This study examined the utility of the K-correction procedure (i.e., raw score weights added to scales 1, 4, 7, 8, 9) to adjust for a defensive response set on the MMPI. The sample consisted of 51 patients with end-stage heart disease undergoing psychosocial evaluation for transplantation. Participants were separated into "defensive" and "nondefensive" groups using a median split on the K scale (defensive defined as T-score ≥ 59). The MMPI was scored once in the standard manner and then rescored omitting all K-scale items from the clinical scales. As hypothesized, raw score analysis after omitting K-scale items showed the defensive group endorsed significantly fewer items on three of the five clinical scales involving K-correction (scales 1, 7, 8). In contrast, analysis of K-corrected T-scores using standard procedures showed a significant group difference only on scale 4, with a higher T-score found among the defensive group. The defensive and nondefensive groups were not significantly different on numerous demographic, medical, and psychiatric characteristics, suggesting that the tendency to respond in a defensive manner is the major characteristic that distinguishes the two groups. Given this assumption, the K-correction procedure appears to appropriately adjust for a defensive response set on scales 1, 7, 8, and 9, whereas K-correction may overadjust on scale 4. © 2000 Plenum Publishing Corporation.