Presentation of Us and Uk Advertising in Journal and Tabloid Sunday Newspapers: Cross-Cultural Content Analysis

ABSTRACT - This study is a cross-cultural content analysis of presentational features of newspaper advertising. A sample of ads from both US and UK joumals and tabloids were analyzed by the number of items advertised, their shape, type and mode of illustration, and the use of color.

Citation:

Mark A. P. Davies and Melvin Prince (1999) ,"Presentation of Us and Uk Advertising in Journal and Tabloid Sunday Newspapers: Cross-Cultural Content Analysis", in E - European Advances in Consumer Research Volume 4, eds. Bernard Dubois, Tina M. Lowrey, and L. J. Shrum, Marc Vanhuele, Provo, UT : Association for Consumer Research, Pages: 7-12.

European Advances in Consumer Research Volume 4, 1999      Pages 7-12

PRESENTATION OF US AND UK ADVERTISING IN JOURNAL AND TABLOID SUNDAY NEWSPAPERS: CROSS-CULTURAL CONTENT ANALYSIS

Mark A. P. Davies, University of Loughborough, U.K.

Melvin Prince, Pace University, U.S.A.

ABSTRACT -

This study is a cross-cultural content analysis of presentational features of newspaper advertising. A sample of ads from both US and UK joumals and tabloids were analyzed by the number of items advertised, their shape, type and mode of illustration, and the use of color.

Presentational features differ between markets for number of items shown, whereas they differ between newspapers within the same market for type of illustration; and vary in both markets and newspapers for the use of color. These results suggest both crosscultural factors and audience demographics are selectively important for newspaper advertising.

'Ihis study involves testing an integrative theory of newspaper advertising, which postulates that presentational features are reflectors of advertising strategy. Presentational features represent design and layout dimensions, such as the number of items advertised within an ad, the shape, the use of photographs or illustrations; whether ads are in descriptive or decorative type-- and the use of color. Edwards and Liebowitz, (1981: 172-173) assert that a decorative type portrays the basic character of the merchandise, either in pictures and/or words, but leaves detail to the readers' imagination, designed to create an atmosphere associated with prestige. Conversely, a descriptive ad illustration shows the merchandise pictorially with accurate detail, designed to trigger immediate sales.

Our theory builds on theories of cognition and perception from advertising psychologists. Tolley and Bogart (1994) assert that newspaper readers who are not necessarily prospects at the time of ad exposure will use peripheral vision to scan vast quantities of newspaper material and then focus more exactly on specific ads that attract their attention. According to Berlyne (1966:182-187), attention is increased by an unexpected or novel stimulus. Our theory purports that presentational features, in concert with other creative approaches, serve to build attention and to uniquely position the merchandise or store by improving reputation, exclusivity or prestige. Thus single-item ads convey distinctiveness, unique ad shapes attract attention (Lund, 1947; Bogart, 1967), photographs convey authenticity that artistic illustrations lack (Ball and Smith, 1992); color advertising communicates attention (Lund, 1947) and exclusivity; whilst decorative illustrations enhance store or product prestige (Edwards and Liebowitz, 1981).

Advertising strategies incorporating presentational decisions can be evaluated for their cost effectiveness. Our theory suggests there are strategic trade-offs between the costs associated with the choice of advertising presentation alternatives (as described above) and advertising frequency. This would suggest that academics might devote more attention to the presentational aspects of advertising, which is often neglected in advertising strategy research.

The theory focuses on strategies in advertising presentation fo-r newspapers with different audience profiles. Marketers segment their audiences on a variety of demographic characteristics such as age, income level, educational achievements and geographical area(Zeithaml, 1985; Slama and Tashchian, 1985; O'Brien and Ford, 1988; and Kahle, 1986). Targeting strategies capitalize on the strategic choice of presentational alternatives. All things being equal, education level and socioeconomic segments vary in responsivity to varying types of advertising presentation. When different newspapers cater for different socio-economic audience profiles serving different regions, we expect differences in the presentational features of advertisements.

Newspapers can be categorized as quality journals that offer more news coverage than the compacted tabloids. Journals report on the news more authoritatively, and less sensationally than tabloids. Journals are designed for a readership comprised of predominantly middle to higher socio-economic groups, whereas the readership profile of tabloids tends to be comprised of predominantly lower socio-economic groups (NRS). Applying our integrativc theory, we hypothesize that presentational features vary between newspapers according to their distinctive reader profiles within a country. Specifically, we examine whether journals are more likely to use presentational features that we have suggested can uniquely position the store or merchandise. These presentational features may vary more between journals and tabloids within a country than between newspapers comprised of similar audience profiles from different regions or countries. We argue that socioeconomic segmentation is more attuned to both the needs of advertisers (insofar as the medium supports the message) and with the expectations of their respective audiences.

Our study adopts a cross-subcultural approach, by focussing on regions rather than countries, as advocated Samiee and Jeong (1994). We test for whether differences in presentational features will be greater between journals and tabloids within the same subculture compared to that between journals or between tabloids across subcultures.

We suggest this will be moderated by the characteristics of retail advertising. Since much of regional advertising is retail (Hathcote, 1995; Nowak, Cameron and Krugman (1993), any similarities in presentational features found between tabloids and journals might reflect the conditions germane to retailing. First, retailing often deals with merchandise that people need regularly with some buying urgency (Park, 1993), so ads are created quickly for the market, reflected by fast track advertising from initial conception to production in media. Quick decisions may appear best served by following popular presentational approaches in advertising. Second, many retail advertisers are constrained by smaller ad budgets, severe competition and cost pressures (Jackson and Parasuraman, 1988). If retailers need to base their design decisions on economies of scale, by virtue of their vast volume of newspaper insertions, investment may be prioritized into delivery and service rather than creative diversity in presentation. This might help explain advertising similarities between newspapers as compared to other media, even if readership profiles are different.

The retail connection amplifies our theory. Retailing requires a need for advertising exposure to be direct and fast (reducing the time available for investing in the quality of exposure). Cost is likely to feature heavily in advertising strategy, which may be reflected in the choice of presentational features. Due to the fast track and cost-conscious nature of retail ads, we expect a greater incidence of multiple items, vertical formats, drawings, descriptive illustrations, and use of black and white.

Our study examines ads for two distinctly different demographic readership profiles represented by journals and tabloids within each region, but which are broadly similar across regions. We choose two quality broad sheet newspapers and two tabloid newspapers, one of each representing either the Greater London and South East market of the United Kingdom, or the New York metropolitan market of the United States. This is the first transnational advertising study that compares variations in presentational features, based on newspaper vehicles between regions or subcultures. Ads are also examined by store format and product, should there be different distributions between the two regions. In essence, region or area stands as a proxy for subculture, whereas newspaper vehicle might account for intra-subcultural differences.

METHOD

In order to summarize a vast amount of manifest pictorial and text data, content analysis was considered an appropriate way of classifying data objectively and systematically (Berelson, 1952; Kassarjian, 1977; Weber, 1990). We developed a classification scheme for coding ads based on selected products and presentational features of advertising design that ensured mutual exclusivity of categories (Ball and Smith, 1992). The classification was pretested for its coding suitability. The scheme offered explicit guidance to coders in making appropriate selections that best describe the manifest content for each ad.

Coding was by newspaper section, product, and presentational feature. Presentational features of ads (with response categories) were coded by number of items featured (as single, omnibus or group items), shape (as vertical, horizontal or square), illustration mode (as drawings or photographs), illustration type (as descriptive or decorative), and use of color or black and white, based on their manifest content as described by Edwards and Liebowitz (1981). Products were coded in terms of the merchandise featured in the ads, classified as auto, red goods (i.e., fashion goods and jewelry), electronics, home furnishings and 'other' (containing one or more of the preceding items). Each response category was tightly defined operationally in the form of a code book, following definitions advanced in Edwards and Liebowitz, (1981); Moriarty, (1991).

Sampling

A census of ads for the most frequently advertised products was taken. These products were pre-determined by content analyzing and tabulating a smaller sub-sample of ads drawn from the same newspapers. This procedure ensured products important for advertising revenues were examined. Replicate ads were then removed from analysis to reduce any skewing tendencies, in accordance with Stem and Resnick, (1991).

Selection of medium

Newspapers are typically the most widely used ad medium (Frederick-Collins, 1992; Dunn, Barban, Krugman and Reid, 1991), and have been rated as most appropriate medium for local businesses in a study for using non-national advertising media (Otnes and Faber, 1989).

Selection of newspaper vehicles

Journal and tabloid newspapers were chosen that provided a complementary readership composition between US and UK cities, both in terms of geography (i.e., the urban coverage of the respective cities), and level of sophistication (i.e. tabloid versus journal). The London Times and the New York Times (NY Times) represented journals, and The London Express and New York Daily News (Daily News) represented tabloids. Although all newspapers are available in a number of major cities within their respective countries, specific retail advertising varies regionally by city and should be expected to reflect retail distribution patterns. We chose Sunday editions for our analysis because they are replete with advertising. 'Be NY edition of the NY Times was chosen because it is the closest parallel to the London edition of the Times. Each newspaper covers an entire metro area, with the NY edition covering four or more States, while the London edition covers Greater London and the South East. In order to generate comparisons between different readership profiles, we selected tabloids comparable in profiles, but different from the quality journals. The Daily News has a similar coverage (in terms of numbers of readers) to the Times, and was chosen as a prototypical, popular tabloid. An equivalent newspaper chosen for the Greater London area was the Sunday Express (hereafter referred to as the London Express). According to National Readership Survey data, the percent of readers who are classified "middle class" by a socioeconomic index is 86% for the London Times and 60% for the London Express. The 1998 Gallup Media Usage Study a showed a roughly parallel contrast between the New York Times and the Daily News, using education as a proxy variable for social class (80% vs. 43% attended college),

Selection of time period studied: Examining retail ads leading up to Xmas

In order to keep our study manageable, we restricted our sample to ads shown in the Xmas season. All issues from November 17th to the 15th December 1996 were used, considered to provide a sufficient sample size to offer generalizations about the data (n=694 ads in total). We chose this period because we expected more advertising over this period, reflecting an intensity of retail competition, for several reasons. First, the Xmas period is the most important time of the trading year for most retailers, in terms of generating extra volume, improved margins on seasonal purchases (Schiffman and Kanuk, 1997). Traditionally, the ritual of gift giving over Xmas is celebrated in similar ways in both the States and the UK (Belk, 1975). This has meant a prosperous time for retailers because additional items arc purchased beyond those normally considered by the purchaser.

Selection of countries and regions

By treating heterogeneous populations as simple cultures, true cross-cultural differences cannot be detected. For practical purposes, we focus our attention on two complementary regional areas of business importance, in which these differences arc considerably reduced. The United States and the UK were chosen as countries from which to draw regional comparisons because they are widely regarded as having similar economic and cultural orientations (Samovar and Porter, 1994). The advertising industries for both nations are at a similar level of maturity for each country (Katz and Lee, 1992), with similarities in media, language and literacy rates. Both countries enjoy a substantial middle class, of great potential for segmenting media by traditional demographics (Schiffman and Kanuk, 1997). Nevertheless, despite several cultural similarities, some important culturally derived differences in ads might still be expected (Frith and Wesson, 1991). Moreover, ad presentation may show a customization that represents different groups or segments, reflecting the different subcultures studied.

Our choice of subcultures was selected because of their importance to advertisers in terms of their concentration of population and similar spread in terms of demographics. For example, the comparability of educational standards for each region indicates that literacy rates would be similar.

The judges

These were selected from students from the two countries who had management degrees. Two judges were selected who independently coded the ads for both the US and UK data. The ads for each newspaper were numbered for each edition to ensure the coders used the same order of coding. Both judges were subjected to a similar length of training and exposure to the advertising material, in accordance with Wimmer and Dominick (1991, 157158). Each judge was provided with the same code-book for study and then independently coded a two-week sample of data to test their understanding of the coding system. Both obtained feedback from the investigators.

TABLE 1

RELIABILITY DATA

Reliability Checks of Coding

After refining the explanation of categories, a further subsample (20% of the data) was reanalyzed to ensure no future problems. The complete sample was then coded. Intercoder reliability was then tested. Table 1 shows the number of categories for each feature, the number of ads whose classification was originally agreed on for each feature, the percentage of advertisements initially agreed on, and the reliability index Ir, recommended by Perreault and Leigh (1989). This measure of inter-coder reliability is more robust than the more prevalent coefficients of agreement technique, since the latter does not take account of the laws of chance in reaching a coding decision. All reliability values were above 0.9 and well within the acceptable levels recommended by Kassarjian (1977) and Nunnally (1978). In accordance with Kolbe and Burnett (1991: 249), intercoder reliability was calculated for individual categories to prevent potential concealment of pooled results overstating any reliability estimates overall. According to Krippendorf (1980: 3), a content analysis with many features cannot make use of a single figure to assess overall reliability. It cannot be assumed that reliable items compensate for unreliable ones.

Intracoder reliability was interpreted by asking each coder how they had coded the prior sub-sample to test their level of understanding and concentration levels. In the pre-testing, some responses showed differences between coders of >l 0%. Originally, one coder had some difficulty in distinguishing between ads containing group items and those containing omnibus items, treating group items too narrowly because the term omnibus refers to a diverse range of items, and indiscriminately treating too many ads are decorative when there was a prevalence in the descriptive mode. Further training involved feedback from the researchers that clarified definitions and reduced confusion.

ANALYSIS

Multi-way frequency tables were constructed for each research question and differences between newspapers were analyzed by chi-square analysis. Where contrasts between the regional markets were observed to be different, further significance tests were conducted using Yates' correction for 2x2 tables, where the critical value for the original table was used as the standard for the reduced tables.

Retail formats

Due to the difficulty in classifying ads by retail format, these were post-analyzed by the researchers for both regional markets. We categorized ads as retailer, manufacturer or co-operative advertising, with retail ads subdivided into specialty chains, independent specialists or department stores. Specialty chain stores and specialty independents were defined according to Jarrow et a]., 1987, and actual names were verified by the Retail Directory , 1998 (for the UK sample) and Dun's Regional Directory for the US sample. Since there were relatively few out-of-town stores in our sample, these were classified with specialty chains. Department stores were verified by the same sources, associated with a wider range of merchandise than specialty stores. The greatest difficulty in classification was found to be auto ads. Some of these were found to be co-operatively funded between manufacturer and retailer, with both names prominent in the body copy of the ads.

RESULTS

There were 694 ads examined in total. Of these, the NY Times accounted for 399, the Daily News: 157, the London Times: 62, and the London Express: 76. The relatively smaller sample size from the UK does not reflect differences in sampling design between the two countries. It reflects the comparative incidence of advertising between the two countries during the period of data collection.

Table 2 provides frequencies and percentages in support of our findings. Results for the differential incidence of multiple item advertising showed that certain newspapers were more likely to have advertising with several items (c2=85.5, p=0.001, d.f.=3).

Contrary to our integrative theory, the incidence of multiple items was not significant between journals and tabloids, but significant between the newspapers in New York and Greater London, with c2=72.2, p=0.001, d.f.=3. The percentage of ads with multiple items was 74.1% and 88.5% for the NY Times and Daily News respectively, compared to only 40% of the advertisements for each of the Greater London newspapers.

As expected, we found that the vast majority of newspaper ads were vertical in shape. The proportion ranged from 80.3% for the London Express to 88.2% for the Daily News. However, the use of vertical ad shapes did not differ significantly across newspapers, (not supporting our integrative theory), with no statistical contrast necessary.

TABLE 2

ADVERTISING LAYOUTS OF NEWYORK AND GREATER LONDON NEWSPAPERS

Indicates a statistically significant difference in the distribution of presentational features across newspapers at p<05. Indicates a statistically significant difference in the distribution of presentational features across newspapers at p<01. + At the time of this study, New York newspaper advertising was entirely in black and white. A totals less than 694 due to additional categories that were subsequently incidental in frequency.

Unexpectantly, the majority of ads showed photographs rather than drawings. The Daily News used photographs in 86.7% of advertising illustrations, while the other newspapers used photographs in 95% of ads. Only a contrast between the Daily News and the other newspapers resulted in overall significance (c2=8.4, p=0.05, d.f.=3).

Findings for the type of illustration varied by the kind of newspaper within each country which support our integrative theory. The NY Times is significantly more likely than the Daily News to emphasize decorative illustration (71.8% vs. 65.0%), with c2=21.4, p=0.001, d.f.=3). The difference is even more pronounced for the Greater London newspapers sampled. The London Times is significantly more decorative than the London Express, with 83 % and 60 % respectively (c2=15.5, p=O. 01, d.f.=3). Differences between the newspapers on type of illustration were significant (c2=9.9, p=0.05, d.f.=3).

In examining variations in the use of color vs. black and white advertising, this was only relevant for the UK, at the time of the research. It was found that the use of color was significantly greater for advertising in the London Times than for the London Express, supporting our integrative theory (41.7% vs. 16%, with c2=9.8, p=0.05, d.f.=3).

DISCUSSION

Due to the small sample sizes of the Greater London ads, the research questions were not analyzed by retail format or product. Future research requires larger sample sizes and sample replicates over the same time frame for several years. Instead, observational inferences are reported where warranted based on the distribution of product and retail formats. From all the presentational features studied, only the findings of multiple item ads suggest a greater cross-cultural influence than a demographic influence by newspaper profile. The prevalence of multiple ads in the NY area may reflect the different retail mix between the NY and Greater London areas, possibly with NY retailers requiring a greater support for the amount of advertising from promotional stores and discounters there. Manufacturer ads are far more prevalent in the Greater London market (80% for the London Times, 56% for the London Express), compared to the prevalence of retail advertising in the NY market (85% for the NY Times, 97% for the Daily News). Proposition 1: Retail advertisers in the NY market consider the use of multiple-item ads more appropriate than in the Greater London market, with comparatively more retail ads than manufacturer ads in the NY market, and more retail ads associated with promotional stores. Promotional stores may be more acceptable to a greater number in the US due to their bargain-driven culture, which may be associated with a down-market image in the UK market.

Analyzing ads by product shows the percentage of home furnishings and red goods advertised is 51 % in the NY Times and 45% in the Daily News, compared to 12% in the London Times and 14% in the London Express. It is noticeable that home furnishings and red goods are often advertised as multiple items within department stores or specialist chains. Advertisements from these store formats appear more frequently in the NY market. There is a prevalence of business-to-business products (e.g., office electronics) advertised in the Greater London market. Collectively, by observational inference, the distribution of product and retail format appears to influence advertising presentation. Proposition 2: The prevalence of multiple items in the NY market reflects differences in the distribution of products advertised there compared to the Greater London market.

According to a US copywriter, when merchandise is emphasized, price-featuring might also be promoted. Examining the merchandise advertising more closely revealed a pattern of promoting bargains using special promotions for the NY market rather than promoting clearance lines. This may reflect the competitive intensity at Xmas, in which many retailers may need to offer loss leaders to generate store traffic and hence stimulate additional sales. A managing director of a UK advertising agency suggested the greater emphasis of merchandise advertising in the States is partly a reflection of different business philosophies between UK and US advertising. We were advised that customers are generally less trustworthy in the States, with an emphasis on direct, hard sell approaches. Proposition 3: The greater weight of multiple item ads in the NY market reflects different selling philosophies. In NY, they are more likely selling the merchandise than the image or reputation of the supplier (whether manufacturer or retail store).

The prevalence of vertical formats was expected, but differences were not found between newspapers or regional subcultures. Proposition 4: Ad shape is not directly influenced by demographics nor culture, but more by the constraint of the shape of newspapers. This suggests the use of vertical formats logically improves readability of the advertising and is consistent with the columnar appearance of newspapers.

Neither culture nor demographics appear to influence choice of mode of illustration. Proposition 5: The prevalence of photographs over drawings for each newspaper is attributable to the greater opportunity for projecting the merchandise of retail ads in a more realistic, and therefore convincing manner.

Proposition 6: The relatively greater amount of decorative ads relative to descriptive types of illustration in the journals reflects the greater incidence of luxury or added value items advertised there, that are more likely to be uniquely positioned. However, the predominance of decorative ads, even in the tabloids, is surprising. One explanation for the popularity of decorative ads is the sense of store atmosphere they generate, which can help improve receptivity to new products designed for the New Year.

Proposition 7: The popularity of using color ads in the London Times suggests they are more appropriate than black and white ads for projecting high quality and taste, geared to a younger, upscale audience. But care should be taken in weighing up the cost effectiveness of using color ads, which ultimately should depend on media objectives linked to the products advertised. Hence the London Times, with a greater incidence of premium priced products advertised, may offer greater margins to advertisers to sustain the additional costs of color advertising.

IMPLICATIONS

The findings offer some insight into how advertisers use presentational features to support their advertising strategy in the targeting of particular audience profiles. We conclude this is contingent on the specific presentational feature studied. The different frequencies found in type of illustration and use of color between newspaper types appear to support our integrative theory, together with the popularity of multiple items, vertical shapes, and black and white ads, can be explained by a strong retail presence. But, making decisions about the shapes of ads may simply reflect newspaper advertising practice without a conscious decision to consider shape to reinforce advertising strategy (since unique shapes that can grab attention were found to be rare). In addition, the incidence of multiple items may be a better reflector of cultural effects. The comparative abundance of multi-item ads for the NY area reflects the greater degree of competition there for ad space, which ironically further adds to clutter. The average amount of display ads per page for the US newspapers combined is 1.49 in comparison to only 0.91 for the Greater London newspapers. For some key areas of presentational features, competition may serve to encourage repetition rather than to contribute to a unique position.

Academics can learn from comparative studies such as this by reflecting on normative practice across newspaper types, stores and regions serving different audiences. Advertising practitioners may wish to consider the marginal effects of using additional promotional features to support their advertising strategy, by comparing practices across audience profiles and regional markets.

LIMITATIONS

First, we discuss the I imitations for analyzing visual data using content analysis. A general problem of content analysis is that it cannot reveal underlying processes that determine content patterns. The symbolic character of communication is restricted to its manifest content (Ball and Smith, 1992). Therefore, our explanations and discussion of our findings are simply conjectures that cannot be proven by content analysis. However, they offer useful propositions for further research.

Second, observed differences may arise from the presence of uncontrollable variables such as product category specific effects rather than cultural orientations. We acknowledge this as a separate issue for further research.

Third, it can be argued that examining design elements alone are only one aspect that offers a total image to the reader. Further research might investigate how advertising content might mediate the effectiveness of design and layout elements. Fourth, it can also be argued that a content analysis merely highlighting differences between design and layout practices across different audience profiles need not necessarily point to good practice nor effective targeting. But at least differences suggest retail advertisers are thinking beyond a mass marketing approach, by putting some thought behind different types of design. Fifth, an extension of the study across other newspapers and/ or regions might be instructive and offer more support to making generalizations on advertising design practice. Further research might fill the void in these limitations and explore further cultural parameters, perhaps those in high context societies.

REFERENCES

Belk, Russell W. (1975), "Situational Variables and Consumer Behavior," Journal of Consumer Research, 2 (December), 159-164.

Ball, Michael S. and Smith, Gregory W.H., (1992), Analyzing Visual Data, Newbury Park, CA: Sage, 1-75.

Berelson, Bernard (1952), Content Analysis in Communications Research, Glencoe, Illinois: The Free Press.

Berlyne, Daniel L., (1966), "Conflict and Arousal," Scientific American, 82-87.

Bogart, Leo, (1967), Strategy in Advertising, Harcour Brace and World Inc., New York, 1- 336.

Dunn, Watson, Arnold M. Barban, Dean M. Krugman and Leonard Reid (1989), Advertising: Its Role in Modern Marketing, 7th ed., Chicago: The Dryden Press.

Edwards, Charles M., Jr. and Carl F. Liebowitz (1981), Retail Advertising and Sales Promotion, 4th ed., Englewood Cliffs: NJ: Prentice-Hall, 58-188.

Frederick-Collins, J. (1992), "Measuring Media Image: Expectations, Perceptions and Attitudes of Local Retail Advertisers," Advertising Division, Annual Conference of the Association for Education in Journalism and Mass Communications, Montreal, Quebec, 53-73.

Frith, Katherine Toland and David Wesson (1991), "A Comparison of Cultural Values in British and American Print Advertising: A Study of Magazines," Journalism Quarterly, 68 (1/2), 216-223.

Hathcote, J. M., (1995), "Institutional and Promotional Newspaper Advertising Practices in the US Apparel Retailing Industry from 1971 to 199 1," International Journal of Advertising, 14, 165-179.

Jackson W. and A. Parasuraman, (1988), "The Yellow Pages as an Advertising Tool for Small Businesses," American Journal of Small Business, Spring, 29-35.

Jarnow, Jeannette, Miriam Guerreiro and Beatrice Judelle, (1987). Inside the Fashion Business, NY: Macmillan.

Kahle, L.R., (1986) "The Nine Nations of North America and The Value Basis of Geographic Segmentation,"Journal of Marketing, 50, 37-47.

Kassarjian, Harold H. (1977) "Content Analysis in Consumer Research," Journal of Consumer Research, 26 (December), 8-18.

Katz, Helen and Wei-Na Lee (1992), "Oceans Apart: An Initial Exploration of Social Communication Differences in US and UK Prime-Time Television Advertising," International Journal of Advertising, 11 (1), 69-82.

Kolbe, Richard H. and Melissa S. Burnett (1991), "Content Analysis Research: An Examination of Applications with Directions for Improving Research Reliability and Objectivity," Journal of Consumer Research, 243-250.

Krippendorf, K. (1980), Content Analysis: An Introduction to Its Methodology. Beverley Hills, CA: Sage.

Lund, John V., (1947), Newspaper Advertising, Prentice Hall Inc: New York, 117-133.

Moriarty, Sandra E. (1991), Creative Advertising: Theory and Practice, Englewood Cliffs: Prentice Hall.

Nowak, Glen J., Glen T. Cameron, and Dean M. Krugman, (1993), "How Local Advertisers Choose and Use Advertising Media," Journal of Advertising Research, November/ December, 39-49.

Nunally, J. (1978), Psychometric Methods, New York: McGraw Hill.

O'Brien, Sarah and Rosemary Ford (1988), "Can We At Last Say Goodbye To Social Class?" Journal of The Market Research Society, 30 (3), 289-332.

Otnes, C. and R.J. Faber (1989), "An Examination of Variables Influencing Local Advertiser Media Selection," Proceedings of the American Academy of Advertising, K.B. Rotzoll, ed., American Academy of Advertising.

Park, W., (1993), "Let's do something about retail advertising before it's too late", Editor and Publisher, December 4, 38 and 48.

Perreault, William D., and Lawrence Leigh (1989), "Reliability of Nominal Data Based on Qualitative Judgments," Journal of Marketing Research, 26 (May), 135-48.

Samiee, Saecd and Insik Jeong (1994), "Cross-cultural Research in Advertising: An Assessment of Methodologies," Journal of Academy of Marketing Science, 22 (3), 205-217.

Samovar, Larry A. and Richard E. Porter (1994), Intercultural Communication: A Reader, 7th ed., CA.: Wadsworth Publishing Co.

Schiffman Leon and Charles Kanuk. (1997), Consumer Behavior, Englewood Cliffs, NJ: Prentice-Hall.

Slama, M.E. and A. Taschian (1985) "Selected Socioeconomics and Demographic Characteristics Associated with Purchasing Involvement," Journal of Marketing, 49, 72-82.

Stem, B.L. and A.J. Resnik (1991), "Information Content in Television Advertising: a Replication and Extension," Journal of Advertising Research, 31 ( 3), 36-46.

Tolley, Stuart B and Leo Bogart (1994) , "How Readers Process Newspaper Advertising," in Attention, Attitude and Affect in Response to Advertising, eds., Eddie M Clark, Timothy C. Brock and David Stewart, Hillsdale, NJ: Lawrence Erlbaum Associates, 69-77.

Weber, Robert P. (1990), Basic Content Analysis, 2nd ed., CA: Sage.

Wimmer, Roger D. and Joseph R. Dominick (1991), Mass Media Research: An Introduction, 3rd edition, Belmont, CA: Wadsworth, 156-179.

Zeithaml, V.A., (1985), "The New Demographics and Market Fragmentation," Journal of Marketing, 49, 64-75.

----------------------------------------