“…The technique of controlling companies by industry has proven to be preferable to other past studies that have looked for CSR relationships without controlling for this variable (e.g., Moskowitz, 1972;Parket & Eilbirt, 1975;Vance, 1975;Heinz, 1976;Alexander & Buchholz, 1978). By studying within an industry, a more accurate and homogeneous comparison between firms is possible since certain industries have different "accounting practices, operating leverage and other variables, [such as risk], [which] may influence test results" (Cochran & Wood, 1984, p. 47).…”