The indifferent success in achieving the goal of reliable and valid police officer work performance measures has been attributed to the complexity of the job. This study explored reasons for unsatisfactory police performance evaluation, reviewing and integrating previous studies and new recently collected but not yet published data. Performance criteria studied include subjective supervisory assessments and several objective criteria such as arrests, commendations and disciplinary actions. A quantitative job analysis at three different tenure points was implemented. The job analysis survey (Job Functions Inventory for Police Officers) was completed on a volunteer basis by police officers (N=37) in two mid-sized police departments. The three length of service groupings were 0-6 years, 8-16 years, and 18 years or more. A discriminant function analysis produced four bipolar functions. The two poles of each function were treated as separate factors and thus produced eight scoring dimensions for the Inventory. Each of the eight dimensions had a counterpart among the eight dimensions identified in a 1968 study (Baehr, Furcon, Froemel). A possible imaginative description of these results showed the low-tenure officers engaged in street patrol, and active and hazardous criminal investigation activities. The latter activities continued in the middle-tenure group which also undertook special promotion and training assignments. Finally, the high-tenure group was chiefly concerned with enforcing police procedures and regulations and carrying out public duties. (ABL)