Skip to Content View Previous Reports

Audience

Audience

By the Project for Excellence in Journalism

Local television news audiences continued to decline in 2007, and evening newscasts were hardest hit.

In response, many stations shifted their news programming to new timeslots, trying to fit changing lifestyles. It is too soon to know whether the strategy will work.

Ratings and Share

By our analysis, local news ratings, at least for the four major network affiliates that command the largest share of the audience, dropped throughout the day and throughout the year. Their share of viewers, a different metric, at best held their own.

For the second year, we examined ratings and share data from the four sweeps months — February, May, July and November – that stations use to set their rates for advertising. In each of those months, we looked at the ratings and share of all stations affiliated with the Big Four networks (ABC, CBS, NBC and Fox) across the country.1

Ratings are the percent of households watching a program at a given time among all households in the market. Share is the percent of households watching a particular program of just those households that have their sets on. They are the key metrics for audience in television. Ratings give you a number for a program’s average audience. Share tells you the percentage of television viewers at that moment who are watching that program, their market share.

Local News: Change in Ratings
Sweeps, 2007 vs. 2006

Sweep Month
Evening News Late News Morning News
February
0%
3.2%
6.7%
May
-3.3%
-6.3%
0%
July
-6.5%
-7.4%
0%
November
-5.7%
-6.7%
0%

Source: Nielsen Media Research, used under license
Note: Numbers include ABC,CBS, Fox and NBC Affiliates

Local News: Change in Share
Sweeps, 2007 vs. 2006

Sweep Month
Evening News Late News Morning News
February
-7.1%
0%
-5.6%
May
-7.7%
-14.3%
-5.9%
July
0%
0%
0%
November
-7.7%
-8.3%
-6.3%

Source: Nielsen Media Research, used under license
Note: Numbers include ABC,CBS, Fox and NBC Affiliates

Early Evening News

Increasingly people are not at home or not opting to watch news (or any other programming) in the early evening around the dinner hour.

Local newscasts between 5 p.m. and 7 p.m. saw a drop in ratings and share in three of the four sweep months in 2007, according to our analysis.4 The only exception was in February when numbers saw no change from the year before.

Year to year, ratings were down 3% in May, 7% in July, and 6% in November. In February, ratings were flat from 2006.5

Average Early Evening News Ratings
Comparing Performance Year-to-Year in Sweep Months
Design Your Own Chart
Source: Nielsen Media Research used under license
Note: Numbers Include ABC, CBS, FOX and NBC affiliates

Share followed a similar trajectory. It was flat in July, and down in February (7%), May (8%) and November (8%). This suggests that increasingly people are not watching television at that hour, let alone news.

If we examine an hour earlier, 4 p.m., which is becoming popular with local stations that often advertise it as the first evening news hour, ratings look no better. The same is true if one examines the 7 p.m. to 7:30 timeslot, a time when some local stations are adding newscasts to follow the national network news broadcasts.6

Evening News Ratings
Sweeps Months, 2007

Time
February May July November
4 p.m. – 5 p.m.
1.8
1.6
1.7
1.6
5 p.m. – 7 p.m.
3.6
2.9
2.9
3.3
7 p.m. – 7:30 p.m.
2.6
2
1.9
1.8

Source: PEJ Analysis of Nielsen Media Research, used under license
Note: Numbers reflect the average rating of local newscasts on ABC, CBS, FOX and NBC affiliates that air during these timeslots in their time-zone.

Late News

Late-night news, the shows that follow prime-time entertainment programming, fared similar to early evening.

The only month in which ratings were up over the year before was February – when they rose 3% (an increase of 0.1 points). For the remaining three sweeps months, ratings were down more than 5% (6% in May, 7% in July and 7% in November).

Average Late News Ratings
Comparing Performance Year-to-Year in Sweep Months
Design Your Own Chart
Source: Nielsen Media Research used under license
Note: Numbers Include ABC, CBS, FOX and NBC affiliates

Share in late night was flat in February and July, according to our analysis, and fell 14% in May and 8% in November.

(The Project has begun to collect data for the news that comes earlier, during prime time, typically on Fox stations at 10 p.m. in Eastern and Pacific Time zones and 9 p.m. Central and Mountain. A year from now we will have comparisons to track their audience trends. (See Methodology Sidebar.)

Local News in the Morning

Early morning news, the local programs that come on before the network morning shows at 7 a.m., had been a lone growth area for many local stations around the country in recent years. In 2006, however, we found even this time slot had begun to suffer audience declines, with both ratings and share falling in every sweep month.

In 2007, the news was not quite so bleak. While the size of the audience remains less than half that of evening or late night, the newscasts between 5 a.m. and 7 a.m. held on to their viewers.

Ratings were unchanged in the sweep months of May, July and November. And in February 2007, the most positive month in this accounting for all timeslots, they were up 7% (0.1 rating point).

Morning News Ratings
Comparing Performance Year-to-Year in Sweep Months
Design Your Own Chart
Source: Nielsen Media Research used under license
Note: Numbers Include ABC, CBS, FOX and NBC affiliates

Share was flat in July and down the same amount (6%) in February, May and November.

A number of stations, particularly Fox affiliates, also extend their local morning news beyond 7 a.m., to 8 a.m. and even in some places to 9 a.m. (Fox airs a nationally broadcast morning show of its own, the Morning Show with Mike and Juliet, in 15 major markets that begins at 9 a.m.) A few CBS affiliates that decline to use the Early Show feed from the network also air local news during these hours.

These later morning programs saw ratings in 2007 similar to and in some ways more favorable than the early morning block. (The Project in future years will track the growth patterns of these audiences.)

Morning News Ratings
Sweeps Months, 2007

Time
February May July November
5 a.m.-7 a.m.
1.6
1.4
1.1
1.4
7 a.m.-8 a.m.
1.6
1.5
1.2
1.4
8 a.m.-9 a.m.
1.4
1.2
1.2
1.2

Source: PEJ Analysis of Nielsen Media Research, used under license
Note: Numbers reflect the average rating of local newscasts on ABC, CBS, FOX and NBC affiliates that air during these timeslots in their time zone.

Non-Affiliates

If we examine independent stations, those not affiliated with the big four networks that air local news, the universe shrinks considerably. Only about 40 of these stations in the country that run news have large enough audiences for Nielsen Media to track, according to data from the ratings company. Their performance in 2007 mirrors that of the affiliate stations.

In the early evening block, according to the last 2007 sweeps period (November), the independent stations monitored saw ratings unchanged, but a drop in share. Comparing November sweeps of 2007 to those in 2006, ratings stayed the same (0.5) but share dropped a point (from 2 to 1).7

In the late news block, both ratings and share were down. Again, comparing November to November, ratings fell to 0.2 points in 2007, from 0.7 in 2006 to 0.5 in 2007. Share fell a point from 2 to 1.

The morning news hours numbers were flat. Of the stations for which Nielsen has data, ratings rose 0.3% and share held steady at 3 in November 2007 compared with the same period the year before.

Shifting Schedules

As stations contended with falling ratings, there was good deal of talk in 2007 about stations shifting their newscasts to new timeslots to adjust to changing lifestyles.

To find out how widespread the phenomenon of time shifting might be, PEJ this year examined data for stations affiliated to the Big 4 networks in the top 25 designated market areas (DMA) in the country. From data licensed from Tribune Media Services, we took one weekday in November 2007 and compared it with the corresponding weekday in 2006.

Stations in two-thirds (15) of the top 25 markets added or shifted newscasts in 2007, according to the analysis. In all, 24 stations (of 106) in these markets did some time shifting, or just under 25%. The majority of these changes occurred during the early morning hours.8

New newscasts were added starting as early as 4:30 a.m. There were also instances of stations repeating their late night newscasts in the overnight hours, with additions virtually every hour – from midnight (Phoenix, Dallas), 1 a.m. (Tampa), 2 a.m. (Los Angeles) to 3 a.m. (Atlanta and Portland).

Shifting Schedules
Top 25 Markets, November 2007 vs. November 2006

Rank
Market Added Newscasts at Shifted Newscasts
1
New York 7 a.m. (Fox); Noon (ABC) 5 p.m. to 7 p.m. (NBC)
2
Los Angeles 2 a.m. (CBS); 7 a.m. (Fox); 9 a.m. (Fox)
3
Chicago 4 30 a.m. (NBC); 10 p.m. (Fox)
4
Philadelphia 5 p.m. (Fox)
5
Dallas-Ft. Worth Midnight (CBS); 5 a.m. (Fox); 6 a.m. (CBS)
6
San Francisco-Oakland-San Jose
7
Boston (Manchester)
8
Atlanta 3 a.m. (NBC)
9
Washington, DC (Hagerstown, Md.) 6 p.m. (Fox) 10 a.m. to 11 a.m. (NBC)
Noon to 11a.m. (Fox)
10
Houston
11
Detroit
12
Phoenix (Prescott) Midnight (ABC) 11 a.m. to 1 p.m. (NBC)
13
Tampa/St. Petersburg (Sarasota) 1 a.m. (ABC)
14
Seattle-Tacoma
15
Minneapolis-St. Paul
16
Miami-Ft. Lauderdale 4 30 p.m. (Fox); 7 p.m. (NBC); 10 p.m. (Fox) 10 a.m. to 11 a.m. (NBC)
17
Cleveland-Akron (Canton) 11 a.m. to Noon (NBC)
18
Denver
19
Orlando-Daytona Beach-Melbourne 6 p.m. (Fox)
20
Sacramento-Stockton-Modesto
21
St. Louis 10 p.m. (Fox)
22
Pittsburgh
23
Portland, OR 3 a.m. (Fox); 4 30 a.m. (Fox); 4 30 a.m. (NBC)
24
Baltimore
25
Charlotte

Source: PEJ Analysis of Tribune Media Services data, used under license

Evening was not as popular a time for adding new newscasts among the network affiliates in the top 25 markets. But even among the top markets, we did find additions, in three markets – Orlando (6 p.m.), Washington, D.C. (6 p.m.) and Philadelphia (5 p.m.).

Another trend among the big markets appears to be to eliminate earlier newscasts in favor of later ones. In both Miami and New York, for example, the NBC affiliate (WTVJ and WNBC, respectively) got rid of the 5 p.m. newscast in favor of one at 7 p.m.9

Robert Papper, a television analyst at Hofstra University, also notes that stations can time shift in another way, by producing news for other stations in the market that run at times when their own newscasts do not. This can allow them generate revenue for news without altering their own schedules. It amounts to time shifting of a sort, across channels. Papper estimates that this is particularly popular during the evening. Many of these newscasts run in the last hour of prime time, competing often against Fox.

Our analysis also does not reflect the trends outside the top 25 markets. And there are a growing number of independent and local cable stations that air competing newscasts in the evening, which are often effective in luring viewers away from the Big Three (ABC, CBS and NBC) newscasts. The traditional early evening block also barely has any Fox affiliates, which tend to air news later, at 10 p.m.

Thus, if you look at the larger picture, local television news is now populated by not just the big network affiliates, but also by local cable and non-affiliated stations, and not just by news at the traditional early evening and post prime-time news slots.

Changes in Audience Measurement

As television audiences decline, the industry adopted a new system or “currency” in the industry language, to measure viewership in 2007 (see Advertising Chapter).

Rather than measuring viewers during program minutes, the industry began measuring viewers of commercials aired during the program. The new metric looks at average rating for all commercials aired in a particular show – starting with when it aired live and continuing over the next three days (giving it its industry nickname “C3”).10

Until now, advertisers decided what time to air and what price to pay for their ads based on ratings for a specific program – the higher-rated a program, the more advertisers wanted to place their ads during it and, consequently, the more they were willing to pay. Now, they will purchase advertising time based on average viewership of the commercial breaks during that program.

This marks a huge departure, and the new measuring system was preceded by wide debate about technical validity (see last year’s Report). After years of back and forth, in March 2007, all the key stakeholders in the ratings metrics battle – broadcasters, advertising agencies (which were agreeable) and cable companies (which weren’t) – finally agreed to adopt the new ratings as the currency for buying and selling ads.

Despite all the concerns and speculation, preliminary results of the new ratings did not really alter the results.

Nielsen Media – the predominant television audience measurement agency in the United States – released commercial ratings for the week of September 24 in October, 2007 and found that total viewership for commercial breaks on the broadcast networks was, on average, just 3% lower than it was for viewership of the respective programs. Comparing the top programs using the C3 ratings to the older ratings saw the same top 10 programs, with a slight difference in their ordering.11

These results were similar to those found during the trial run of the new ratings, released in July 2007. Those had found a 5% to 10% drop in the number of viewers who watched ads, compared to the show itself. The drop was somewhat higher for cable than broadcast networks.

The move was also seen as an acknowledgment to consumption realities. Fewer people in the U.S. now watch a show when it airs live and are increasingly adopting other means – including digital video recorders, video-on-demand options and/or Web video – to view them at their convenience.12

The new C3 ratings were one of the many changes that were part of Nielsen Media’s “A2/M2” endeavor to revise television ratings to keep pace with actual consumption practices (see last year’s Report).

The other big change was Nielsen’s aim to measure more markets electronically. Nielsen reduced the number of markets with handwritten diaries and rolled out more Local People Meters to take their place – these electronic devices provide a more accurate representation of viewing habits (although Nielsen still does not measure out-of-home TV viewing.)

In 2007, the Local People Meters were working in the top 13 markets, up from the 10 the year before. The plan is to reach 56 markets by 2011.13 According to the company, 56 markets would mean that 70% of U.S. households will be measured by the People Meters.

How Nielsen Media Measures Viewers
In the 210 U.S. Television Markets

Type of Market
Method Used Number of Markets Covered Percent of TV Households Covered
Local People Meter Markets
Local People Meters measure total viewership and viewer demographics daily
13
30%
Combined Meter/Diary Markets
Electronic meters measure total viewership daily; handwritten diaries track viewer demographics four times a year
43
40%
Diary Markets
Handwritten diaries track total viewership and viewer demographics four times a year
154
30%

Source: Nielsen Media Research Web site; Broadcasting & Cable

Alongside this transition, Nielsen Media also aims to triple its core national television ratings panel to 100,000 people (from its current figure of 35,000).14 This sample is what is used to determine television ratings and share.

Footnotes

1. The four sweep months are when Nielsen Media Research measures television audiences to help the industry determine advertising rates for TV stations. We took Nielsen data for all the stations affiliated with the four biggest local television networks in all designated market areas (DMA). That gave us the ratings and share for an average local newscast in each time slot in each sweep month. According to Nielsen Media Research, the DMA “identifies an exclusive geographic area of counties in which the home-market television stations hold a dominance of total hours viewed.” There are 210 DMA’s in the United States. See Nielsen Media Research Web site, http://www.nielsenmedia.com.

2. Share tells a station how it is performing compared with the other stations in the local area. Ratings give a sense of the total audience and are used by advertisers to determine what price they are willing to pay for an ad on the particular program. Webster, J., Phalen, P., & Lichty, L., (2000) Ratings Analysis: The Theory and Practice of Audience Research, Lawrence Erlbaum Associates.

3. Data in previous reports were limited to stations affiliated with ABC, CBS or NBC. We remained consistent with this approach year to year for the purposes of a trend over time. Other stations were not included either because they did not carry news or, as in the case of Fox network affiliates, they aired news in non-traditional time slots, particularly during prime time. The time slots we measure represent the traditional timing of local newscasts.

4. For early evening news, we took newscasts between 5 p.m. and 7 p.m. in the Central and Mountain Time zones and 6 p.m. to 8 p.m. in the Eastern and Pacific Time zones. For late news, we took 10:00 p.m. to 10:30 p.m. in the Central and Mountain Time zones and 11:00 p.m. to 11:30 p.m. in Eastern/Pacific.

5. The evening news broadcast in February 2007 had an average rating of 3.6 points. Each rating point represented 1,114,000 viewers in 2007, according to Nielsen Media Research. One rating point refers to 1 percent of the television viewing universe (or 111,400,000 as of spring 2007, according to Nielsen).

6. There are instances where the network news itself airs at 7 p.m., such as the NBC news broadcast in markets like Washington, D.C.

7. According to Nielsen Media, the non-affiliated stations air news in the early evening in only about 20 markets (of 210), in the late night in about 30 markets, and in the morning in only 18 markets.

8. While it does not constitute time shifting per se, stations can also generate revenue by producing news for other stations in the market at different times than their own newscasts. Some 206 stations in the United States now purchase newscasts from other places.

9. Marisa Guthrie, “Stations Kill ‘Live at Five,’ ” Broadcasting & Cable, September 10, 2007.

10. Viewers who watch the commercials on the day the program is aired (“live” ratings) and those watch it over the next three days using digital video recorders. MediaPost, May 14, 2007.

11. Advertising Age, October 16, 2007

12. Nielsen Media has the technology to measure either the average rating of all commercials in a program or the ratings of each individual commercial. MediaPost, March 21, 2007.

13. Alex Weprin, “Nielsen National TV Ratings Panel Expanding More than Threefold,” Broadcasting & Cable, September 26, 2007.

14. There are plans to make the people meters active in five more markets in 2008, 12 in 2009 and 2010 each and 14 in 2011. In 2007, there were 12,000 homes in the national panel and Nielsen aims to increase it to 37,000 homes in 2011 (reaching 100,000 people). Nielsen pointed out, however, that the effective sample size will only be 17,000 homes after the geographic distribution of the 56 people meter areas is considered. The “effective” size in 2007 was 10,000 homes. Alex Weprin, “Nielsen National TV Ratings Panel Expanding More Than Threefold,” Broadcasting & Cable, September 26, 2007.