 10,
                        total = 12,
                        event_type = "success",
                        notes = "executive")

-- Compare Event Rate with a Benchmark---------------------------------------------
Executive: Based on the success rate of 84%, the probability that this rate exceeds
a benchmark of 80% is 44%
  term           result  
  event              10  
  total              12  
  benchmark         80%  
  probability     0.442

Compare Time with a Benchmark

Example:

7 people successfully took 60, 53, 70, 42, 62, 43, and 81 seconds to complete a task. What is the probability that time taken to complete this task is above a time of 60 seconds?

You can use the compare_benchmark_time() function to answer this question. This function requires a benchmark value and a list of times. To pass a benchmark value of 60s, you can specify this as 60.

compare_benchmark_time(benchmark = 60,
                       time = c(60, 53, 70, 42, 62, 43, 81),
                       alpha = 0.05)

-- Compare Time with a Benchmark ------------------------------------------------------------
  term          t.result.  
  lower_ci           45.8  
  upper_ci           71.7  
  t                 0.509  
  probability       0.314