Abstract The present study used a bogus pipeline methodology to investigate the extent to which grandiose narcissism and other narcissism-related constructs were sensitive to bias in reporting. In addition, we sought to test the psychodynamic mask model by examining the association between narcissism variables and deep-seated feelings of self-esteem for men and women when both narcissism and self-esteem were assessed simultaneously under three different conditions: a bogus pipeline condition, an anonymous condition, and an exposure threat condition wherein participants believed that somebody else might be reviewing their responses. Results revealed that the assessment of narcissistic grandiosity and global self-esteem was sensitive to study conditions whereas assessment of narcissistic personality and psychological entitlement was not. Grandiose narcissism and self-esteem were positively correlated within each study condition, a finding which contradicts the psychodynamic mask model and has implications for understanding narcissistic functioning.