Increasingly, hospital administrators have been concerned about the public's perception of the facility. Nationwide, they have engaged marketing firms to study how consumers rate their local facilities in comparison to others. This type of information has been important to develop effective marketing and advertising programs (Steiber and Boscarino 1985). In this study, hospital ratings were analyzed for 65 short-term (nongovernment), medical and surgical hospitals across the United States. These hospitals represented different regions of the country (east, west, north, south, and central), as well as urban, suburban, and rural areas. Over 14,000 consumers were surveyed in these local market surveys. The public's ratings of these local hospitals were analyzed in terms of hospital size (number of beds), inpatient census, the "urbanicity" level of the local area, the level of care provided (primary, secondary, or tertiary), geographic region, and the 1984 Health Care Financing Administration death rate reported for Medicare patients. A multivariate analysis of the data indicates that hospital ratings are significantly related to the level of care provided and to the hospital's census level. Both of these are positively related to the public's attitude toward that facility (the higher the rating, the more specialized the care provided and the higher the census at that facility). Other variables are also positively related to ratings for example, bed size), but this is because of the relationship of these variables to either census or care level.