Content-Type: text/html

Readability and Content Differentiation of Business News








A Comparative Analysis of On-line versus Print Media:
Readability and Content Differentiation of Business News






Abstract




The purpose of this study was to examine the readability and content differentiation of business news. Specifically, three newspapers and three Internet sites were content analyzed to see differences based on the media type. The findings suggest that the Internet business news showed more difficult readability with longer sentences, lower reading ease scores and more complex business terminology. The other results showed differences in the topics and visual usages between newspapers and Internet sites.







"K-Tel International is planning to launch a retail Website. Posting on Yahoo's bulletin board said that K-Tel's stock would soar. CNBC news picked up on the news, helping propel the company's stock up 125%. Pfizer's Viagra was being touted on Yahoo's board as a monster drug even before the FDA approved it. Pfizer's stock has climbed 20%." (Galarza, 1998). The Internet is the ideal medium for financial news, because of its real time information. It is not surprising, then, that more and more people rely on the Internet to check market fluctuations for their individual investments.
On the other hand, the decline of daily newspaper circulation in the U.S. that began in the early '70s has caused considerable concern in the newspaper industry and among scholars. Since 1973, when the total newspaper circulation had reached its maximum point, it has plummeted continuously until recent years (Wall Street Journal Almanac, 1999).
The growth of competitors in broadcast, weekly magazines and, most recently, on the Internet inevitably has some impact on the dropping newspaper circulation. And changing lifestyle, with greater emphasis on recreation and other outdoor activities, apparently cause some people to believe they don't have time to read a newspaper (Morton, 1999). In addition, in the visual age, the difficult readability of newspapers may be one of the reasons causing continuously dropping circulations.
If newspapers are difficult to read, readers may turn away from them as a source of information. Fusaro and Conover (1983) noted that if the readability levels of a newspaper demand too much of a reader's skills, reader satisfaction will be reduced. In particular, if newspaper can't provide satisfactory information to readers, the situation will worsen. In the case of business news, the needed and understandable information is the key factor to attract readers. If there's nothing to get and stories are difficult to understand, nobody will want to read business news.
In the capitalistic society, many people try to get economic, business, and financial information through a variety of ways to make financial decisions. Sometimes, those decisions are relatively trivial, but sometimes they have far reaching consequences. The most desirable information sources are the mass media such as newspapers and television or Internet. Even though many people search for personally tailored information through on-line news, mass media are also still profound and a broad information source. The influence of business news cannot be exaggerated in the society. Therefore, the function of business news information providers needs to be examined thoroughly. Nevertheless, business news has rarely been addressed before in the studies of newspaper and on-line news studies. Because the impact of business news is very critical for industry and general readers, it is necessary to have a better understanding of the current content of business news. In particular, since Web
 content is produced by a different system and consumed differently from printed news, it is expected to pursue a different audience through the differentiated content.
Therefore, there are several questions that need answers. How does the newspaper content vary from the Internet business news sites? Is it understandable or too difficult to read?
The purpose of this study is to examine how the most popular newspapers and Internet business Web sites in the U.S. covered business news, especially, focusing on readability and content differentiation. Three newspapers -- Wall Street Journal, New York Times, USA Today -- and three on-line business news Web sites -- CBSMarketWatch, CNBC, CNNfn -- will be examined.

Literature Reviews

A flurry of research has been done in readability studies, initially in the field of education research. If the reading materials are too difficult for students to read, their usefulness in the classroom maybe questioned (See Severin & Tankard, 1997, for the detailed history).
Flesch (1948) developed a readability formula, which is one of the best-known readability measures. He determined that reading difficulty was predicted by length of words and length of sentences.
Journalism scholars have attempted to apply Flesch's readability notion to journalistic writing styles to analyze the readability of mass media such as newspapers and magazines. Danielson and Bryan (1964), for example, showed that hard news stories were written in a more difficult to understand writing style than other soft news stories. Other researchers who have looked at newspaper readability have found newspapers to be more difficult to read than is popularly assumed; often articles are in the "very difficult" classification on the Flesh scale (Stempel, 1981; Johns and Wheat, 1978).
A number of researches have asked for newspapers to simplify their writing styles to appeal to more readers. Burgoon, Burgoon, and Wilkinson (1981) argued that newspapers should lower their readability levels if they want to increase their effectiveness. Other researchers warned that large segments of the population may become turned away if newspapers continue to be written above the reading level of general readers (Fowler, 1978; Smith, 1983).
Gunning (1952) developed reading ranges that were based only on sentence length. He argued that the longer sentence, the more difficult it is to comprehend, since longer sentences mean more relationships between words and require the reader to expend more mental energy, which reduces reader satisfaction. Stapler (1985) found that sentence length in lead paragraphs exceeded lengths in the three paragraphs that followed, and leads for most papers were "difficult" to read.
Hoskins (1973) found that half of the AP stories and 83% of the UPI stories were in the "difficult" or "very difficult" levels according to the Flesch reading ease score scale. Monznette and Rarick (1968) found that editorials in ten West Coast dailies were more readable than news stories. However, overall difficulty exceeded "standard" levels. Stempel (1981) compared the readability of six kinds of content and concluded that international news was clearly most difficult and sports and women's articles were easier than others. Johns and Wheat (1978) investigated the readability of newspapers to determine the difference in the readability of political and sports stories and to determine the difference between wire and non-wire stories. They found that non-wire sports stories were more readable than other formats. This result is consistent with the previous research (Anderson, 1966; Danielson and Bryan, 1964; Razik, 1969; Reiley, 1974). Later, Catalano (1990) proved similar results
with a wire-service news analysis.
Several studies on readability were conducted over time. Fowler and Smith (1979), for example, indicated that the newspapers were generally less readable than magazines through the comparison of the sampled three year period. While the readability of the magazines has remained fairly constant during the 1904, 1933, and 1965 time periods, the readability of newspapers has changed. The findings of the study showed that newspapers are generally as readable today as they were in 1904 although the complexity of the language they use has increased.
Visuals are often discounted or ignored in word-oriented journalism. However, they perform important roles in communication by conveying realism, credibility, and attitudes. Readers also appear to react positively to visual elements, such as photographs and graphics. A study pointed out that photographs are popular with readers and call attention to news stories, highlighting the importance of visuals (Baxter, Quarles, and Kosak, 1978). Polansky (1989) found that the dominant photograph is the typical pattern for a reader processing information on the page. Bain (1980) concluded that large pictures attract readers to an accompanying story better than small pictures and also hold the readers' attention deeper into the story. An experiment by Wanta (1986) found that the size of photos in newspapers could have an immediate influence on readers. The study found that editors have the power to raise their reader's salience on certain issues over a short period of time merely by increasing the size of photographs. He also supported the importance of visuals in interpreting media content in another study. L
arge graphics provide readers with important information in a concise manner and thus make processing information easier (Wanta and Gao, 1995). Waldman and Devitt (1998) presented the results of a content analysis of photographs appearing in five major newspapers during the 1996 presidential campaign. They argued for the presence of a strategic bias benefiting the front-runner.
        Hynds (1980) found that most newspapers are seeking to make their pages more attractive and understandable through the use of pictures and art work. They used pictures, drawings, cartoons, and diverse colors. With advanced computer technology, retrieving graphics from database and updating them as required in an Internet newspaper can be completed in minutes. Thus, the trend of increasing newspaper use of graphics can continue in the Internet environment (Li, 1998).
        Historically, little research has examined the content of business news. Newspaper coverage of business was called dry, listless, and out of touch with the interests of the average readers. Carswell (1938), a financial writer on the New York World-Telegram, resulted the prevailing over-emphasis on Wall Street and argued that the editorial point-of-view must be reoriented to the average businessperson; editing must be done for the reader and not the investor. His argument that the public at large is economically illiterate where the mechanics of money is concerned is consistent with the readability studies of scholars.
        In a 1980 study, Hynds found cause for optimism in a national survey of 186 newspapers. He concluded that newspapers are becoming more aware of the need for business coverage and are devoting more of their news hole to it.
        Barkin (1982) examined the changes in the business section from 1931 to 1979. He found that three major dailies covered the stock market less and international business more. He also emphasized the importance of readable business news and noted the improvement of the diversity of subjects. Through the analysis of trends in newspaper coverage of business and economics between 1968 and 1978, Feldman and Aronoff (1980) found that newspapers moved away from straight news coverage, such as lengthy earnings reports, to articles oriented towards the how and why of business operations and the implications of domestic and international economic policies and problems.
The Internet offers new reporting and research opportunities. While the Internet is mainly used as an efficient tool for reporting (Garrison, 1995) and as a research tool for content analysis through databases, such as Lexis-Nexis, it also gives researchers a chance to compare the web page with the print version of the papers.
Aronson, Sylvie, and Todd (1996) found that Wall Street Journal articles are longer than real-time articles on the Dow Jones Business Wire, but readability levels are nearly the same. A content analysis of three U.S. Internet newspapers by Li (1998) has found that text information is still dominant as newspaper content on the Web. The Web version of three newspapers did not take full advantage of available technology.
Other research has been focused on the characteristics of readers of online news. Mueller and Kamerer (1995) explored reader preference for online newspapers versus traditional newspaper. They found that readers of online news preferred to breaking news, world news, entertainment news, business news, and sports news and online news preference correlated most positively with computer liking. Later, Weir (1999) argued that the use of the electronic newspaper is not correlated with the use of the Internet or knowledge of computers. According to his founding, electronic editions are enthusiastically read by a much younger audience that has been a problem for the print newspaper. Chyi and Lasorsa (1999) found that most people still read the local newspaper in the printed format, while national newspaper sites were gaining more ground online.
Looking at the effect of online media on readers is another interesting perspective. Tewksbury and Althaus (2000) compared the differential effects of exposure to print and online versions of the New York Times. They found that online version readers have read fewer national, international, and political news and were less likely to recognize and recall events that occurred during the exposure period.
Even though many researchers have shown interest in online news analysis, most of studies are comparison of print and online version of the same paper. Despite the fast growth of Internet only news sites, however, few studies have been published in the analysis of Internet only news content.
This study will address two important questions to evaluate business news of print newspaper and Internet business news. First, how does the business news of newspapers differ from Internet only business news in readability? Second, how does the business news of newspapers differ from Internet only business news in content?

Methods

To analyze the readability and content of business news, six media were examined. The top three newspapers (Wall Street Journal, USA Today, and New York Times) were selected based on circulation. The most recent survey reported that Wall Street Journal has 1,774,880, USA Today has 1,629,665, and NYT has 1,074,741 in circulation (Gale Research Co., 1999). The three business news Web sites were selected based on number of sites visits (Gunther, 2000). Even though MSNBC was ranked first, it relies heavily on wired news services. Thus, it was excepted for this study. CNBC, CNNfn, and CBSMarektWatch were selected, because they belong to the most visited top ten sites and provide their own news content.
        The front page of Wall Street Journal and the business section front pages of USA Today and New York Times were content analyzed on a number of variables, including readability and content difference variables. Even though the front-page function is not distinct in Web sites, the top stories at the homepages of three Web business news sites were examined.
The variables were: Average sentence length; Flesch reading ease score; Flesch-Kincaid grade-level readability; number of complex business terminology used; the visual type; attribute of content.
Average sentence length is the average number of words per sentence. As Gunning (1952) said longer sentences that have many words require the reader to expend more effort, which causes reader dissatisfaction. He urged writers to keep their average sentence length under 20 words. Flesch (1951) also stressed short sentences and recommended that news writers use an average sentence length of 19 words.
The Flesch reading ease score is calculated by average sentence length and number of syllables per 100 words. Even though Smith and Smith (1984) reported the inconsistent readability of newspapers according to the several readability measurement scales, Flesch Formula (Flesch, 1974) was used because it is the test most commonly used by journalism researchers. The output for the Flesch Reading Ease formula is a number from 0 to 100. While low Flesch scores indicate the material is difficult to read, high grade levels mean the content is easy to read. Followings are description of writing style for each reading ease score.  0-30, "very difficult"; 30-50, "difficult"; 50-60 "fairly difficult"; 60-70, "standard"; 70-80, Fairly ease"; 80-90, "easy"; 90-100, "very easy."
Like the other Flesch formulas, the Flesch-Kincaid Index uses the number of syllables per word and words per sentence to calculate a grade level required for reading. Its highest grade level is 12th. This index is calculated automatically on the grammar checker readability tests in MS word processing program.
How many readers know what an assumed debenture is? Difficult terms often frustrate readers and take more time to be understood. Dale-Chall readability formula has a list of 3,000 words. If a word is not included in the list, it is counted as an unfamiliar word (1995). For clarification of difficult terms, this study borrowed the concept of Dale-Chall formula. However, instead of using the same 3,000 common words, difficult terms were selected based on the glossary of Warfield's book (1994), "How to read and understand the financial news." Among Warfield's glossary, already generalized terms, such as interest or investment, were not presented. Instead, several terms used recently, such as B2B or B2C were added. Therefore, all terms related to economics, business, and finance in six media were presented to a small group (5 students majoring Journalism). If the students didn't understand the word clearly at first reading, it was classified as the complex terminology.
Visuals were coded separately. The visual aspects were analyzed based on the type. The size was not considered in this study, because the Web sites had different sizes varying on the monitor size. Visual types were composed of three items; photos, computer graphics, and data format. A photo is defined as a factual representation of human, animal, or other objects made by photographs. A computer graphic is defined as a pictorial output produced through a computer technique. Most of them are changed, exaggerated, or combined images that cannot be taken by photos usually. Data format includes table, chart, and graph that is complementary for the explanation of articles. Since WSJ doesn't provide actual photos of persons and uses illustration instead of photos, the illustrations of persons were coded as photos.
Stories were classified in terms of their specific content area, which in turn belonged to one of nine categories:
1) National Economy, which included the current economic status, outlook for the economy, inflation, national news about wages and prices, taxes, federal loans, GNP, GDP, jobs, employment, layoff, and other general economic explanations.
2) International Business and Trade, which included international investment and trade, business relations, foreign business news, etc.
3) Legal and Regulatory issues, which included monetary policy, regulation, antitrust, and court lawsuits.
4) Individual Firm, which included change in corporate status, such as, mergers and acquisition, expansion, bankruptcy, profit, losses, sales, and new product, etc.
5) Industry profiles, which included broad industry news, such as telecommunication, broadcasting, transportation, food industry. It's different from individual firm news, which dealt with only specific company news.
6) Executive Profiles, which included firms' executives or other personal stories.
7) Stock Market, which included stock exchange trading and news about Dow Jones industrial Average, Nasdaq Composite, S&P 500 indexes, earnings, dividends, and bonds. In case of a foreign stock market, it was categorized as International Business and Trade.
8) Consumers Guides, which included government programs for the consumer, offers, services, consumer group activities, union activities, and personal financial investment advice.
9) Others, which included several miscellaneous items, such as education, real estate, etc.
The unit of analysis is the article. Since USA Today and Wall Street Journal publish editions five and six days a week, Saturday and Sunday editions of New York Times were excluded to have equal sample sizes for all newspapers. Every newspaper has four articles exactly on front page each day; therefore the same number of articles was also selected from each Internet business news site everyday. Even though it is hard to choose the top news stories from a Web site, there are top stories that are indicated by bold type and approximately four or five articles were presented as top stories at each site. The readability ease scores were computed for 180 articles (four articles for fifteen days) in three newspapers and for 180 articles (four articles for fifteen days) in three Internet business news sites for three weeks from February 12, 2001.
Since the first a few sentences catch the readers' attention for each article, approximately the first hundred words were taken, stopping at the end of the sentence that contained the 100th word.
Means were computed for the three newspapers and the three Web sites. Readability variables among six companies were compared by one-way analysis of variance. Differences between each media (print vs. Internet) were analyzed by t-test. To identify the content differences, Chi-Squares were tested.
To determine coder reliability, two persons coded randomly selected 10% of the content. For purely quantitative and computerized calculations, categories such as number of words, characters, sentences, and paragraphs were not examined in the intercoder reliability test. Using Holsti's formula (1969), the average agreement for the coding categories was 94.5 %.

Results

Average Sentence Length
Overall Internet business news articles had longer sentences than newspaper articles did. The three Internet news sites showed very similar sentence length. The mean sentence length of CBS MarketWatch was 23.0 (SD = 5.2), CNBC was 24.7 (SD = 5.1), and CNNfn was 24.9 (SD = 7.4). Although two newspapers showed shorter articles length (Wall Street Journal, M = 19.9, SD = 6.2; USA Today, M = 19.7, SD = 4.8) than Internet news articles, the New York Times revealed the longest average sentence (M = 27.7, SD = 7.7). This difference was statistically significant (See Table 1; F (5, 354) = 15.163, p = .000).
These results were replicated when the six companies were grouped by two media type. Since New York Times had exceptionally longer sentence length, the mean difference between newspapers and Internet sites was reduced based on media type. The mean difference, however, was also statistically significant (See Table 2; t (358) = - 2.541, p <. 05).

Reading Ease Score
The average overall readability ease score for these six business news sources was in the range of "difficult" (30 to 50) in the Flesch scale. While three Internet news sources showed quite similar results of reading ease score (CBS MarketWatch, M = 40.4, SD = 10.2; CNBC, M = 39.4, SD = 10.4; CNNfn, M = 37.1, SD = 12.6), newspapers showed different results. WSJ and USA Today showed considerably high scores, 48.3 (SD = 13.9) and 47.0 (SD = 10.7) respectively. The mean score for NYT, however, was 36.2 (SD = 12.1), which was the lowest score among the six. This result was significantly different in ANOVA test (See Table 1; F (5, 354) = 11.097, p = .000).
When regrouped by media type, Web site news were more difficult to read than newspapers. The overall mean for newspapers of 43.8 (SD = 13.6) and mean for Web sites of 38.9 (SD = 11.2) correspond to what Flesch terms "difficult" reading. Its difference was statistically significant in t-test (See table 2; t (358) = 3.744, p = .000).

Grade Level
The grade level also showed difficulty in reading business news. The overall mean of Flesch-Kincaid grade level for six business news was 11.3 grade (SD = 1.2). While NYT showed the highest grade level (M = 11.8, SD = .53), WSJ showed the lowest level (M = 10.5, SD = 1.8) among the six business news sources. The difference among six media showed significant difference in ANOVA test (See Table 1; F (5, 354) = 14.179, p = .000).
When regrouped by media type, the overall mean for newspapers was 11.0 (SD = 1.56) and mean for Web sites was 11.6 (SD = .83). It also showed a significant difference in statistics (See Table 2; t (358) = - 4.731, p = .000).
Since the Flesch-Kincaid grade level test has a maximum of 12th grade level, it doesn't show the college level. The articles that have scores between 30 and 50, which is "difficult" description style and college level in reading ability, were computed. While WSJ had twenty six articles out of sixty, the other newspapers and Web sites had more than 50 % articles that belong to 30 - 50 score level.
Another computation revealed that business news is highly above in standard reading ease level. The score less than 30 means that it is "very difficult" style and college graduate level. New York Times and CNNfn had serious problem in reading ease level. Both media had 18 articles each, which was 30 % of entire articles. Even though the other four versions showed better situation in terms of less than a score of 30, they also had considerably high portion from 8 to 15 %.

Complex Business Terminology
Overall, newspapers had only a few terms that were selected by students as complex. On the other hand, three Web Sites had more complex terminology. The following were selected ones.
WSJ had only one complex term, which was insolvency. New York Times had five, which were mutual fund, B2B, B2C, Treasury securities market, and solvency. USA Today had two, which were Consumer Price Index, and bear market. CBSMW had eight, which were chip, overnight rate, liquidate, passive stakes, ratings of Double-A-Plus, Aa1, Nikkei Average, and call money. CNBC had six, which were options trading, net-interest margin, operating earnings, B2C, mutual fund, and non-attribution rules. CNNfn had nine, which were Nikkei Average, TOPIX index, Consumer Price Index, CAC 40 index, FTSE 100 index, MIB 30 index, AEX index, valuations, and mutual fund.
It should be noted that this results only reflect the college students' reading level. If respondents differ, the results will be changed.

Visual Difference
The visual types were coded into photos, computer graphics, and data. The data included tables, charts, and graphs. In terms of quantity, the three Web sites used considerably more visuals than the three newspapers did. The number of visuals used in the articles was 70 in CBSMW, 103 in CNBC, and 79 in CNNfn. On the other hand, newspapers showed fewer visuals. Wall Street Journal used 42, New York Times used 56, and USA Today used 38.
While all three Internet news sources mainly used the data visual formats, such as chart, table, and graph, (CNBC, 80.6%; CBSMW, 64.3%; CNNfn, 53.2%), the three newspapers used photos as the main visual type (Wall Street Journal, 73.8%; New York Times, 66.1%; USA Today, 50%). Based on the company, the visual types differences were significant statistically (See Table 3; (2 (10, N = 388) = 119.744, p = .000).
Results based on media type showed apparent differences using visual type between newspapers and Internet news. While 64% of visuals in newspapers were photos, photos accounted for only 16.3% of visuals in on-line news. On the other hand, 22.1% of visuals was devoted to data format in newspapers and 51.5% was allocated in on-line news. It should be noted, however, that the two media types used a similar portion of computer graphic visual (See Table 4; (2 (2, N = 388) = 96.547, p = .000).

Content Difference
        Overall three Internet news sties were very close in what they covered. They heavily focused on stock market news: 41.7% in CBSMW, 56.7% in CNBC, and 35% in CNNfn. Other content showed similar results in the three Web sites. For instance, they also often reported individual firm stories following stock news. CNBC devoted 23.3% into individual firm news and CNNfn did 18.3%. Even though CBSMW had 18.3% general economy news, the three Internet sites rarely dealt with other news stories except stock market and individual firm news. In particular, combined across the three web sites, industry news and legal issues were reported the least -- all three gave less than 5%.
On the other hand, the three newspapers showed more diverse content than Internet sites did. In particular, the Wall Street Journal devoted 20.2% in national economy news, 18.3% in international news, and 15% to legal issues and executive news each. International, legal, and executive news in Wall Street Journal were the highest portion among six media companies' coverage. It also covered individual firm and industry news for 8.3% and 13.3% respectively. However, stock and consumer news were hardly seen on the Wall Street Journal front page.
USA Today also covered the national economy very often (21.7%) as did the Wall Street Journal, but the overall coverage was quite different from WSJ. In USA Today, industry news (21.7%) tied with national economy. Following those two categories, executive news (13.3%), individual firm news (11.7%), stock market news (11.7%), and consumer news (10%) were shown. In the New York Times, the most frequently covered area was individual firm news (23.3%). Following this, industry news (16.7%), national economy (15%), and legal issues (13%) were covered (See Table 5; (2 (40, N = 360) = 138.590, p = .000).
        Based on the eight categories of coverage, the content varies by media type. For the three newspapers, 18.9% of coverage was devoted to the national economy news, 17.2% to the industry news, 14.4% to the individual firm news and 12.8% to the executive news. On the other hand, the three Internet news sites focused 44.4% of content on the stock market news, 18.3% on the individual firm news, and 11.7% on the national economy news (See Table 6; (2 (8, N = 360) = 88.047, p = .000).

Discussion

The purpose of the present study was to examine the readability and content differentiation of business news, especially, focusing on the difference between newspapers and Internet News sites. The overall findings of this study suggest that the Internet business news showed a higher grade level in readability. The longer average sentence, low read ease score and higher frequency of complex business terminology also demonstrate the more difficult writing style of the Internet news.  It should be noted, however, that the New York Times was very exceptional among newspapers. It showed the longest average sentence, the lowest read ease score, and the highest grade level among the six examined companies.
        Although this study focused more on readability of business news, the content difference based on media type also showed significant differences. While newspapers focused on national economy, industry, and executive news, the Internet news devoted more coverage to the stock market and individual firm news. In addition, newspapers primarily used photos, and the Internet news used more graphics such as table, chart, and graph to analyze the data on the economy.

Readability
Every Journalism writing textbook urges students to write articles as concisely as possible. However, both newspapers and Internet news appear more difficult than "standard" reading. This is consistent with previous newspaper readability studies.
The results of readability in the Internet news were found to be higher than that of newspapers in every readability measurement scale. Since the Internet news sites are directed toward the financial information seekers, who have better understanding of background information, detailed explanations often might be not necessary.
The Internet media's primary emphasis on real time information might cause quicker writing and editing and could lead to decreased readability levels when compared to newspaper articles.
Even though all the news of the six companies is far beyond the "standard" level, the Wall Street Journal and USA Today showed higher reading ease scores compared to others. The reading ease score of New York Times appears to be the lowest, and it is close to the "very difficult" level. Since the New York Times is an elite medium and assumes its readers are well educated, it is no wonder. However, the New York Times Company has many media outlets including the Boston Globe and 15 other newspapers. In addition, it has its own Wire service, which spreads its stories to other local subscribers. Its low reading ease articles might frustrate general readers' attempts to understand the news story.
The long sentences of NYT especially seem to be a main reason to cause the lowest reading ease score. According to Flesch, 19 words per sentence is most desirable. The Wall Street Journal and USA Today showed the same number of words per sentence that Flesch suggested. On the other hand, the number of words per sentence of NYT was almost one and half times the desirable amount. Consequently, the grade level of NYT reaches the post college level.
Of course, readability is not the only factor to increase circulation, nor does a more readable item necessarily cause people to read it. Wanta and Gao (1995), for example, found that there is no relationship between reader satisfaction and readability. The Los Angeles Times jumped by 5 percent from October 1999 through March 2000, in part thanks to its new program of distributing the newspaper in partnership with Spanish and Korean language dailies in the Los Angeles area (Barringer, 2000). The industry, however, should notice the audience's basic need for concise and understandable stories.

Content Difference
        Product differentiation is the strategy that firms can employ when confronted with competition (Porter, 1980). In the multi channel environment, media firms especially need to differentiate their products from their competitors.
In the overall analysis of content categories, while most coverage of newspapers was devoted to the general economy and industry news, the stock market and individual firm news were main issues in the Internet business news.
WSJ covered the general economy, international trade, legal issues in business world, and executive news, quite intensively. Also, since WSJ isn't as timely as Internet, it compensates for this shortcoming by providing a wider range of stories. In particular, international trade and executive news are the only features that can be shown frequently in WSJ. The reputation of Wall Street Journal as the premier source for business and economics news was shown in this study, not only in the readability but also in content differentiation. It's no wonder that WSJ has the highest circulation and maintains the only successful fee-based Web version newspaper.
In the past, newspaper readers scanned the stock exchange tables more avidly than football news. However, currently someone who checks stock tables in newspapers might be considered to be primitive.
Internet news sites attract audiences with real time stock information, which cannot be provided by newspapers. Even though the stock market news includes difficult terminology and yields difficult readability, real time information is a niche that cannot be achieved by newspapers. They also utilize diverse interactive charts to give detail explanation on stock market. Their usages of visual data surpass by three times that of the newspapers. Visuals attract more attention and convey more information more quickly. Even though the Internet news sites are not desirable in readability, much usage of visuals can overcome the disadvantages. This trend may lead to increases in the usage of visuals in newspapers as well.

This present study examined newspapers and Internet news sites to compare the readability and content differentiation of business news. Even though it found significant differences, other media also provide business news, such as broadcast news, cable news, on-line news service, wire service, and magazine, etc. Future studies may examine business news through these media as well. There are several factors that can be contributed to the readability. For example, font, color and size are also important factors to influence readers in this visual age.
Content difference on media type reflected the product differentiation pursuing different target audience. The Internet news sites focused on stock market news, which appeals to the investors, and used more visuals such as chart, graph, and tables.
This could be an example of media pursuing their own target audience in this multi-media, multi-channel age. As the numerous media try to attract more advertisers by getting more readership and viewership, the content will be more differentiated for the target audience. In this context, this study also provides a foundation for other research to examine the influence of media competition to attract larger audiences.
The impetus for more profit and the possibility of an upcoming economic recession could drive people to pursue even more financial, business, and economic news. The only way to approach the rapidly growing audiences' need for business and economics is to make sure that the content is available and understandable.
References
Anderson, J. (1966). The Readability of Australian Newspapers. Australian Journal of Psychology, 18, 80-83.

Aronson, K., Sylvie, G., & Todd, R. (1996). Real-time Journalism: Implications for News Writing. Newspaper Research Journal, 17 (3-4), 53-67.

Bain, C. (1980). Newspaper Design and Newspaper Leadership: A Series of Four Experiments. Research Report 10. Bloomington: Indiana University School of Journalism Center for New Communications.

Barkin, S. M. (1982). Changes in Business Sections, 1931-1979. Journalism Quarterly, 59 (3), 435-439.

Barringer, F. (2000, May 2). Newspaper Circulation on the Rise. NewYork Times, p. C 12.

Baxter, W. X., Quarles, R., & Kosak, H. (1978). The Effect of Photographs and Their Size on Reading and Recall of News Stories. A paper presented at the AEJMC Annual Convention, Seattle, WA, 1978.

Burgoon, M., Burgoon, J., &  Wilkinson, M. (1981). Writing Style as a Predictor of Newspaper Readership, Satisfaction and Image, Journalism Quarterly, 58 (2), 225-231.

Carswell, H. (1938). Business News And Reader Interest. Journalism Quarterly, 15, 191-195.

Catalano, K. (1990). On the Wire: How Six News Services are Exceeding Readability Standards. Journalism Quarterly, 67, 97-103.

Chanll, J. S., & Dale, E. (1995). Readability Revisited: The New Dale-Chall Readability Formula. McNaughton & Gunn, Saline, MI.

Chyi, H. I., & Lasorsa, D. (1999). Access, Use and Preferences for Online Newspapers. Newspaper Research Journal, 20 (4), 2-13.

Danielson, W. A., & Bryan, S. D (1964). Readability of Wire Stories in Eight News Categories, Journalism Quarterly, 41, 105-106.

Feldman, H. D., & Aronoff, C. E. (1980). Trends in Newspaper Coverage of Business and Economics, 1968-1978. Newspaper Research Journal, 1 (4), 54-65.

Flesch, R. A. (1948). A New Readability Yardstick. Journal of Applied Psychology, 32, 221-233.
Flesch, R. A. (1951). The AP Writing Handbook. New York: Associated Press.

Flesch, R. A. (1974). The Art of Readable Writing, 25th Anniversary Edition. New York: Harper and Row.

Fowler, G. L. (1978). The Comparative Readability of Newspapers and Novels, Journalism Quarterly, 55 (3), 589-592.

Fowler, G. L., & Smith, E. J. (1979). Readability of Newspapers and Magazines Over Time. Newspaper Research Journal, 1 (1), 3-8.

Fusaro, J. A., & Conover, W. M. (1983). Readability of Two Tabloid and Two Non-Tabloid Newspapers, Journalism Quarterly, 60 (1), 141-144.

Galarza, P. (1998 June). Buzz Board. Money, 27 (6), 36.

Gale Research Co. (1999). Gale Directory of Publications. Detroit: Gale research.

Garrison, B. (1995). Online Services As Reporting Tools: Daily Newspaper Use of Commercial Databases in 1994. Newspaper Research Journal, 16 (4). 74-86.

Gunning, R. (1952). The Technique of Clear Writing. New York: McGraw-Hill.

Gunther, M. (2000, February 21). The Top Financial Website is Nobody's Fool. Fortune, 54-56.

Holsti, O. R. (1969). Content Analysis for the Social Science and Humanities. Reading, MA: Addison-Wesley.

Hoskins, R. L (1973). A Readability Study of AP and UPI Wire Copy. Journalism Quarterly, 50, 360-363.

Hynds, E. C.(1980). Business Coverage is Getting Better. Journalism Quarterly, 57,  297-304, 368.

Johns, J. L., & Wheat, T. E. (1978, December).  Newspaper Readability. Reading World, 141-147.

Li, X. (1998). Web Page Design and Graphic Use of Three U.S. Newspapers. Journalism and Mass Communication Quarterly, 75 (2), 353-365.

Monznette, J., & Rarick, G. (1968). Which are More Readable: Editorial or News Stories? Journalism Quarterly, 45, 319-324.

Morton, J. (1999, July-August). Bad News About Newspaper Circulation. American Journalism Review. p. 96.
Mueller, J., & Kamerer, D. (1995) Reader Preference for Electronic Newspapers. Newspaper Research Journal, 16 (3). 2-13.

Polansky, S. (1989). Address to the Pittsburgh Design Conference, in Daryl R. Moen, ed., Newspaper Layout and Design, 2nd ed., Ames, IA: IA State University Press.

Porter, M. E. (1980). Competitive Strategy: Techniques for Analyzing Industries and Competitors. New York, NY: The Free Press.

Razik, T. A. (1969). A Study of American Newspaper Readability. Journal of Communication, 19, 317-324.

Reiley, C. C. (1974). Can They Read What We Write? American Press Institute's Seminar, 33, 5-8.

Severin, W. J., & Tankard, J. W. (1997). Communication Theories, 4th Ed., New York: Longman.

Smith, R. F. (1983). How Consistently Do Readability Tests Measure the Difficulty of Newswriting? Newspaper Research Journal, 5 (4), 1-8.

Smith, R. F., & Smith, K. L. (1984, February). A Comparison of Readability Formulae as Applied to Newspaper Stories. Journal of Reading, 432-434.

Stapler, H. (1985). The One-Sentence/ Long-Sentence habit of Writing Leads and How It Hurts Readership. Newspaper Research Journal, 7 (1), 17-27.

Stempel, G. H. (1981, October). Readability of Six Kinds of Content in Newspapers, Newspaper Research Journal, 32-37.

Tewksbury, D., & Althaus, S. L. (2000). Differences in Knowledge Acquisition among Readers of the Paper and Online Versions of a National Newspaper. Journalism and Mass Communication Quarterly, 77(3), 457-479.

Waldman, P. & Devitt, J. (1998). Newspaper Photographs and the 1996 Presidential Election: The Question of Bias. Journalism and Mass Communication Quarterly, 75(2). 302-311.

Wall Street Journal Almanac. (1999). New York, NY: Ballantine Books.

Wanta, W. (1986). The Effects of Dominant Photographs: An agenda Setting Experiment. Journalism Quarterly, 63, 728-734.

Wanta, W., & Gao, D. (1994). Young Readers and the Newspaper: Information Recall and Perceived Enjoyment, Readability, and Attractiveness. Journalism Quarterly, 71 (4), 926-936.
Warfield, G. (1994). How to Read and Understand the Financial News. (2nd Ed.), New York, NY: Harper Perennial.
Weir, T. (1999). Innovators or News Hounds? A Study of Early Adopters of the Electronic Newspaper. Newspaper Research Journal, 20 (4). 62-81.



















Table 1. Mean Sentence length, Ease score, and Grade level of Six Companies

WSJ
NYT
USA
CBSMW
CNBC
CNNfn
F value
Sentence
19.9 (6.2)
27.7 (7.7)
19.7 (4.8)
23.0
(5.2)
24.7 (5.1)
24.9 (7.4)
15.163 *
Score
48.3 (13.9)
36.3 (12.1)
47.0 (10.7)
40.4 (10.2)
39.4 (10.4)
37.1 (12.7)
11.097 *
Level
10.5 (1.8)
11.8 (.53)
10.7 (1.6)
11.6
(.64)
11.7 (.78)
11.5 (1.0)
14.179 *
Mean (Standard Deviation)
* significant at p < .001












Table 2. Mean Sentence length, Ease score, and Grade level of Two Media

Newspaper
Internet
t value
Sentence
22.4 (7.36)
24.2 (6.03)
- 2.541 *
Score
43.8 (13.6)
38.9 (11.2)
3.744 **
Level
11.0 (1.56)
11.6 (.83)
- 4.731 **
Mean (Standard Deviation)
* significant at p < .05
** significant at p < .001














Table 3. Number and Percentage of Visual Usage in Six Companies

WSJ
NYT
USA
CBSMW
CNBC
CNNfn
Total
Photo
31
(73.8)
37
(66.1)
19
(50.0)
10
(14.3)
10
(9.7)
21
(26.6)
128
(33.0)
C.G.
5
(11.9)
5
(8.9)
9
(23.7)
15
(21.4)
10
(9.7)
16
(20.3)
60
(15.5)
Data
6
(14.3)
14
(25.0)
10
(26.3)
45
(64.3)
83
(80.6)
42
(53.2)
200
(51.5)
Total
42
(100)
56
(100)
38
(100)
70
(100)
103
(100)
79
(100)
388
(100)
Frequency (Percentage)
(2 (10, N = 388) = 119.744, p = .000











Table 4. Number and Percentage of Visual Usage in Two Media

Newspaper
Internet
Total
Photo
87(64.0)
41(16.3)
128(33.0)
C.G.
19(14.0)
41(16.3)
60(15.5)
Data
30(22.1)
170(67.5)
200(51.5)
Total
136(100)
252(100)
388(100)
Frequency (Percentage)
(2 (2, N = 388) = 96.547, p = .000












Table 5. Number and Percentage of Articles by Category in Six Companies

WSJ
NYT
USA
CBSMW
CNBC
CNNfn
Total
Economy
12
(20.0)
9
(15)
13
(21.7)
11
(18.3)
2
(3.3)
8
(13.3)
55
(15.3)
International
11
(18.3)
5
(8.3)
1
(1.7)
5
(8.3)
2
(3.3)
8
(13.3)
32
(8.9)
Legal
9
(15.0)
8
(13.3)
4
(6.7)
2
(3.3)
2
(3.3)
3
(5.0)
28
(7.8)
Firm
5
(8.3)
14
(23.3)
7
(11.7)
8
(13.3)
14
(23.3)
11
(18.3)
59
(16.4)
Industry
8
(13.3)
10
(16.7)
13
(21.7)
1
(1.7)
0
(0)
2
(3.3)
34
(9.4)
Executive
9
(15.0)
6
(10.0)
8
(13.3)
5
(8.3)
2
(3.3)
1
(1.7)
31
(8.6)
Stock
2
(3.3)
6
(10.0)
7
(11.7)
25
(41.7)
34
(56.7)
21
(35.0)
95
(26.4)
Consumer
1
(1.7)
2
(3.3)
6
(1.0)
3
(5.0)
4
(6.7)
5
(8.3)
21
(5.8)
Others
3
(5.0)
0
(0)
1
(1.7)
0
(0)
0
(0)
1
(1.7)
5
(1.4)
Total
60
(100)
60
(100)
60
(100)
60
(100)
60
(100)
60
(100)
360
(100)
Frequency (Percentage)
(2 (40, N = 360) = 138.590, p = .000





Table 6. Number and Percentage of Articles by Category in Two Media

Newspaper
Internet
Total
Economy
34(18.9)
21(11.7)
55(15.3)
International
17(9.4)
15(8.3)
32(8.9)
Legal
21(11.7)
7(3.9)
28(7.8)
Firm
26(14.4)
33(18.3)
59(16.4)
Industry
31(17.2)
3(1.7)
34(9.4)
Executive
23(12.8)
8(4.4)
31(8.6)
Stock
15(8.3)
80(44.4)
95(26.4)
Consumer
9(5.0)
12(6.7)
21(5.8)
Others
4(2.2)
1(.6)
5(1.4)
Total
180(100)
180(100)
360(100)
Frequency (Percentage)
(2 (8, N = 360) = 88.047, p = .000