Table 2.

Characteristics of participant speech by sex (means, standard deviations, and ranges)

	Females	Males	Effects
Participant speech behavior
Part A
Social word frequency per 1000 words	127.04 (24.8) 59–161	110.59 (24.3) 59–171	p < .001** est: − .13
Friend word frequency per 1000 words	7.72 (4.92) 2–23	6.13 (3.81) 0–15	p = .01* est: − .22
Family word frequency per 1000 words	8.88 (5.84) 0–26	8.32 (5.61) 0–28	p = .18 est: − .11
Part B
Total length of conversation (min)	21.86 (4.24) 14.3–38.7	21.64 (6.67) 10.8–52.3	p = .97 est: − .05
Total time speaking (min)	7.66 (3.29) 2.9–15.4	6.96 (3.59) 0.8–19.3	p = .15 est: − .87
Word count	1218.72 (545.97) 318–2420	1024.68 (544.47) 132–3091	p = .11 est: − 150.9
Characters per word	3.83 (0.11) 3.6–4.1	3.77 (0.15) 3.4–4.1	p = .18 est: < .001
Type-token ratio	0.40 (0.08) .30–.63	0.37 (0.07) .24–.60	p = .05 est: < .001

Effect sizes for GLM are reported as unstandardized effects (estimates [94]). The final GLM model [glm(variable ~ age.z + IQ.z + sex, data = lang.par, family = ‘poisson’)] accounts for age (centered), IQ (centered), and examines sex as primary predictor variable. Part A includes the primary variables of interest. Part B includes additional variables used to characterize the language sample. Effect of sex is significant p < .01