Miscommunication in Commercial Aviation: The Role of Accent, Speech Rate, Information Density, and Politeness Markers

ABSTRACT

Objective

This study is specifically designed to examine the effect of accent (native or non-native English sounding), rate of speech, complexity of transmission (information density), and politeness markers on commercial pilot (mis)communication.

Background

Aviation accident reports often cite miscommunication as a contributing factor. Anecdotal reports from pilots, along with limited empirical studies on pilot communication, further confirm that miscommunication remains a problem.

Method

Approximately 250 ATC-Pilot transmissions from each of four international airports: Kingsford Smith, Sydney, Australia (YSSY); Hong Kong International Airport (VHHH); Los Angeles International Airport, USA (KLAX); and Haneda, Tokyo, Japan (RJTT) were analyzed. Pilot communication errors were compared between the four locations based on pilot and ATC accent alignment, rate of speech, number and order of items in transmission, and politeness markers.

Results

Native English-sounding pilots committed more errors than accented pilots. Alignment of pilot and ATC language background reduced communication errors, but not when native English speakers were involved. Longer messages increased the number of communication errors. Politeness markers did not affect communication and pilots committed fewer errors when the readback order was not scrambled.

Conclusion

Communication errors still occur in ATC-Pilot radio communication. These errors appear more common with native English sounding pilots than accented pilots. Hence, the origin of the problem appears to stem from proficiency in the lingua franca of Aviation English, rather than with the English language.

Disclosure Statement

No potential conflict of interest was reported by the authors.

Notes

1. “Pilots were classified as ‘native English sounding’ if no foreign accent could be detected and if they were flying with an airline registered in a country where the official language is English (e.g., Australia, New Zealand, the United Kingdom, or the United States of America). Pilots were classified as ‘accented English’ if a non-native English accent could be detected and if the aircraft was registered in a country where English is not one of the official languages (e.g., Japan, Korea, China, Chile). Pilots with an English sounding accent, but who were on an aircraft that was registered in a country where English is not the official language or is one of several official languages (e.g., Hong Kong, Singapore, Malaysia, Fiji) were excluded, as it was less certain whether their native language was English.” (Wu et al., 2019, p. 3).

2. Coder 1 (AD; native Sinhala speaker proficient in English – IELTS Overall Score 8.0) coded all transmissions. Coder 2 (BM; native English speaker) and coder 3 (DE; native French speaker proficient in English and English Language Proficiency assessor for CASA) coded 20% of YSSY and KLAX transmissions, while coder 4 (DL: native Cantonese speaker proficient in English) coded all VHHH transmissions and coder 5 (YN; native Japanese speaker proficient in English) coded all of RJTT transmissions. Inter-rater agreement between coders 1, 2 and 3 was 100%. Following two repeated coding exercises between coders 1 and 4, 100% agreement was achieved (92% agreement on first coding exercise). Following two repeated coding exercises between coders 1 and 5, 99% agreement was achieved (85% on first coding exercise).