Perceived quality features of aircraft sounds: An analysis of the measurement characteristics of a newly created semantic differential Martin Schütte a, * , Uwe Müller b , Stephan Sandrock c , Barbara Griefahn a , Catherine Lavandier d , Benoit Barbot d a Leibniz-Institut für Arbeitsforschung an der Technischen Universität Dortmund, Ardeystraße 67, D-44139 Dortmund, Germany b Deutsches Zentrum für Luft- und Raumfahrt e.V., Linder Höhe, D-51147 Köln, Germany c Institut für angewandte Arbeitswissenschaft (IfaA), Uerdinger Str. 56, D-40474 Düsseldorf, Germany d Laboratoire MRTE, Université de Cergy Pontoise, 5 mail Gay Lussac, Neuville sur Oise, 95031 Cergy Pontoise Cedex, France article info Article history: Received 5 February 2008 Received in revised form 16 January 2009 Accepted 23 January 2009 Available online 10 March 2009 Keywords: Semantic differential Sound quality feature G-theory Departure sounds Arrival sounds abstract The increasing air traffic in the past has led to the fact that and more and more people were affected by aircraft noise. Since the potential of reducing noise level of current aircraft will be limited in the future after recent developed techniques have been realized in practice. Therefore, additional efforts are necessary to improve the sound quality of aircraft noise and give hints to aircraft manufactures concerning components that have to be acoustically optimized. The present study describes the development of an item list for a semantic differential (SD) that is appropriate for the determination of different perceptive features of aircraft sound quality and focuses on the estimation of the reliability of this new instrument. The selected 10 SD items have relation to the sound components emitted by the different technical devices of airplane engines. The analyses of the measurement characteristics of the different items were based on G-theory. This method presupposes data possessing interval-scale-level characteristics. Therefore, the used rating scales were examined with respect to this attribute. The results verified that the data fulfill this precondition. The results gave evidence that the judgments show large inter-individual differences. The proportion of variance which can be traced back to the factor ‘‘aircraft sound” amounted to at most one-fifth of total variance indicating that the differences between the aircraft sounds are rather small. Nevertheless, the findings demonstrate that it is possible to measure not all but several decisive aspects with sufficient reli- ability and expense. Ó 2009 Elsevier Ltd. All rights reserved. 1. Introduction The current ‘‘Eurocontrol medium forecast of instrument flight rule movements” predicts that 11.9 million movements can be ex- pected in Europe in 2013. This corresponds to an average annual growth of 3.4% in relation to the reference year 2006 [1]. Even if such predictions are based on many assumptions concerning for instance the economic development, the number of operating low-cost carriers etc., the reported prediction nevertheless shows that air traffic will steadily increase and thereby also the number of people concerned. Correspondingly, effort is necessary to investigate additional means for the acceptance of aircraft noise, since the potential of reducing noise will be limited after the currently developed techni- cal improvements have been realized in practice. One way to re- duce annoyance of people affected by aircraft noise is to remove the most disturbing aircraft sound components and define target sounds for aircraft manufacturers which are perceived as more pleasant [2–6]. A precise description of the quality of aircraft sounds is necessary in order to get more detailed information about possible technical sources bringing about noise components perceived as disturbing. The quality of aircraft sounds is studied here in the frame of the ‘‘sound quality” approach proposed by Blauert and Jekosch [7]. Recently, Jekosch [8] defines ‘‘sound qual- ity” as ‘‘the result of an assessment of the perceived auditory nat- ure of a sound with respect to its desired nature”. The perceived nature depends on the perceived quality features of the sound. These features are the recognisable characteristics that can be ob- served or detected and denominated by listeners. The desired nat- ure of a sound corresponds to the ‘‘totality of features as projected by individual expectations and/or functional requirements and/or social demands”. Accordingly the measurement of sound quality presupposes data concerning the perceived nature of a sound which can be collected via the semantic differential method. The semantic differential (SD) approach, already used in psy- cho-acoustic research for the measurement of particular auditory characteristics [e.g. 9–11], offers the possibility to characterize 0003-682X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.apacoust.2009.01.004 * Corresponding author. Tel.: +49 (0) 2311084377. E-mail address: schuette@ifado.de (M. Schütte). Applied Acoustics 70 (2009) 903–914 Contents lists available at ScienceDirect Applied Acoustics journal homepage: www.elsevier.com/locate/apacoust