Women remain under-represented in the fields of Science, Technology, Engineering and Maths (STEM), constituting only 28% of science researchers worldwide. Research has identified implicit gender-bias as a major barrier to women's progression in these fields. Previous research using the Implicit Association Test (IAT) suggests that individuals studying or working in STEM fields exhibit different levels of implicit male-STEM bias than those in non-STEM fields. The Implicit Relational Assessment Procedure (IRAP), a non-relative measure, was compared with the IAT which allowed us to probe this difference further among STEM and non-STEM students. The IRAP revealed a more detailed and complex picture of gender-STEM bias. All groups demonstrated a significant implicit pro-male-STEM bias. However, there was also evidence of a pro-female-STEM bias, significant only among female STEM students. A number of correlations were observed between the explicit measure and the IRAP. The presence of a pro-female-STEM bias has implications for the development of interventions. If this relational response can become more normative it may influence attitudes and behavior towards women in STEM.