Skip to main content
. 2017 May 22;114(23):5964–5969. doi: 10.1073/pnas.1613117114

Table S3.

Social belonging and engineering self-efficacy mediate the relation between mentor condition and engineering career aspirations

Independent variable (X) Multiple mediators (M) Dependent variable (Y) a paths b paths c path a*b path indirect effects for ΔM1, ΔM2, ΔM3 a*b 95% CI using ΔM1, ΔM2, ΔM3
X → ΔM1 ΔM1 → ΔY X → ΔY
X → ΔM2 ΔM2 → ΔY LLCI ULCI
X → ΔM3 ΔM3 → ΔY
Female mentor vs. no mentor Belonging (M1) Engineering career aspirations 0.58 (0.27)* 0.41 (0.07)*** −1.19 (0.98) 0.24 (0.12)* 0.03 0.49
Self-efficacy (M2) 0.65 (0.25)** 3.95 (0.12)*** 2.58 (0.98)** 0.65 4.49
Threat/challenge (M3) −0.26 (0.14) 6.94 (0.18)*** −1.77 (0.93) −3.59 0.05
Female vs. male mentor Belonging (M1) Engineering career aspirations 0.55 (0.26)* 0.41 (0.08)*** −0.87 (0.96) 0.22 (0.11)* 0.02 0.46
Self-efficacy (M2) 0.32 (0.24) 3.95 (0.12)*** 1.27 (0.96) −0.61 3.14
Threat/challenge (M3) −0.10 (0.13) 6.94 (0.18)*** −0.71 (0.91) −2.50 1.08
Male mentor vs. no mentor Belonging (M1) Engineering career aspirations 0.03 (0.27) 0.41 (0.07)*** −0.32 (0.22) 0.01 (0.11) −0.21 0.24
Self-efficacy (M2) 0.33 (0.25) 3.95 (0.12)*** 1.31 (0.98) −0.61 3.24
Threat/challenge (M3) −0.15 (0.13) 6.94 (0.18)*** −1.06 (0.93) −2.90 0.75

Note: P < 0.10, *P < 0.05, **P < 0.01, and ***P < 0.001; values for a, b, c, and a*b paths are unstandardized slopes with SEs in parentheses.