Using multiple performance parameters in testing small graphical symbols

Sammanfattning: The main objective of the study was to analyse the understanding of small graphical symbols (pictograms and icons) among different potential user groups around the world using different test indices or parameters. Pilot studies using videophone symbols designed in Japan, USA and England were initially done. The main study utilised videophone symbols studied by the Human Factors Technical Committee (HFTC) of the European Telecommunication Standards Institute (ETSI). Test subjects were from Indonesia, Malaysia, Philippines, Thailand, Sri Lanka, Finland, Sweden, and the USA. The four tests used yielded 10 parameters and used to compare symbol understanding of the different country groups. These were: (1-4) spontaneous identification and cued response hit rates and certainty ratings; 5) misses; 6) false alarms or confusions, 7) missing values; 8) individual symbol preference, 9) symbol set preference; and 10) semantic rating scales. Results from the free recall tests in all countries revealed very poor identification of most symbols in contrast to the cued response test results. In other cases when symbols were recognised for their intended meaning, they could still be easily confused as representing another (wrong) function. Overall, the results also showed the use of several parameters or indices are very helpful in understanding how different groups perceive symbols, as well as to help decide which ones to retain, re-design, or even replace.