Content area
Full text
Introduction
Background
South Africa is a country with a distinctly unique amalgamation of languages, dialects, cultures and linguistic communities (Swanepoel, 2006). Audiologists in South Africa face a predicament when conducting speech audiometry tests on English Second Language (ESL) speakers. The majority of audiologists registered in South Africa are English- or Afrikaans-speaking individuals (Penn, Frankel, Watermeyer & Muller, 2009) who are unlikely to speak an African language (Khoza, Ramma, Mophosho & Moroka, 2008). This is in stark contrast to the demographics of South Africa, where the home language of 77% of the population is an African language (Statistics South Africa, 2012). Only 9.6% of South Africans speak English as their first language (Statistics South Africa, 2012).
However, a cycle perpetuates wherein English remains the dominant language in South African society and the majority of the population (between 32% and 69%; there are broad ranges in the statistical estimates) uses English as one of their multiple languages, particularly in urban areas (Minow, 2010). English is used extensively in South Africa in education, law, government, news broadcasts, business, commerce, the army and parliamentary debate (Alexander, 2000; Minow, 2010). English is the language of learning and teaching (LoLT) for more than 90% of South African learners (De Wet, 2002). It is within this context that audiologists in South Africa are aiming to provide quality audiological assessments to the population.
Audiology uses various measures to determine hearing function, including pure-tone testing and speech audiometry. The measurement of the speech recognition threshold (SRT) is a speech audiometry test that relies on the participant's recognition of familiar spondaic words from a closed set. SRT is defined as the lowest intensity at which the spondaic (spondee) words are identified correctly 50% of the time (Martin & Clark, 2003). Spondaic words are made up of bi-syllabic words, typically nouns, with equal stress placed on each syllable, such as ‘sunset’ (Gelfand, 2009). In 1947, Hudgins, Hawkins, Karlin and Stevens (1947) first developed a list of spondaic words at the Harvard Psycho-Acoustic Laboratories (PAL). Following the development of the PAL wordlists, the words were evaluated for homogeneity of audibility, within ±2 dB for mean recognition thresholds (Gelfand, 2009). Six words were excluded from the original 42 spondaic words because of a lack of homogeneity (McArdle &...