Measuring speech privacy

The acoustics researcher Manuj Yadav modelled acoustic dissatisfaction in 28 offices, combining room acoustic measurements, sound recorded while people were working and a survey of 349 occupants. The model has two latent factors, lack of privacy and noise disturbance, and lack of privacy contributed 25% more to acoustic dissatisfaction than noise disturbance did. Metrics based on the decay of speech sound pressure level predicted both factors better than distraction distance did, which is the measure ISO 3382-3 is built around, and the trends for those metrics ran against what the standard leads one to expect. Yadav attributes part of that to the cross-sectional design.

The building researcher Toby Cheung surveyed 666 occupants of seven Green Mark certified air-conditioned buildings in Singapore on 18 parameters. 78% were satisfied with their workspace overall. The three worst-rated were sound privacy, with 42% dissatisfied, personal control at 32 and temperature at 30. The three best were flexibility of the dress code at 86% satisfied, electric lighting at 84 and cleanliness at 82. Satisfaction with cleanliness had the largest effect on overall satisfaction of any of the 18 parameters.

The acoustics researcher Pyoung Jik Lee and Jeon separated being able to understand speech from finding it hard to listen. 20 subjects rated 240 Korean sentences at source-receiver distances of 8, 16 and 24 metres, with the spatial decay rate of speech modelled at 4, 8 and 12 dB and the speech level at four metres varied from 43 to 57 dB, against a fixed background of 30 dBA. At eight metres, intelligibility scores did not move with either quantity. Listening difficulty ratings changed significantly with both. At the distance where colleagues actually sit, a specification changes the listening effort and leaves the intelligibility score where it was.

The engineering researcher Samin Marzban analysed 1,403 post-occupancy surveys from 14 open-plan offices in Australia, New Zealand and Hong Kong, ten of them WELL certified and four not. Satisfaction with visual privacy and with access to the outdoor environment were the two strongest predictors of productivity. Layout and interior design predicted creativity. The five offices with the highest satisfaction on layout, environmental quality, privacy and connection to the outdoors were all certified, and there was no significant difference in health between certified and uncertified buildings. More than 20% were dissatisfied with the physical environment either way.

A survey by Di Blasio compared 1,078 workers in shared offices of two to five people with those in open-plan offices of more than five, asking about the noise colleagues make: conversation, telephone calls, laughter. About 70% in both types named difficulty concentrating as the main effect, with no significant difference between them. Everything else differed. Annoyance, self-rated performance and mental health symptoms were all worse in open plan. The people most likely to be annoyed were women, those in rooms with more than 20 occupants, and those in southern cities without acoustic treatment. The coping strategies split by room type too. In open plan people put on headphones; in shared offices they took a break, moved, closed the door or went home.

Marzban surveyed 1,403 people and Di Blasio 1,078, and neither took a measurement in a room. Cheung surveyed 666 people in seven buildings. Yadav combined instrument readings with 349 surveys across 28 offices. Lee and Jeon worked in a laboratory with 20 subjects.