126 Zsuzsanna Fagyal
32 Thus, contrary to RNM who were able to compute single-factor ANOVAs with
rhythm type as their main factor, such a comparison could not be carried out here.
Levene’s test of equality of error variances was non signi¿ cant (F(9,1359) = 1.632,
p< 0.101).
33 Levene’s test of equality of error variances was non signi¿ cant (F(9,1359) = 1.632,
p< 0.101).
34 Stepwise regression analyses in SPSS do not return correlation coef¿ cients allow-
ing to check for cross-linearity effects. For this reasons, Pearson linear regression
analyses were conducted (see Field 2005).
35 EF speakers were coded with value 1, AF speakers with value 2 for the dichoto-
mous variable “ethnicity.”
36 The possible impact of this co-linearity effect on the model can be evaluated by
other statistical means.
37 Model summaries:
—5 predictors (rhythmic variables only): adjusted r^2 = 0.156, standard error of
estimate = 0.464, F change (5,210) = 7.783, p<0.01.
—6 predictors (rhythmic variables and grade): adjusted r^2 = 0.256, standard error
of estimate = 0.430, F change (6,209) = 13.329, p<0.01.
—6 predictors (rhythmic variables and GPA): adjusted r^2 = 0.253, standard error of
estimate = 0.431, F change (6,209) = 13.130, p<0.01.
—6 predictors (rhythmic variables and speaker): adjusted r^2 = 0.393, standard
error of estimate = 0.410, F change (6,209) = 24.194, p<0.01.
—7 predictors (rhythmic variables, grade, and GPA): adjusted r^2 = 0.324, standard
error of estimate = 0.410, F change (7,208) = 15.731, p<0.01.
—8 predictors (all predictors in): adjusted r^2 = 0.413, standard error of estimate =
0.435, F change (8,207) = 19.890, p<0.01
38 However, such differences existed in archival recordings of speakers born in the
19 th century.
39 In Northern varieties of European French, both syllables can be accented (accent
initial and accent ¿ nal).
40 “Pour se faire entendre dans les groupes des pairs, il faut non seulement parler fort,
il faut aussi parler vite. La rapidité d’élocution de certains adolescents est en ce
sens tout à fait étonnante. Cette vitesse s’applique aussi bien à l’articulation qu’à
l’enchaînement des mots et des phrases, et au rythme des échanges. [. .] A l’inverse,
un locuteur trop lent, qui s’exprime en faisant traîner ses syllabes [... ] s’expose de
façon quasi systématique aux sarcasmes appuyés, aux éclats de rire.. .”
References
Antoine, Fabrice. 1998. Des mots et des oms: verlan, troncation et recyclage formel
dans l’argot contemporain. Cahiers de Lexicologie 72 (1): 41–70.