Table 2.

Parallel mediation analyzes testing indirect effects of anxiety and perceived competence (PC) on the association between Impostor Phenomenon (IP) and willingness to communicate (WTC).

	b	SE	95% CI	p
Baseline model (R2 = 0.029)
IP ➔ WTC	−0.01	0.003	[−0.014, −0.001]	0.017
Mediation model (R2 = 0.398)
Direct paths
IP ➔ WTC	0.00	0.003	[−0.006, 0.007]	0.867
IP ➔ Anxiety	0.04	0.004	[0.030, 0.048]	<0.001
IP ➔ PC	−0.01	0.003	[−0.017, −0.003]	0.003
Anxiety ➔ WTC	−0.08	0.046	[−0.175, 0.005]	0.066
PC ➔ WTC	0.52	0.059	[0.407, 0.638]	<0.001
Indirect paths
IP ➔ Anxiety ➔ WTC	0.00	0.002	[−0.007, 0.000]	0.078
IP ➔ PC ➔ WTC	−0.01	0.002	[−0.009, −0.002]	0.008
Total indirect effect	−0.01	0.003	[−0.014, −0.004]	0.001
Total effect	−0.01	0.003	[−0.014, −0.001]	0.017

N = 400. Bold coefficients: bootstrapped 95% CI (k = 5,000 samples) excludes zero. The χ² model test indicated that the mediation model was superior to a null model assuming no associations among the study variables (χ² = 498.09, df = 9, p < 0.001). The covariate gender had an effect of b = −0.26 (β = −0.13, p = 0.011) in the baseline model and b = −0.20 (95% CI [−0.343, −0.055], SE = 0.073, p = 0.007) in the mediation model.