The corporate annual report is considered to be a formal, comprehensive and important convential mean of communication documents between managers and shareholders. It comprises "quantitative information" such as financial status; "narratives"; "photograph" and "financial graphs" four sections (Stanton et al., 2004). It is also an accurate format style of firms' advertisement to the public with company's business history, management faith and the future direction. Marston and Shrives (1991) described the corporate annual reports as the "main disclosure vehicle".
The corporate annual reports of the list companies are required to be published by many countries' governments and associations (AAA, 1966; ASSC, 1975). The requirement fulfills the demands of the public since besides the shareholders of the firms, many outside financial reports users are also of great interest to the annual reports, such as the employees of the corporate, bondholders, suppliers, customers, potential investors, public or social community and the government as well. To the readers with interests and expectations, the annual report is also a confirmation, revision and formation. It allows the investors and analysts to make forecasts and invests by gathering and analyzing past annual reports. Courtis (1997) commented that compared with other mediums, the corporate annual report is less timely but the most comprehensive data base to reflect the past performance and achievements.
The importance of the readability of annual report
Because of the importance medium function of financial annual reports, the clarity and effectiveness of the reports are extraordinary essential to the outside users. However, Smith and Smith (1971) demonstrated that financial report was no more the fairly reflect of comprehensive financial information but the preparation of the selected financial information for a special purpose.
Stanton and Stanton (2001) have observed that the main purpose of the financial reports have already moving to reduce the unfavorable effects of the corporate performance. Beattie and Jones (2000) commented that the annual report had been adopted by the managers to achieve the purpose of providing self-interested views of the company performance to the public. The gradually important role of impression management in the annual reports is mentioned in several studies. Philips and Drew (1991), Schipper (1989), Smith (1992), Tweedie and Whittington (1990) and many researchers all focused on this area.
Readability Ease Formula
The users of the corporate annual report expect an effective communication by reading the annual reports. They require simple, quick and inexpensive yardstick techniques to achieve the aim of effectiveness, Coutis(1986) pointed out that readability formula is just the technique available for the annual report users. The formula of readability is considered as a quantitative method in predicting whether the target audience can obtain the expected, valuable and comprehensive information from the annual reports. They were firstly introduced by Soper and Dolpin in 1964 atfer the pioneering work of Pashalian and Crissy(1950). Various indexes are introduced to applied in many accounting researches, especially the readability ease tests of annual reports (Klare, 1974). Over the years, various readability formulae have been derived to measure readability. Klare(1964) counted in her studies and gained a conclusion that there are more than 80 elements of styles and 70 versions of the readability had been developed at that time. Some of the derived formulas and popularly used over the years are dale-chall index, fog index, lix index, flesch index and cloze. The first four measures of readability mentioned above assume that readability prediction is a function of sentence length, unfamiliar words, syllables/and or polysyllables (Jones and Shoemaker 1994; Hewaidy, 2007). The cloze on the hand is used to capture understandability. This concept of understandability can be captured by cloze since it does not only look at textual factors but looks also at the interaction between text and subjects and respondents factors (Smith and Taffler, 1992a; Jones and Schoemaker 1994; Hewaidy, 2007). The flesch index is further divided into two types being flesh reading ease and flesch-kincaid grade level. The latter, derived by Peter Kincaid in the 1970's is reformulation of the original flesch test to measure readability basing on the minimum education level of the reader (Stockmeyer, 2009).
In 1986, Courtis (1986) concluded that there were 82 elements of styles and several formulas had been recalculated and revised after these 20 years. Furthermore, a series of new formula was added and advocated, such as Lix and Rix was suggested by Bjornsson (1983) and Anderson (1983). Several formulas even developed in to the computer visions which were much easy for the application.In 1980s, even the widely used writing soft WORDS added a new application of readability formula calculation which is a computerized analysis technique (Ingram and Frazeier, 1983).
Readability measures mentioned above have been used in studies reviewed in this paper either in isolation or in collaboration with others. According to Sydserff and Weetman(1999), Shriver (1989) held that readability formulas are inexpensive to use, objective and reliable and can detect obvious errors like excessive sentence length. This is one obvious reason for choosing one of the readability measures to conduct this study. The flesch-based method (especially flesch reading ease) is the most popular with twenty-six out of thirty-two studies using this method either alone or in addition to other tests (Jones and Schoemaker, 1994). The popularity of the flesch reading ease continuous in later year as observed in Courtis (1995a, 1998), Hossain and Siddiquee (forthcoming). Apart from it being the most widely used method, flesch reading ease compared to other existing readability methods was chosen mainly because of it being readily available on MSWord package and thus making it easier to use and error free than having to use the formulas manually. As observed by Jones and Shoemaker (1994), most prior studies manually calculated readability with the exception of Baker and Kare (1992); Smith and Taffler (1992a, 2000); Subramanian et al. (1993); Courtis 1998; Hossain and Siddiquee (forthcoming). Degnan (1983) held in Jones and Shoemaker (1994) noted that where computers are used, reproducibility, which means that more than researcher will find the same results when using the same text is not, prima facie, a problem. Thus, to enhance the reliability, that is, reproducibility of the results, a choice of a computer program deemed appropriate. Flesch-Kincaid grade level is also available on MSWord. However, it has not been selected mainly because it is not as widely used as its counterpart flesch reading ease.
2. Prior Research
2.1 Readability and Evaluation Approach Studies
Annual reports components, especially the chairman's address, footnotes and financial statements are defined as the communication tool for the reports users and other interest groups. The majority researchers agreed that easy and understandable both in communication and styles are the basic standard of effective communication (Flesch, 1964; Fry, 1988; Gunning, 1952; Karlinsky and Koch, 1983). Studies that research on the readability of financial reports were first appeared in 1948, and continued since then (Jones, 1988).
After the first introduction of Rudolf's Flesch readability formula in 1943, plenty of studies accumulated in various fields, such as Agriculture and government agencies (Hotchkiss and Paterson, 2006). Later in 1952, Pashalian and Crissy introduced Flesch reading formula into accounting research by investigating the readability of narratives of list companies' annual reports (Courtis, 1997). Courtis (1997) believed that since 1952, there have been more than 25 pieces of studies concentrate on readability of annual reports, spanning USA (Pashalian and Crissy,1952; Soper and Dolphin, 1964; Smith and Smith, 1971; Schroeder and Gibson,1990), Canada (Courtis,1986,1987, 1995,1997), UK (Still, 1972; Jones,1988; Smith and Taffler, 1992a, 1992b) and other countries and regions which official language for is English. Jone and Shoemaker (1994) gathered the previous studies of readability. They mentioned 32 studies up to 1994 and 26 pieces among them used Flesch when testing their samples. Courtis(1995) concluded that with very few exceptions, most findings of those research were consistent with the situation that annual reports are ranked as very difficult or difficult reading ease level for the bulk of adult population. The passages in annual report are far beyond the fluent comprehension skills for public as well.
Pashalian and Crissy (1952) selected 26 annual reports for year 1949 from United States Firms to investigate the readability. They introduced Flesch formula to test several 100-word passages which chosen from every other page of each report. They examined a total number of 211 samples from the 26 annual reports which means a average 8.1 samples per annual report was chosen. They gained an overall average score of 34.37 by Flesch Reading ease, with a range of 6 to 58. The scores stand for a range of very difficult level of readability to fairly difficult and an extremely low human interest in reading the annual report. That equals reading a scientific material, a professional journal and/or a high quality magazine. The analyzed their result was a proof of apathy to the importance of readability to the financial report users especially the outsiders since in 1949's the educated population was not the same as today's United States'. Only 4.5% population had completed the college and no more than 40% of the population had some high school educated background. The potential audience of the annual report was a rather small group in 1949's USA.
Twelve years after Pashalian and Crissy's(1952) study, Soper and Dolphin (1964) attempt to evaluate new changes in annual report readability level after Pashalian and Crissy's study. Soper and Dolphin (1964) use 25 of the 26 companies which examined in Pashalian and Crissy's sample. They expected to investigate the trend of the readability. They also applied Flesch ease to the samples and found that the readability level of annual report was declined from 34.37 to 28.76 in thirteen years after the research of Pashalian and Crissy(1952).
Smith and Smith (1971) focused on the application of Flesch ease to investigate the readability of financial statement footnotes. 49 corporations were chosen as the experiment's samples randomly from first 50 companies from the Fortune list of 500 large US companies for the year of 1969. The readability tests applied on four pieces of 100-word samples every 2000 words. In their research, the Flesch reading ease formula and the Dale-Chall formula were applied to the samples. The application of the two readability formulas all indicated that the annual reports were extremely difficult which require the users had a college graduation background. More specific, the Flesch score of their study was only 23.49 which meant the footnotes of these large scare firms were even difficult for postgraduate students. The scores of using Dale-Chall formula was rated as at least college level, too.
Still (1972) selected 50 United Kingdom annual reports randomly and applied Flesch reading ease to Chairman's Address of these reports. The average score gained from the readability test processed by Still was 42.51 in a range of 18.0 to 71.9 which means very difficult or fairly difficult to understand by the audience and was in low human reading interest. Still conclude that the UK's large majority of Chairman's Address were 'beyond the fluent comprehension' of most financial statement users.
Healy (1977) used the Flesch formula to examine footnotes of 50 New Zealand public companies. He applied the Flesch reading ease to the samples and gained an overall average score of 30.19 which were in a range from a high of 53.04 and a low of 7.35 only. The scores represented that the footnotes of the annual reports were very difficult to understand by the report users especially the public and parties of interest. He found that 48% of the footnotes tested were very difficult for readers and 46% can be marked as difficult due to the Flesch Ease scores. He believed that most users and readers of these financial statements will lose temple to carry on reading and would probably ignore the footnotes.
Pound (1980) chose and compared the readability of 30 Australian public companies annual reports and employee reports. He tested the readability scores using Flesch Reading ease respectively: 89% of the tested employee reports required the readers' education background at least as a student of school age 10 to 12. Meanwhile, 93% of the scores of the annual report tests were below 50 which mean the reports can be understood by the high grade students in high school or at least have the ability to enter the college. Then, he compared the scores of employee reports to annual reports. As the Flesch score shown, 75% of the employee reports were below 50 and so did 90% of annual reports. That can be concluded as the employee reports were easier for the audience than the annual reports. Furthermore, the scores gained by Pound (1980) were concluded as 'comprehension of the annual reports generally requires a higher educational attainment than employee reports' by Courtis (1986).
Then, Pound (1981) processed another readability test on 120 audit reports since he reckoned that the "audit report is an integral part of the communication of economic information" as well as "to add credibility to the financial representations of management". The 120 audit reports were chosen randomly and were applied with the Flesch readability ease. The Flesch scores were consistent with all the previous studies that the level of readability was quite low: 65% of the audit reports fell into the range from 0 to 30 and another 21% of them were in the range of 30-50. Pound deemed that the result indicated in order to understand and successfully receive the information contained in these audit reports, the reader and reports users required at least an education level of university undergraduate.
Parker (1982) argued the only approach ---Flesch reading ease which was widely applied in testing may have the possibility of lie or mass communication. He reckoned that the failure in communication to annual reports users were caused by misunderstood the needs of financial information users or by the difficulty the users met when reading annual reports. In order to investigate social accessibility of the annual reports, he suggested another approach-Fog index to test a small simple (ten reports) selected randomly from Australian companies. The Fog index test results for chairman's address was from 14 to 25.7 with an average score of 19.1 and footnotes from 16.1 to 24.4 with an average score of 20.1. These results represented that the readability of these annual reports were equal the scientific literature or technical books. He also gathered plenty of evidence to prove that readers in various countries experiencing the similar difficulty when reading the financial reports.
Jones (1988) carried out a longitudinal study to investigate the readability of the chairman's narratives for a UK listed corporate annual report for 34 years. He pointed out that the previous four longitudinal readability studies all had various limitations. Three of them were lack long period evidence as well as the other one was not only too short period but also concentrated on employee narratives instead of report to shareholders. To sum up, no studies were accurate until Jone's research. Moreover, Jone's study was also the first focus on UK listed companies. H.P. Bulmer Holdings PLC was selected by Jones to applied Flesch reading ease. 32 selected Chinaman's statements were set up as the database which dating back to 1952 covered 34 years. Seven hypothesizes regarding to the difficult level of the annual report, the trend of obfuscation and potential correlations were developed by Jone(1988). The test results showed that the annual report remained difficult for the majority readers while the readability declined in these 34 years significantly. Furthermore, the assumptions about the positive correlation between readability and company performance were unsupported by the study. The suspicion that chairman's personal preference and characterize effects the annual report readability were ruled out.
Baker and Kare (1992) processed a research to explore the relationship between annual report readability and companies' financial performance. They selected president's letter to the shareholders form 44 firms' annual report as the database. Financial variables for each corporate were obtained from the same database as well. They set up two proxies, the logarithm of the total assets and profitability of the firm, when estimating companies' performance. The return of stockholders' equity (ROE) and the net profit margin (NPM) are represented to describe the profitability. They adopted Flesch Reading Ease to analyze the readability of the 44 annual reports. The average readability index of their research is 13 which stands for the readers must at the level as a college student. The result of 39 companies in the 44 sample companies are between level 13 and level 16 which means the readers are required to have a college educated background at least and even a PHD background. Their result was consistent with earlier studies. Moreover, a correlation test between the readability and firm-scale was carried out as well. After the calculation of the coefficients, Baker and Kare (1992) indicated that the scale factor is positive correlation with the Flesch Reading Ease scores (larger firms' annual report are more readable than the smaller ones). Besides, another positive correlation, relationship between the length of president's letter and the number of common shares outstanding was proofed by Baker and Kare (1992). However, the main research result which assumed by the authors were not such consistent with expected. The result of the relationship between the Flesch Reading Ease and return on equity (ROE) were positive correlation as the assumption. However, the analysis shows that no significant correlation between net profit margin (NPM) and the Flesch test result. An unexpected result gained that more profitable firms prefer to be more verbose in president's letter.
Courtis (1983, 1986, 1988, 1995a, 1955b, 1997) started research in the area of annual report readability since 1983. His researches covered many branches such as first applied Fry, Smog, Lix and Rix to the president's letter to test the readability (Courtis, 1988); Investigated the relationship between readability and several corporate running indexes(Courtis, 1986); introduced and tested the obfuscation hypothesis in accounting narratives (Courtis, 1995).
In Courtis 1986's paper, he carried out an investigation into the relationship between annual reports readability and companies' risk return. Samples of his study were 100-word prose passages selected from Canadian annual reports. 46 different reports for 1982 and 96 different reports for 1983 were selected via the Financial Post's annual report request service. Courtis (1986) applied the Flesch reading ease and Fog readability measurement technique on three samples of 100-word from Chairman's address and footnotes of each annual report. The results of the Flesch reading ease on Chairman's address were 31.34 for the 1982 and 28.96 for 1983. The Flesch scores for footnotes were 28.06 for 1982 and 25.96 for 1983 respectively. These result indicated that to understand the chairman's statement and footnotes, the readers needed an educational attainment level of postgraduate degree to honor undergraduate degree. The findings gained by Fog index supported the Flesh reading ease. Moreover, the result didn't Mann-Whitney U tests and failed to be the proof of any assumed relationship between risk, return and readability.
In the year of 1995, Courtis (1995) first introduced annual report readability test into the area outside Western countries where the first language is English. He studied the prior researches on readability of annual reports and discovered that no studies had proceeded in Asian countries. He selected Hong Kong as the object and investigated Hong Kong companies' annual reports from 1986 to 1991. He analyzed that although the annual reports are required to prepared in English, English is normally second language in Asian countries. Under this circumstance, he selected 32 Hong Kong public companies randomly and applied the Flesh, Lix and Fog reading ease on the financial reports ended of the year 186 and 1991 to explore whether the results consistent with the western studies. A cross-country comparison between USA, UK, Canada, New Zealand and Hong Kong was carried out as well. The 100-word passage were gathered from chairmen's address and footnotes of the sample companies and assumed four hypotheses to be tested. The first hypotheses which assumed the readability improved between 1986 and 1991 were rejected. The second hypothesis was concluded by Courtis (1995) as "industrial classification does not appear to habe any systematic bearing on annual report readability levels." Moreover, the message gain by the tests showed that the readability was out of the fluent comprehension for 90% of the population in Hong Kong.
2.2 Studies of Chairman's statement
The number of studies which demonstrated the importance of Chairman's Statement is plenty. Epstein and Pava (1993), Anderson and Epstein (1996), Bartlett and Chandler (1997), Clatworthy and Jones (2001, 2006) all had focused on this area. Clatworthy and Jones (2001) represented that an easy-to-read summary of company performance and activities can be gained by Chairman's statements which is widely used by financial analysts and investors of interest (Smith and Taffler, 1992; Bartlett and Chandler, 1997). Bartlett and Chandler (1997) carried of an investigation to shareholders in UK to explore the most read section of the annual reports. The result showed that among 17 sections of the annual reports, the Chairman's Statement is indeed the most read one. Shareholder had a faith that future management decision and company truly financial activities can be founded in the Chairman's Statement. Hence, it is an important factor which influences the investors' decision making (Abrahamson and Amir, 1996). For instance, Smith and Taffler (1995) reckoned that all the future financial activities even bankrupt can be predicted from the Chairman's Statement. The crucial role the Chairman's statement plays are confirmed by many researchers. Abrahamson and Amir (1996) demonstrated that the two main characteristics decided this important role in impression management: its relative freedom in reporting selected information instead comprehensive information and its lack of regulation on the reporting style and disclosure details compared to other part of corporate annual reports. Due to these characteristics, the Chairman's Statement significant affects the individuals' investment decision and expectation of the future performance. Moreover, it also may affect the investors' judgment of the future share price (Kaplan et al, 1990). Although Holland (1998) demonstrated Chairman's statement is not response for or at least is not the direct influence factor of the share price, Clatworthy and Jones (2006) represented that the financial statement has the effects of enhancing and contextualize a series of information which are price-sensitive.
3. Present Study
3.1 Design and Structure
Major of prior studies focused on the readability of annual reports were published in last century. Especially the majority tests of readability were undertaken in 1980s and 1990s (). Almost twenty years past, seldom researches have been carried out in this field. The current information of readability ease of the listed companies is lack in financial area.
Courtis (1995) assumed that the readability of corporate annual reports should have improved along with the growth of public caution and academic studies. However, his assumption was rejected by statistic tests (Courtis, 1995). The present study is attempting to investigate the current readability of corporate annual reports. By comparing with the prior studies, a trend of the annual reports' readability is expected to be discovered.
Furthermore, the potential relationships between the readability, corporate business index and company performance indexes such as profit index and size factor will be tested in this paper as well. In previous studies of the corporate annual report readability, for instance, Jones' work at the year of 1988, his assumption of the correlation between readability ease and corporate performance was failed the statistic test and finally rejected. However, Jones' (1998) assumption advocated and supported by many other researchers (Coutris, 1995). This paper will carry on his assumption to investigate and detect possible evidence from the readability ease test results of these 35 London Stock Exchange companies.
At last, the industry classification of readability is assumed and carried out. Feng Li, who was an assistant worked in University of Minigo investigated in ? ? and advocated that the readability ease are different in different industries. His research showed the Flesch scores of monopolized industry, airlines industry, stones industry, clay industry, glass industry and concrete industry were higher than the insurance, health industry; electric industry and sanitary industry. It represented that the annual report of companies in first group were easier and more comprehensible than the rest companies. In the present study, the industry difference will be investigated and the new evidence will add to the series of studies.
3.2 Hypothesis development
Five hypothesizes are assumed in this study according to the previous studies concluded in literature review section.
Hypothesis 1. The readability of the annual reports is difficult or very difficult to the readers.
The first hypothesis is derived from the previous studies in readability ease tests. Despite the research methods, readability formulas and samples of the previous studies are diversified, the test results and conclusions are identical. They all gained the result that the corporate annual reports are in very difficult category without any exception and the readability beyond the fluent comprehension of the bulk of the shareholders, employees and other users.
Hypothesis 2. The experiment result of the Readability ease shows that the readability declined compared with previous studies.
The second hypothesis was assumed by many previous researches.
Hypothesis 3. The index of readability of annual reports has the industry distinctions.
Hypothesis 4. The level of readability of the chairman's address is positive correlated with the companies' performance.
ROCE and profit to sales will be used as the measurements to analyze the company's performance.
Hypothesis 5. Financial crises which happened in 2008 and 2009 appeared in the chairman's address as an important factor especially stressed in the less profitable companies.
4. Methodology
4.1 Data and Variables
The focus of present study is to examine the current extent of the readability of listed companies' annual report after these 20 years. A random sampling technique is applied in this paper to select the samples. London Stock Exchange is selected as the big sample since it is a mature stock exchange and also the largest one in Europe as well as the fourth largest stock exchange all over the world. Moreover, plenty of previous researches had selected it as the sample which means choosing London Stock Exchange as the sample will be more convenient in the future process of comparison with past data.
40 listed companies' annual reports were selected randomly from 2713 listed companies in London Stock exchange for the financial year started from January of 2009. All the official statement is ordered by email via London Stock Exchange official website and the annual reports service website of Investor Calendar.com. However, after gathering all the annual reports together, five companies' annual report were threw off from the total sample space. The reasons are various: one of them is lack of clear classification of the chairman's address or president's letter; other four companies' "annual reports" are actually the quarter reports or half year's reports. Moreover, the re-ordering work of the annual reports through writing emails to the corporate financial or customer service departments was failed since no one reply the email. Under these circumstances, only 35 companies' annual reports are set up as the final sample for this research.
These 35 companies are chosen randomly and covered various industries, such as: financial service, holdings, biotechnology, building and construction, chemicals, computer, technology, electronics, engineering, energy and retails and so on. All previous studies on readability of the corporate annual report have mainly focused on three sections: Chairman's Address, Footnotes to the financial statement and Auditor's Report (Courtis, 1997). Coutris (1997) deemed that these sections are constituted by prose and passages which contains sufficient length and plenitudinous information to apply the computation of formula-based readability ease scores. Due to the special and important location of the chairman's statement has, it is considered as the announcement of the management. Moreover, according to Courtis (1982), Parker (1982) and Jones (1988), the chairman's address or the President's Letter is considered to be the most the section in corporate annual report. Besides, more than 15 previous studies selected Chairman's address as the sample (Jones and Shoemaker, 1994; Courtis, 1995). According to the importance and research value, Chairman's Address or President Letter is selected to apply the readability computation in this study.
Courtis (1997) described the formula-based readability ease test as followed: "the usual approach has been to select a passage of at least 100-words, count syllables and average sentence length, insert these values into the formula and com". The similar descriptions represented by () in their researches too. According to the prior researches' result mentioned in this part, passages of 100-words are required for the readability formula approach. Hence, 100-word passages are selected randomly from each page of the Chairman's statements as the sample. There is one more thing needed to be declared, quite a large amount of companies only represent one page or two pages Chairman's Statement. After consideration for the accuracy of the sample, for these fewer pages Chairman's Statements, the 100-word passages are obtained from the beginning, middle and the end part of the Chairman's Statement. To conclude, for each company, at least three 100-words passages are obtained. The arithmetic average of these samples are calculated later to represent the annual's readability in the later section. The mean is rather important in the future process of research due to its basic role in observation for analyzing the readability levels and corporate attributes (Courtis, 1997).
5. Choice of readability formula
In prior accounting literature, plenty of articles analyzed these measurements with its merits and drawbacks. Among the indexes, Dale-chall (1948), Flesch (1948), Fog (1952) and Kwolek (1973) are much more popular techniques adopted in readability tests. Fry Graph (1977) and Lix, Cloz More specific, Pashalian and Crissy (1950), Soper and Dolphin (1964), Smith and Smith (1971) and Pound (1981) all applied Flesch as the readability measurement technique to complete their studies; Smith and Simth (1971) and Worthington (1978) applied Dale-chall in their works. Moreover, Parker applied Fog in his study in 1982 as well as Adelberg chose Cloze in 1979.
In this study, four readability formulas are selected. They are SMOG, Fog, Flesch and
SMOG was advocated by G. Harry Mclaughlin in the year of 1969 to calculated the texts. SMOG is short for the Simple Measure of Gobbledygook. Gobbledygook is to describe the text contains difficult contents and convoluted English and results in the difficulty to understand and comprehended for the readers. SMOG here is a readability formula to test the difficult level for readers to obtain the information from the piece of writing. The test result of SMOG is connected with the education years of the reader or in other words, the test result shows the schooling year requirement in order to complete the piece of writing and comprehend all the information discoursed in the annual report. SMOG is widely adopted in many researches especially in health-related documents (Florio, 1996 and Hedman, 2008). SMOG is considered as an easier and simpler formula and also the more accurate one as well. It counts texts and meanwhile counts a fixed number of sentences. At least 30 sentences are counted and tested by this formula, ten from the beginning, ten from the middle and ten from the end part.
The SMOG formula yields a 0.985 correlation with a standard error of 1.5159 grades with the grades of readers who had 100% comprehension of test materials.[3]
In those sentences, count the polysyllables (words of 3 or more syllables).
A version is also given which is more easily used for mental math and is sometimes known as the SMOG Index:
Count the number of polysyllabic words, excluding proper nouns, in a sample of thirty sentences.
Take the square root of the nearest perfect square
Add 3
In linguistics, the Gunning fog index is a test designed to measure the readability of a sample of English writing. The resulting number is a rough estimate of the number of years of formal education that a person requires in order to understand the text on a first reading. That is, if a passage has a fog index of 12, it has the reading level of a U.S. high school senior (which is around 18 years old). The test was developed by Robert Gunning, an American businessman, in 1952.
The fog index is generally used by people who want their writing to be read easily by a large segment of the population. Texts that are designed for a wide audience generally require a fog index of less than 12. Texts that require a close-to-universal understanding generally require an index of less than 8.
The Gunning fog index can be calculated with the following algorithm:
Take a full passage that is around 100 words (do not omit any sentences).
Find the average sentence length (divide the number of words by the number of sentences).
Count words with three or more syllables (complex words), not including proper nouns (for example, Djibouti), familiar jargon or compound words, or common suffixes such as -es, -ed, or -ing as a syllable.
Add the average sentence length and the percentage of complex words
Multiply the result by 0.4
While the index is a good indication of reading difficulty, it has limitations. Not all multisyllabic words are difficult. For example, the word "asparagus" is generally not considered to be a difficult word, even though it has four syllables.
The Flesch/Flesch-Kincaid readability tests are designed to indicate comprehension difficulty when reading a passage of contemporary academic English. There are two tests, the Flesch Reading Easiness, and the Flesch-Kincaid Grade Level. Although they use the same core measures (word length and sentence length), they have different weighting factors, so the results of the two tests correlate approximately inversely: a text with a comparatively high score on the Reading Ease test should have a lower score on the Grade Level test. Both systems were devised by Rudolf Flesch.
The Coleman-Liau Index is a readability test designed by Meri Coleman and T. L. Liau to gauge the understandability of a text. Like the Flesch-Kincaid Grade Level, Gunning-Fog Index, SMOG Index, and Automated Readability Index, its output approximates the U.S. grade level thought necessary to comprehend the text.
Like the ARI but unlike most of the other indices, Coleman-Liau relies on characters instead of syllables per word. Although opinion varies on its accuracy as compared to the syllable/word and complex word indices, characters are more readily and accurately counted by computer programs than are syllables.
The Automated Readability Index (ARI) is a readability test designed to gauge the understandability of a text. Like the Flesch-Kincaid Grade Level, Gunning Fog Index, SMOG Index, Fry Readability Formula, and Coleman-Liau Index, it produces an approximate representation of the US grade level needed to comprehend the text.
The formula for calculating the Automated Readability Index is given below:
As a rough guide, US grade level 1 corresponds to ages 6 to 8. Reading level grade 8 corresponds to the typical reading level of a 14 year-old US child. Grade 12, the highest US secondary school grade before college, corresponds to the reading level of a 17 year-old.
Unlike the other indices, the ARI, along with the Coleman-Liau, relies on a factor of characters per word, instead of the usual syllables per word. Although opinion varies on its accuracy as compared to the syllables/word and complex words indices, characters/word is often faster to calculate, as the number of characters is more readily and accurately counted by computer programs than syllables.
(838)
Results and analysis
To test hypothesis1, four readability formulas were applied to the samples. The overall mean score for the 35 selected companies' Chairman's Statement is 44.73 with a range of 28.9 to 61.
Table 1 Results of Readability tests
Reading ease Formula
Average Scores
Reading Ease
44.73
Flesh
12.48
Fog
15.28
Colaim-Liau
12.65
Smog
11.65
Table 2 Pattern of Reading Ease Ratings
Reading EaseRating
Description of Style
Eductional Attainment Level
Typical Style of Magazine
0-30
30-50
50-60
60-70
70-80
80-90
90-100
Very difficult
Difficult
Fairly Difficult
Standard
Fairly Easy
Easy
Very Easy
Postgraduate degree
Undergraduate degree
Grades 10-12
Grades 8-9
Grades 7
Grades 6
Grades 5
Scientific
Academic
Quality
Digests
Slick fiction
Pulp fiction
Comics
