The most notable finding here is that cable news has all but abandoned what was once the primary element of television news, the written and edited story. In doing so, it has de-emphasized the story package's strengths, namely the chance to verify, edit and carefully choose words and pictures. The stress in cable news is on mmediacy and cost efficiency of the live interview and unedited reporter stand-up.

Next, rather than covering a comprehensive menu of issues, each morning the cable channels settle on a limited number of core stories that are then repeated, and only occasionally substantively updated, as the day proceeds. The level of repetition on cable is enormous. The level of updating is minor.

There are four distinct parts to the cable day -- morning news, daytime, early evening and prime time -- and each has different qualities. Prime time is remarkable for the fact that, for channels label themselves news, it is almost totally bereft of newscasts. If viewers wanted a comprehensive prime time survey of the national and international news of the day on these cable channels, only CNN offers such a newscast, and then for only one hour of its prime time schedule.

While there are differences among the three cable channels, the similarities in how they are put together and what they choose to cover stand out.

The dominant impression is that managers in the control room, rather than the on-air talent, function as the real agents of influence in cable. They decide what pictures to air, what stories to cover, where to go next, who gets to express expert knowledge and analysis. They define the personality of the product.

To get a sense of the nature of cable news, the study looked at five sample days for the three cable news networks, CNN, Fox News, and MSNBC, and studied those five days intensively, for 16 hours each day, or 240 hours of programming, some 5,570 story segments.

This intensive approach, examining days in depth over many hours, allowed researchers to get a sense of how cable news is constructed throughout the day. More conventional studies have tended to examine short periods of time on cable, usually a single program, over a longer number of days, the way that traditional network evening newscasts are studied. This conventional approach, while useful, leaves too much of the cable day unexamined.

For this study, instead, one of each of the five weekdays was selected at random from May to October of 2003.¹ [1] For those dates, our partner, ADT Research, publisher of the Tyndall Report, monitored and coded the cable programming continuously from 7 a.m. to 11 p.m. Eastern time.² [2]