Griffin and Mahon (1997) extended earlier research in the area concerning the relationship between CSP and CFP, with particular emphasis on methodological inconsistencies in three critical areas. First, it focused on a particular industry, the chemical industry. Second, it used multiple sources of data-two that were perceptual based (KLD Index and Fortune reputation survey) and two that were performance based (TRI database and corporate philanthropy) in order to triangulate toward assessing CSP. Third, it used the five most commonly applied accounting measures in CSP and CFP literature to assess corporate financial performance. Their results showed that the a priori use of measures may actually predetermine the CSP/CFP relationship outcome. Amazingly, Fortune and KLD indices very closely track one another, whereas TRI and corporate philanthropy differentiate between high and low social performers and do not correlate with the firm's financial performance.