Exploring the Effectiveness of Metaphors in Video Advertising - the Interaction Effect of Different Cultural Groups and Different Metaphors

Exploring the Effectiveness of Metaphors in Video Advertising - the Interaction Effect of Different Cultural Groups and Different Metaphors

Lesley Wu

7th September 2022

Metaphors are often used in contemporary advertising, and previous research has confirmed that advertisements with metaphors are more effective than literal ones. At the same time, research into the role of metaphors has become more comprehensive, moving from traditional metaphor theories based solely on literal language to the study of the interactive effects of different modalities of metaphor (multimodal metaphor). The aim of this study was to understand the differences in the responses of different cultural groups when exposed to advertisements containing different types of metaphors (needs-highlighting metaphor vs. feature-highlighting metaphor). Based on this expectation, a 2 (cultures: British, Chinese) x 3 (advertisement types: feature-highlighting metaphors, needs-highlighting metaphors, and literal advertisements) designed experiment was conducted to test.

Marketing
Psycholinguistics

Design
To obtain statistics on the extent to which creative metaphor in video advertising contribute to the effectiveness of advertisement, a quantitative research method was used in this study. To test if there was an interaction effect between cultural group and metaphor types, this experiment had a 3×2 mixed design, with a within-subjects factor of advertisement type (feature-highlighting metaphors, needs-highlighting metaphors, and literal advertisements), and a between-subjects factor of participants’ culture (British and Chinese). The dependent variables were attitude toward the product/advert and purchase intentions.
Participants
Fifty-three participants were recruited through convenience sampling and participated in the study by completing an online survey. The responses obtained from participants who either did not complete the consent form or did not answer all the questions were excluded from the analyses. This led to a total of 40 responses retained: 20 from Westerners, 10 men and 10 women; 20 from Chinese participants, 9 men and 11 women. Table 1 provides an overview of the participants’ information in this experiment. Most of the participants are currently studying at Lancaster University, some of the Chinese participants are currently living in China. As the aim of the experiment was to look at cultural differences, therefore, no specific age restrictions were set.
Materials
In the current experiment, the selection of stimulus classification conditions was based on the setting of Pan's study in 2020. However, in order to investigate the pattern and consistency of people's responses under different conditions, the number of stimuli under each condition classification was larger in this experiment. The stimuli consist of 9 video ads in total: 3 ads each for the literal, feature-highlighting metaphor, and needs-highlighting metaphor conditions. All ads featured tangible products: perfume, body wash and deodorant, with 3 ads per product covering all 3 conditions. The experimental manipulation was based on the metaphorical dimension of the advertisements. Table 2 provides an overview of advertisement conditions and the links to view them.
The metaphor condition contained at least one metaphor in the stimuli, while the literal advertisement was used as a control condition. The length of the selected advertisements was controlled to be less than 120 seconds (about 2 minutes). Advertisements that have been created in recent years were chosen, between 2012 and 2021.
Phau and Prendergast' study (2000) found that consumers associated the image of a brand with the image of its country of origin. In order to minimise the influence of consumers' previous perceptions of brand image, the advertisements chosen for this experiment were made for well-known brands, whose countries of origin were all developed countries, such as the USA, the UK and Japan.
Advertisements created from different countries were chosen; therefore, the language of the original advertisements were Chinese, English and Japanese. All advertisements were translated into Chinese and English with subtitles, which were checked by native Chinese speakers with undergraduate degrees in Japanese translation and English translation. As the video exceeds the size of the attachments that could be added to the Qualtrics questionnaire, the video advertisements with bilingual subtitles were uploaded to OneDrive and the link was added to the questionnaire for participants to view. All selected video advertisements were sourced from internet platforms.
To measure attitudes toward the ad and purchase intentions, questions were formulated based on questions previously used in marketing research (Jeong, 2008; Kim, Baek & Choi, 2012; Pan, 2020).
Attitudes towards advertisement. Participants were asked to rate/evaluate the ad on 4 scales, i.e., to what extent they agreed that the ad is ‘good’, ‘favourable’, ‘pleasant’, and ‘appealing’; the scales ranged from 1 (Strongly disagree) to 7 (Strongly agree) (Jeong, 2008).
Purchase intentions. Participants were asked to rate the value of the item being promoted, the probability of purchasing the promoted product, and the probability of recommending the products to their family or friends (Maheswaran & Meyers-Levy, 1990).
The original questions above were in English and were translated into Chinese for Chinese participants who took part in this study. The translations were checked for equivalence of meaning by a native Chinese translator researcher in English. Variables and measures in this study are provided in Table 3.

Procedure
All ethical guidelines related to data collection, and informed consent were reviewed and approved by the Faculty of Science and Technology Research Ethics Committee at Lancaster University. The data collected were anonymised upon extraction from Qualtrics. no participant information beyond the critical data is included.
All participants were asked to complete an online questionnaire. They could access the survey either via a QR code or via the shared link from Qualtrics. The questionnaire was set up on Qualtrics in English and Chinese versions. The first section included a participant information sheet and the consent form, followed by the experimental section.
In this section, each video advertisement and the corresponding questions were grouped into a separate question block, each with a link to a specific advertisement for participants to view. This was to make sure participants focus on watching and evaluating one advertisement at a time. To move to the next block, participants had to complete the question evaluating the current video and press a button to access the next question block. Participants rated the properties of each advertisement immediately following exposure to it. The order of ads presented was fully randomised and differed for each participant. To prevent participants' overall liking of the advertised brand, product or brand spokesperson from influencing their assessment of each attribute of the advertisement and obtain valid data, participants were reminded in each question block of the cautions for rating the advertisement itself with a sentence, "If you have any knowledge of the brands or products, please try to rate the following ads, by excluding your liking of them (including the celebrity spokesperson) and your current purchasing needs. " At last, participants clicked the submit button, and were debriefed and thanked for their participation. The study took approximately 40 minutes and participants were paid £6.50 for their time.
Statistical analysis
The data was examined and analysed using SPSS software. A two-way mixed ANOVA (analysis of variance) was used to examine the two independent variables, i.e., advertisement condition, within-participants, with 3 levels (needs-highlighting metaphor, feature-highlighting metaphor, literal) and culture, between-participants, with 2 groups (Chinese, British), and their effects on two dependent variables, i.e., attitude towards the advertisement, and purchase intentions.

Lancaster University

SPSS.sav

Wu2022

Chrisie Pullin

Open

None

English and Chinese

Data

LA1 4YF

Francesca Citron

MSC

Marketing
Psycholinguistics

40

Mixed ANOVA