Explore Harvard's Nieman network Nieman Fellowships Nieman Lab Nieman Reports Nieman Storyboard

Where's Leo Bogart when we need him?

COMMENTARY | February 23, 2006

Phil Meyer finds problems with reports of really high newspaper circulation for some newspapers and offers a little history lesson, including how sociologist and newspaper analyst Bogart came up with the ‘read yesterday’ measure.

By Philip Meyer

Newspaper success stories are so rare these days that they have a man-bites-dog appeal to our news instincts. But reporters who want to find good news about newspapers can easily be led astray if they don’t understand the numbers that measure newspaper performance.

An Investor’s Business Daily report called Gannett’s Democrat & Chronicle of Rochester, NY, the nation’s readership “champion.” It said the paper reaches 84% of its market when its Web site is counted. Other high readership papers cited were the Austin-American-Statesman (83%), the Des Moines Register (75%) and the Syracuse Post-Standard (75%).

“Those are impressive figures given that, nationally, daily readership stands at 52%,” said writer Brian Deagon. (Some Newspapers Have the Right Stuff, posted Feb. 17, 2006)

Not really.

That national 52% is rounded from the 51.6% reported by Scarborough Researach as the proportion who read or look into a newspaper on an average weekday. The comparable figure for The Democrat & Chronicle’s designated market area is 52.3%, according to Jeffrey A. Kapuscinski, vice president for market development.

So where do the 84% and the other large numbers described by Investor’s Business Daily come from? They represent the net total of readership accumulated across a seven-day period plus those who look at the newspaper’s Web site at least once in a 30-day period. Big difference.

Newspapers have always looked for numbers that put their situation in the best possible light, and they have every reason to do so. Lately, the industry has been emphasizing seven-day reach. "Reach" is a concept used since the 1960s to show advertisers how multiple insertions would increase the audience size. It is even more important today as average daily readership continues to fall. The advertiser needs more insertions to get the same number of potential customers that used to be exposed to just one ad.

Unpacking them, we find that Rochester reports that a seven-day ad would reach 80.3 percent of the adults in its home market at least once in those seven days, and that another 3.6 percent would be netted from 30 days of exposure on the paper's web site.

The problem is to convince the advertiser to run an ad for seven days to get the same unduplicated exposure that used to result from a one-day ad. (Yes, in 1961, the year that standardized readership measures began, 80 percent of adults read a newspaper on an average day.) The Web is proving its worth as an ad medium, but it is too competitive to allow the kind of pricing to which newspapers are accustomed.

Readership measures were introduced in 1961 as a response to falling household penetration, which is simply circulation divided by households in a given geographic area. Leo Bogart, who died last year, invented the “read yesterday” measure to avoid contamination from social desirability bias. It was necessary because newspaper readership, like voting, tends to be reported by more people than actually do it.

Bogart’s method, endorsed by the Advertising Research Foundation, was to ask the survey respondent a series of questions about what newspapers he or she had read in the past week. Then, for each paper claimed, the interviewer asked, “When was the last time you read or looked into any weekday copy of the (name of newspaper)?” If the person said, “Today,” the next question was, “And when was the last time before today?”

The fact that “yesterday” had to be volunteered for the respondent to count as a reader made the measure credible to advertisers, and “read yesterday” became the gold standard for measuring newspaper performance for the rest of the century.

Measuring the cumulative reach is done with variations on that technique, and reach-and-frequency tables have been standard tools, since the 1960s, for persuading advertisers to run an ad more than once. The confusion comes when the term readership is used to describe five-day or seven-day accumulation rather than the traditional one-day level.    

Because readership numbers come from surveys performed by a variety of suppliers, the industry decided in 1999 to give them greater uniformity and credibility by bringing them under the surveillance of the Audit Bureau of Circulations (ABC). The resulting “reader profiles” are posted here publicly, if not always promptly.  

You can also find current compilations and comparisons here, on the Web site of the Newspaper Association of America.  

Another kind of error is to compare a newspaper’s total circulation with its city size, either in population or number of households. The Wall Street Journal, in a story describing the success of the Bismarck, N.D., Tribune, fell into this trap. (“Unlike Big Dailies, a Paper Prospers in Bismarck, N.D.” February 8, 2006.)

“With little competition, the Tribune sells nearly 28,000 papers – or about one copy for every two people,” said writer Joseph T. Hallinan. The problem with this comparison is that those 28,000 papers are distributed across a vast 24-county area, far beyond the city limits of Bismarck with its population of about 56,000.

One uniform way to compare newspaper’s success is to look at household penetration in their home counties. Outlying circulation is not always important to advertisers, but the home county is a place where most newspapers have to do well. And because counties are political units, their boundaries are not likely to change, as often happens with the newspapers’ own designated market areas.

In Bismarck, the trend is not good. In the 2000 ABC report, the paper enjoyed weekday household penetration of 58% in Burleigh County. By the 2005 report, that had fallen to 53% – a pretty good figure by today’s standards, but revealing a discouraging trend.

That same apples-to-apples comparison can be used to check out the situation in Rochester, home of the readership “champion.” ABC reported in 2000 that the daily Democrat & Chronicle enjoyed weekday household penetration in Monroe County of 50%. By the spring 2005 report, that had fallen to 46% – better than many, but definitely not a rebound.


Circulation – The number of newspapers sold on a specific day. Usually reported as a weekday or Sunday average, but other specific days of the week can be broken out. Includes both subscriptions and single-copy sales. Rules for ways to count discounted copies and third-party sales, i.e. to hotels for their guests, are set by ABC and its member newspapers. These rules are changed from time to time.

Household penetration – For a given geographic area, circulation divided by households. A paper with 50% penetration in the stated area sells one copy for every two households in that area.

Readership – Historically refers to “read yesterday” or average daily readership. Because copies are passed along, readership can yield a higher number than penetration.

Reach – Readership accumulated across a given period of time. Also called “cume.” Currently, the industry is promoting seven-day reach which yields a very high number because a person needs to read the paper on only one of seven days in order to be counted as a reader.

The NiemanWatchdog.org website is no longer being updated. Watchdog stories have a new home in Nieman Reports.