Colour as visual rhetoric in financial reporting

1 Lack of research interest might be attributed to its perception as irrelevant to some classes of professional users of accounting-based communication. Professional users typically generate their own reports about corporate performance and other matters. Digitized versions of annual reports are available at web sites and in other databases such as the SEC's EDGAR Database in the USA. Financial analysts download digitized data into computer models and generate their specific reporting needs. In such cases, it is hard to imagine how colour could impact on their judgements and decision-making. However, not all professional users follow these procedures and not all users are professionals.

2 Stereotypical colour associations are not universal because of the role of culture in colour perception. Different cultures associate some different colours with different feelings and activities.

3 Between 1971 and 2001 more than 11,000 articles containing the keyword colour were cited in ABI/INFORM CD-ROM abstracts and indexing, including approximately 800 management, marketing and business journals.

4 With regard to measurement of the frequency of colour application each page of the annual report (excluding the front and back of the cover pages) was checked for the occurrence of colour in six ways: background, words, graphics, tables, photographs and animation. The frequency of appearance in any one annual report was the aggregate of the total number of times one of the above items appeared in colour. More specifically, decision rules were determined to operationalise each of these six areas. Background was either a complete page of coloured paper (counted as one), or a coloured line occurring frequently throughout the annual report (counted as one application per page) or a colour block used for highlighting page numbers or words. Words with colour were counted as one if only a single word (or part of a word) appeared in colour, or counted as one if a block of words (such as a complete paragraph) appeared in colour. Each paragraph in colour was counted as one. Each graph or chart in colour (bar chart, pie chart, flow chart, line graph) was counted as one. Each table in colour was counted as one. Each non black and white photograph was counted as one. Animation comprised paintings, drawings and logos and were counted as one each time they appeared. The decision rules were applied consistently across the sample of 100 annual reports. Double counting was avoided; for example, paragraphs appearing against a page of coloured background was counted as one for the background, but not counted again as separate paragraphs. With regard to the measurement of colour dedication, the same six headings were used to calculate the total area of colour per page and the overall total per report. A ruler was used to measure the width and height of a colour application and converted into square centimetres, and these were added page by page across the entire report. The total square centimetres of the whole report (number of pages times the area of a single page) was then used as the denominator, and a percentage of colour dedication obtained. This was done for the reports of each company for the two consecutive years studied in order to determine whether the percentage of space used for colour increased or decreased from one year to the next.

5 The population of public companies listed on the Stock Exchange of Hong Kong at the time of the study was 581. The sample was chosen randomly and contains Hong Kong only public companies, i.e., the sample contain no foreign subsidiaries which might have conceivably affected the colours used in the experiment.

6 It is acknowledged that the experimental task is artificial. The amount of investable funds available to each student was HKD 100,000, and for a new accountancy graduate this is about one year's salary. It is therefore not a trivial sum and would take a graduate some years to save. A fair but irrelevant question is whether these subjects would allocate the same percentages of these funds if they were dealing with their own money. We are not interested in the extent to which the investment task corresponds to reality, but rather whether background paper in different colours makes any appreciable difference between group mean judgements.

7 Students were all from the Faculty of Business at one university in Hong Kong. Each tutorial group is of a fixed size of 20 registered students. At any class a few students will be absent because of illness, job interviews, pressure of assignments, time of the day, and other reasons. We could not control actual attendance. However, all attendees participated in the experiment. Group participation sizes ranged from 14 to 19, with an average of 16.75.

8 The literature suggests that ivory is preferred for business communication. Ivory was the second most frequently used colour found in the annual report sample.

9 The sample size of three males in the purple group restricts any interpretation of this finding.