THE STUDY POPULATION INCLUDED 1,700 INMATES AT THE IOWA STATE PENITENTIARY. AGED INMATES WERE DEFINED AS THOSE 55 OR OVER WHEN THEY ENTERED AND LEFT PRISON; THE MIDDLE GROUP WAS AGED 31 TO 54; YOUNG INMATES WERE THOSE AGED 30 OR LESS. THE MAJOR DEPENDENT VARIABLE WAS LENGTH OF INCARCERATION IN MONTHS. LEGAL BACKGROUND VARIABLES INCLUDED THE NUMBER OF PREVIOUS ARRESTS AND THE NUMBER OF PRIOR PRISON COMMITMENTS. INSTITUTIONAL BEHAVIOR VARIABLES INCLUDED THE NUMBER OF MAJOR DISCIPLINARY REPORTS AND THE NUMBER OF DAYS IN SOLITARY CONFINEMENT. OTHER VARIABLES INCLUDED THE LENGTH OF SENTENCE, TYPE OF OFFENSE, AND INMATE'S PHYSICAL CONDITION. ANALYSES OF VARIANCE AND COVARIANCE INDICATED THAT AGE IS POSITIVELY RELATED TO INMATES' LEGAL BACKGROUND BUT NEGATIVELY RELATED TO INSTITUTIONAL BEHAVIOR. THESE AND THE OTHER VARIABLES WERE CONTROLLED TO ASSESS AGE'S NET EFFECT ON LENGTH OF INCARCERATION. AGE WAS FOUND TO HAVE NO DIRECT EFFECT OR INDIRECT EFFECT AS A NUISANCE VARIABLE ON THE LENGTH OF INCARCERATION. AGE, THEREFORE, IS NOT AN EXTRALEGAL VARIABLE. AGE'S INFLUENCE IS MEDIATED BY OTHER VARIABLES SUCH AS OFFENSE TYPE, SENTENCE LENGTH, LEGAL BACKGROUND, AND INSTITUTIONAL BEHAVIOR. THEREFORE, WHEN RELEASE DECISIONS ARE BEING MADE, AGED INMATES HAVE NO SIGNIFICANT ADVANTAGE OVER NONAGED INMATES BECAUSE AGED INMATES MAY NOT BE PERCEIVED AS LESS DANGEROUS TO SOCIETY THAN NONAGED INMATES. THESE CONCLUSIONS ARE SUBJECT TO DATA AND MODEL LIMITATIONS, THE SMALL NUMBER OF INMATES IN CERTAIN OFFENSE CATEGORIES, AND THE AGE GROUPINGS CHOSEN, RESEARCH ON THE INFLUENCE OF A COMBINED 'NUISANCE' FACTOR ON OTHER VARIATIONS IN INMATES AND INSTITUTIONAL RESPONSES TO THEM, AND ON INMATES' BEHAVIOR, IS RECOMMENDED. THEORETICAL AND METHODOLOGICAL DISCUSSIONS, TABLES, DIAGRAMS, AND A BIBLIOGRAPHY ARE INCLUDED. (CFW)