So I'm afraid it's just once again the bad quality of the SZ when it comes to scientific research...Looking on Google, it seems to have been reported like this everywhere, but the abstract (and the actual information in the SZ) say something else:
http://adc.bmj.com/cgi/content/abstract/adc.2..."Participants Children aged 8-12 plus who stuttered (monolingual and bilingual) and fluent bilingual controls (FB). ...
The sample of 317 children had 69 bilinguals (prevalence rate of bilingualism in the stuttering sample was 21.8%). ... "
The full text costs eleven euro, so I don't know what the full results were, but the abstract says they were actually comparing kids who learned a second language simultaneously with English from birth, and those who learned English after the age of 5.
The abstract does not say whether the rate of 21.8% is comparable to the rate of bilinguals in the population of London as a whole (wouldn't be surprised if it was, to be honest), or if they actually deliberately chose to have a higher number of bilinguals in the sample, which would make sense as they were studying bilinguals. (Another possibility could be that they chose a random sample of children referred to speech therapists, but who knows if bilinguals are actually referred more often, as a) parents of bilinguals are more likely to worry about their speech and b) speech therapists might be more likely to consider bilinguals at risk.
The result is that of the bilinguals, the ones who learn the second language later stutter less and for a shorter time. I wonder if in the study they distinguished between those who
officially learned English at school, but who grew up in England and had been passively picking it up since birth, and those who arrived in England at age five.
Quote from another study from 2000:
"The only known survey to date on the topic was carried out in the 30s by
Travis, Johnson and Shover (1937). Travis et al. surveyed 4827 children aged between four and seventeen years (mean age 8.54 years) and found that there was a higher percentage of children who stutter in the bilingual population surveyed (2.8%) than in the monolingual population (1.8%)."
The 2000 study doesn't sound wonderfully representative, but it concludes:
"The current study showed that there is no difference between monolingual and bilingual speakers in term of their likelihood of having stuttering in their life."
http://www.speech.psychol.ucl.ac.uk/PAPERS/PD...Again, this figure may be affected by reporting rates.