David Zelený

en:data:morse

# Rothkopf's experiment with Morse codes

## Source of data

Rothkopf (1957), data used here were distilled from library xgobi1)

## Description of the dataset

The aim of the study was to show, which Morse codes are the most easy to be confused. Each Morse code is a set of short (0.05 sec) and/or long (0.15 sec) beeps. Experiment was joined by 598 respondents, and Rothkopf played each of them different combinations of Morse codes, separated by short pause. Each respondent should quickly decide, whether the played codes were the same or different. The result is confusion matrix, in which each number represents number of cases when respondents consider the given combination of codes to be different. The higher the number, the easier are the codes recognisable.

The example of the matrix morse.dist is below:

 A B C D E F G .- 0 167 169 159 180 164 163 -. . . 167 0 96 79 163 95 139 -.-. 169 96 0 141 166 115 137 -.. 159 79 141 0 172 136 100 . 180 163 166 172 0 183 182 ..-. 164 95 115 136 183 0 156 –. 163 139 137 100 182 156 0

I created also another dataset, which two attributes of each Morse code: the length (number of beeps), and the ration between the number of long and short beeps. The example of matrix morse.attr is as below:

 length ratio.short.long 2 0.5 4 0.75 4 0.5 3 0.6666667 1 1 4 0.75 3 0.3333333

morse.dist <- read.delim ('https://raw.githubusercontent.com/zdealveindy/anadat-r/master/data/morsecodes-dist.txt', row.names = 1, head = T)