AT THE TIME OF THE STUDY FOUR CO-CORRECTIONAL FEDERAL INSTITUTIONS WERE IN OPERATION AT PLEASANTON, CALIFORNIA; FORT WORTH, TEXAS; LEXINGTON, KENTUCKY; AND TERMINAL ISLAND, CALIFORNIA. THE THREE PROGRAMMATIC STUDIES DISCUSSED HEREIN INVOLVED THE DEVELOPMENT OF AN INSTRUMENT TO MEASURE STAFF AND INMATE ATTITUDES TOWARDS CO-CORRECTION, REPLICATION AT THE SAME INSTITUTION 13 MONTHS LATER, AND A SERIES OF COMPARISONS AMONG CO-CORRECTIONAL AND ALL-MALE FEDERAL INSTITUTIONS USING THE TEST INSTRUMENT. THE LIKERT METHOD OF ATTITUDE SCALE WAS USED IN THE FIRST STUDY. ISSUES ARE LISTED. A TOTAL OF 238 STAFF MEMBERS AND 398 INMATES COMPLETED THE SURVEY AT THE LEXINGTON, KENTUCKY, FACILITY. SEPARATE CORRELATION MATRICES WERE COMPUTED FOR INMATES, STAFF MEMBERS, AND THE TOTAL SAMPLE FOR THE 38 ITEMS IN THE SCALE. IN THE SECOND STUDY THE PROCEDURE WAS REPLICATED AT THE SAME FACILITY EXCEPT THAT PARTICIPANTS INCLUDED 182 STAFF MEMBERS AND 521 INMATES. IN THE THIRD STUDY SUBJECTS WERE DRAWN FROM LEXINGTON, KENTUCKY, AND THREE ADDITIONAL MEDIUM SECURITY INSTITUTIONS, INCLUDING TWO FOR MALE INMATES ONLY. ANALYSIS PROCEDURES WERE ESSENTIALLY THE SAME AS THOSE OF THE SECOND STUDY. INMATES TEND TO BE MORE FAVORABLE TOWARD CO-CORRECTIONS THAN STAFFS WITH THE EXCEPTION OF THE LEXINGTON, KENTUCKY, STAFF. IN ALL THREE STUDIES THE RELATIVELY FAVORABLE RESPONSE OF THE LEXINGTON STAFF, AS COMPARED WITH THE STAFFS AT OTHER INSTITUTIONS, APPEARS TO CREATE THE INTERACTION EFFECT. FURTHER APPLICATION OF THE SCALE AT OTHER CO-CORRECTIONAL INSTITUTIONS IS NECESSARY BEFORE THE INSTITUTION BY STATUS INTERACTION CAN BE FULLY UNDERSTOOD. NOTES, REFERENCES, AND FIGURES ARE INCLUDED IN THE PAPER.