Persuasion within Advertising: Metaphorical Expressions vs. Literal Expressions

Persuasion within Advertising: Metaphorical Expressions vs. Literal Expressions

Helen Vale

2015

The present research built upon research conducted by Citron and Goldberg (2014) on figurative language, emotion, and the brain. This study examined the three different data sets: sentences, stories and sentences with taste metaphors collected by Citron and Goldberg (2014). It examined three different data sets: sentences, stories and sentences with taste metaphors. Metaphorical and literal sentences, stories and taste metaphors were rated on emotional valence, imageability, emotional arousal, metaphoricity and similarity in meaning. Familiarity was rated within sentences and taste metaphors and understandability and naturalness were rated within stories. This study explored relationships among variables, relationships between metaphors and literal counterparts, relationships between each data set and lastly, relationships between each data set when split by type: metaphor and literal. Findings from this investigation provide evidence for marketers, of the benefits of using metaphors within advertising to increase persuasion and consumer buying behaviour. A company who wants to portray imagination, develop images within a consumer’s mind and evoke emotional arousal should use metaphorical sentences within their advertisements. Additionally, the more arousing a sentence the more imaginable, therefore, marketers should specifically employ emotionally arousing material to further engage a consumer. This study can add to literature on figurative language and persuasion. Also, provide evidence for marketers who want to increase their sales and further persuade consumers with an effective approach

Metaphor
literal
persuasion
advertising
marketing
figurative language
emotion

All metaphorical sentences and stories were created in German with words that would obtain a metaphorical interpretation. Then each word was replaced with its literal counterpart, which created: one hundred and twenty non-taste related sentences, sixty metaphorical “The bride was very moved by her wedding” and sixty literal “The bride was very happy about her wedding”. Sixty-four stories, thirty-two metaphorical “Lisa was sitting in her physics class and was still digesting the stuff from the lesson before when her teacher announced a task to bite your teeth out on.” and thirty-two literal “Lisa was sitting in her physics class and was still having problems with the stuff from the lesson before when her teacher announced a really difficult task.” Finally, seventy-four taste metaphors, thirty-seven metaphorical “She received a sweet compliment” and thirty-seven literal “She received a nice compliment”.
Specific instructions were created by Francesca Citron for each variable to be rated (See Appendix A). Sentences, stories and taste metaphors were rated on emotional valence, imageability, emotional arousal, metaphoricity and similarity in meaning. Emotional Valence refers to how positive or negative the stimulus is which was rated on a scale from -3 (very negative) to + 3 (very positive) through 0 (neutral). All other variables were measured on a scale of 1 to 7. Imageability is the ability to evoke a mental picture rated: 1 “not imaginable at all” and 7 “very imaginable”. Emotional arousal describes to what extent the stimulus is emotionally stimulating rated: 1 “not intense at all” and 7 “very intense”. Metaphoricity describes the figurativeness of the stimulus rated: 1 “literal” and 7 “very metaphorical”.
Lastly, similarity in meaning which refers to how similar the meaning of both metaphorical and literal counterparts are with regard to contents. For instance, the metaphorical sentence “He praised her to the skies” compared to the literal sentence “He praised her fulsomely”. These have the same meaning, thus the meaning similarity between metaphorical and literal sentence is high. This was rated 1 “not similar at all” and 7 “very similar/equal in meaning”.
Familiarity was rated within sentences and taste metaphors, which describes how familiar the stimulus is rated: 1 “not familiar at all” and 7 “very familiar”. Additionally, taste relatedness was measured for taste metaphors which refers to the extent a sentence is associated with degustation. It was rated as 1 “not taste-related at all” and 7 “very taste-related”. Lastly, understandability and naturalness were rated within stories. Understandability is about the easiness of grasping what the content means rated: 1 “very difficult to understand” and 7 “very easy to understand”. Naturalness is how normal and daily a story or its parts are rated: 1 “not natural at all” and 7 “very natural”.
To evaluate complexity, several measurable parameters were created. For each parameter one “complexity point” was given, therefore, creating one overall complexity score. For all data sets all 9 characteristics were the same: subordinate clauses, relative clauses, passive forms, compound nouns, appearing persons, adverbs and adverbial phrases, conjunctive forms, analytically-formed tenses/infinitive constructions and marked/deviating structure of sentence. For sentences and taste metaphors alone the number of words was also a characteristic and within stories the number of metaphors. (See Appendix B).

Procedure
Participants were each provided with a consent form to sign if they agreed to partake in the study. Once completed, participants were provided with a URL via E-mail to access the questionnaire. General instructions were shown first, followed by the specific instructions for the first variable to be rated. The words were then presented, each one at the centre of the page immediately followed by the 7-point scale. When all words had been rated for one variable, instructions for the next variable rating appeared. The order of variables were random for each participant. This procedure was the same for all sentences, stories and taste metaphors.

Data Analysis
All the means and standard deviations were calculated and used for the analyses for all sentences, stories and taste metaphors. Independent sample t-tests were then used to look at the differences between metaphors and literal counterparts of each variable within the three data sets. When there was a specific hypothesis a one tailed t-test was implemented however, when there was no hypothesis a two tailed t-test was applied.
Next, the variable emotional valenced squared was computed to represent the quadratic relationship between all other variables and then used within the following data analyses. Firstly, a multiple regression was then used to analyse any quadratic or linear relationships between emotional valence and other variables within each data set. In each regression, features of no interest were partialled out by entering them as predictors in the first step; then valence and valence squared entered in the second step. Additionally, partial correlations were conducted within each data set to look at linear relationships between pairs of variables within metaphors and literal counterparts by controlling for other variables. Lastly two types of analyses of variances were conducted, firstly, one-way between subjects ANOVAs to look at the difference between datasets: sentences, stories and taste and their impact on emotional arousal, imageability, emotional valence and metaphoricity. Then one-way between subjects ANOVAs to look at the differences between datasets when split by type, metaphors and literal counterparts and their impact upon variables.

Lancaster University

Vale2015

John Towse

Open

The rating data had been gathered already by Francesca Citron during her research in Berlin and ethical approval had been obtained at that time. The present study has been approved by the Department’s Research Ethics committee at Lancaster University.

English

Data

LA1 4YF

Francesca Citron

MSc

Psychology of Advertising

Sentences were rated by thirty-five males and seventy-eight females aged between twenty-one and sixty-seven (M = 35 years, SD = 12.23 years). Stories were rated by fifty-nine males and one hundred and forty-two females aged between seventeen and seventy-eight (M = 36 years, SD = 15.00 years). Lastly, taste metaphors were rated by seven males and nineteen females aged between twenty-two and seventy-four (M = 27 years, SD = 4.9 years). All participants were native German speakers from the Berlin area.

t-tests
regressions
correlations
partial correlations