The master’s thesis analyzes advertisements in Swedish, American and Croatian women’s magazines and it is divided into three main parts, each dedicated to a different type of analysis. In the first part, a basic quantitative approach to discourse analysis is used to determine the similarities and the differences in language structure of advertisements in the three languages. The corpus comprises a total of 405 advertisements, or 135 advertisements per language. Magazines included in the corpus are Elle, Cosmopolitan, and Allure for the USA; Elle, Cosmopolitan and Ljepota i zdravlje for Croatia; and Elle, Femina, and Plaza Kvinna for Sweden. The advertisements were divided into several different categories with exact numbers of occurrences and percentages in relation to the total sample for each specific language. The advertisements were first divided into product types, which showed that hair products, skin products, and makeup were the most represented catregories in the corpus across all three languages. Then, they were divided into categories based on Cook’s and D’Angelo’s divisions. Cook defines advertisements as either “reason” (hardsell) advertisements which describe the product and give reasons for purchase or “tickle” (soft-sell) advertisements which appeal to the reader through a more indirect way of connecting a certain mood or feeling with the product (15). The majority of advertisements in the corpus belong into the reason category, ranging from 61% to 75% for different countries. D’Angelo divides advertisements into descriptive, narrative and reason-why or argumentative advertisements, which are not exclusive, so one advertisement can belong to more than one category (356). Results show that American advertisements are more argumentative than Croatian and Swedish advertisements, which are mostly descriptive. The majority of reason advertisements are also descriptive and/or argumentative, while tickle advertisements are predominantly narrative. All advertisements for cars and perfumes are tickle advertisements, and all advertisements for travelling, as well as all weight loss and beauty salon advertisements are reason advertisements. The majority of results of the language structure analysis are similar for all three languages. Advertisements were divided by length into short (1-20 words), medium (2150 words) and long (51 or more words) advertisements. The results show that medium length advertisements are the most common ones in all three languages with about 50% percent for each country. After that, the advertisements were divided by type of address into those that address the reader and those that do not address the reader. All advertisements that use imperative and 2nd person singular or 2nd person plural personal pronoun were considered to address the reader. Due to differences in the three languages, it was impossible to divide them into more detailed categories which would still be comparable. In English, the 2nd person personal pronoun “you” is the same in singular and plural. However, in Swedish, the 2nd person personal pronoun is “du” in singular and “ni” in plural, while in Croatian, there are tree forms. The singular form is “ti”, but the plural form can be either true plural “vi” or the honorific plural “Vi”. Croatian imperative verb forms include grammatical number makrings, so it is impossible to tell whether an imperative form in plural is true plural or honorific plural. Between 67% and 83% of all advertisements in the corpus address the reader. Use of 2nd person personal pronoun in advertisements is often connected with creating a closer and more personal relationship with the audience, which is additionally emphasized when using the singular version of the pronoun (“ti”) in Croatian, because it is considered to be much more informal than in the other two languages. Furthermore, the advertisements were split into those that use some form of negation and those that do not. The results were not largely different for the three countries, with American advertisements using negation the most at 31%, Swedish advertisements using negation in 21% of the cases, and Croatian advertisements using negation the least at just 13%. Additionally, the total corpus of reason advertisements for all languages uses negation in 25% of the cases, while this is true for only 13% of the total sample of tickle advertisements. The advertisements were split by use of verb tense as well, simplified into only three main levels due to differences in the three languages, those being past, present, and future. Additional two categories were advertisements that use no verbs and therefore no tense, and those that use a combination of two or more different tenses. The majority of all advertisements use the present tense, ranging from 68% to 82% for the three countries. The future and past tense are used more in reason than in tickle advertisements, since they are usually used to express promises of the product and negative circumstances in the past which the product is portrayed as being able to change (Vaičenonienė 45-46). In the second part, the advertisements in the initial corpus were divided into those that do and those that do not use foreign language elements, and those that do were then further analyzed. The analysis shows a large difference between the three countries. Some foreign language elements are used in 54% of Swedish advertisements, 20% of Croatian advertisements, and only 1% of American advertisements. Only 7 advertisements in the whole corpus use a foreign language which is not English. The advertisements that use foreign language elements were divided into seven categories according to Martin’s division by degree of use of English elements which was adapted for this thesis to fit all languages (387-398). Her term substitution is used in a loose sense, covering both the cases in which advertisements were translated from a foreign language, keeping some of the foreign elements, and cases in which advertisements were created in the language of the target market, with purposefully added foreign elements. Out of the two American advertisements that use a foreign language, one is a phrasal substitution in French (“The Eau de Parfum”), and the other one is an isolated lexical substitution in Italian (“The Life DELIZIOSA”). In the American advertisements, foreign language elements are used to give products made in other countries an exotic and foreign identity connected to their country of origin. The majority of Swedish and Croatian advertisements that use a foreign language use substitution of expressions, which means they use foreign language phrases which are not embedded into a larger unit of language, but are still part of an advertisement containing other Swedish or Croatian elements. In the Swedish corpus, there is at least one example of each of the seven categories. Foreign language was used mainly for slogans in all three languages.In Croatian, advertisements that are fully in English use short text and rely on pictures to show everything that is described in English in order to avoid misunderstanding of the message. In the Swedish corpus, there are also much longer advertisements which are completely in English, some of them even containing word play, such as the Aussie advertisement stating “Overdoing it is Aussome [...] 3 Minute Miracle Collection (or the overdone-it-undo-ers), crammed full of hair-loving shizzle”. There are several cases of partial substitution where a word is formed by creating a compound word out of an English and a Swedish word, such as “enegibooster” (“energy booster”) and “ANTIAGE-HUDVÅRD” (“ANTI-AGE-SKINCARE”). There are even examples of phrasal and isolated lexical substitution of foreign language with Swedish, where a foreign element is directly translated – “charming tvist” for “charming twist” and “Dina” for “Your” using a capital letter, which is a form of address that is not used in the standard Swedish language. In both Croatian and Swedish advertisements, there are examples of local brands using English in advertisements intended for their local market. English is not as nationally marked as other foreign languages – it has become a global language in the West, and use of English is connected with a sense of progress, modernity and trendiness. Local brands most likely decide to use English in order to appear more global and contemporary to the local audience. In the last part, a multimodal analysis is carried out on advertisements promoting the same product and appearing in all three countries. There are two such examples, one an advertisement for perfume by Miu Miu, and the other an advertisement for jewelry by Pandora. The analysis combines the qualitative approach to text and foreign language used in the first two parts of the thesis with a visual analysis of the layout of advertisements according to elements of composition defined by Kress and van Leeuwen (179-203) and Björkvall (319-341). The first comparison of Miu Miu advertisements confirms the results of the foreign language use analysis. The Croatian advertisement is fully localized, while the Swedish advertisement is completely in English. The picture is the central element in all three advertisements, but the American advertisement uses a slightly different picture. The Croatian and Swedish advertisements are a perfect example of composition split into the left, which represents given information, and right, which represents new information. On the left, the young girl is an example of old information, someone the viewer can identify with, while the cat on the right in the domain of the new is a visual metaphor for the Miu Miu brand, offering the product to the reader. The product is placed in the center, it uses red and blue colors in opposition to the neutral colors used in the rest of the picture, and it is facing the viewer, while the girl and the cat are shown from a sideway perspective. A further division into the top, domain of the ideal, and the bottom, domain of the real, shows that the whole picture represents the “ideal”, the promise of the product, while the text in the bottom showing the brand name, slogan, and in Croatian and American advertisements also the store location, represents the “real”, practical information. The second example includes an American advertisement which differs from the other two significantly. All three advertisements use similar style of text and the word “unique” to attract the viewers. However, the Swedish and Croatian advertisements are smaller and only contain a picture of the advertised bracelet, while the American advertisement spans over two pages and is split into left and right. On the right, there is a picture of a young smiling mother with a child, while the bracelet is shown on the right side. Just like in the Miu Miu advertisement, the given left side represents the reader, while the right side represents the product. The Pandora advertisement is a reason advertisement because it describes the kind of silver used in the product, and the Croatian and Swedish advertisement also list the price. However, they also play partially on the projected feeling to attract the viewer through the use of the word “unique”, and especially in the American advertisement, through the use of a happy mother with a child. This advertisement is a clear example of how it is not enough to only analyze the text of an advertisement to get a complete picture of its meaning. The American advertisement uses the same kind of language as the Croatian and the Swedish ones,but because of its use of visual composition, it could be also considered a tickle advertisement. A lot has been written about the way advertisements create and reaffirm ideologies or manipulate readers to buy products. The two examples do show ideal women and implicitly promise qualities such as youth, beauty and happiness to those that buy the product, but it is not so easy to influence people’s opinions as it may seem. Such messages are only accepted by readers who already value those qualities, and they do not interest those who are not already drawn to the depicted lifestyle. A combination of quantitative, qualitative, and multimodal approach enables a thorough overview of the similarities and differences in advertisements in the three countries. Advertising is a field strongly influenced by globalization, so the lack of differences in the structure of advertising texts is not surprising. However, the large differences in foreign language use highlight how English and other foreign languages are used for completely different purposes, as well as how the degree of Anglicization in Sweden and Croatia differs, both due to differing expected levels of knowledge of English in the two countries, as well as openness to accepting foreign language elements in terms of language standards. The multimodal analysis shows how important visual elements are in the overall creation of meaning in advertising. However, such qualitative analyses are much harder to undertake for large samples of data, so quantitative results are used in order to gain easily comparable insights instead.