A Pain Rating Scale [PRS] and a Pain Behavior Checklist [PBC] were developed for use during a clinical pain interview. Norms were based on 395 chronic pain patients referred for pain management. The PRS obtained ratings of present pain, and worst and least pain during the previous 30 days. The PBC recorded 16 pain behaviors during the 45- to 60-min interview. Normative data were given for seven measures: (1) worst pain rating, past month; (2) least pain rating, past month; (3) present pain rating; (4) difference between worst and least pain ratings; (5) pain behavior score; (6) total pain score (based on Nos. 1, 2, 3, and 5), and (7) consistency score. The consistency score reflects the agreement (or discrepancy) between the pain behavior and present pain ratings. A moderate overall relationship was found between pain behavior and present pain ratings (r = 0.46, p < 0.001). The frequency of good agreement between pain behavior and present pain ratings was unaffected by sex, race, age, pain site, type of injury, duration of pain, legal representation, and evaluating psychologist; but it varied markedly with conscious symptom magnification. Patients seen as consciously exaggerating pain (n = 127) gave higher pain ratings (all p values < 0.001); and had frequent (64.6%) discrepancies between their pain report and pain behavior compared to others (14.2%). The report of present pain was unrelated to pain behavior for conscious exaggerators (r = 0.04, NS); in contrast there was a moderately high relationship for other patients (r = 0.68, p < 0.001).(ABSTRACT TRUNCATED AT 250 WORDS)