Differences In The Number Of Adjectives In Online Reviews With High & Low Ratings In Comparison To Neutral Nratings

Emotions are known to play a huge role in the process of thinking, behaviour and decision making. Strong emotions which were prevailing in any kind of personal experience can determine the resulting judgment. The more involvement with an item the person experienced, the more she or he tends to have a personal connection with it (Bagozzi, Gopinath & Nyer, 1999; Ramanathan & Williams, 2007).

Researchers Liljander and Strandvik (1997) found a pattern that people may have an opinion on an item based on their previous experience: if a customer was satisfied he will associate positive emotions with the provided service and if not, he will remember negative emotions. The link between emotions and judgments becomes clearer when we analyse online reviews on products and services. Customers often decide to express their feelings on the Internet using online reviews (Koenig-Lewis & Palmer, 2014; Martin, O’Neill, Hubbard & Palmer, 2006).

When describing experiences with high or low evaluations, people often use implicit emotions. These emotions can be pronounced in different ways, however, the most natural way of expressing them is by using subjective (evaluative) adjectives (Bruce & Wiebe, 1999; Edo Marzá, 2013; Rittman et al., 2004,). The choice of words plays a crucial role in review writing and opinion formation of the public. Only one evaluative adjective can bring a completely different meaning to the text (Edo Marzá, 2013).

These adjectives help the author describe purely objective or more subjective features, stress the qualities that make the service worth using and reveal a lot about the writer’s attitude towards the content. The purpose of this paper is to analyze the amount of implicit emotions, in this case evaluative adjectives, in two types of customers reviews: with high evaluations (negative and positive) and neutral ones.

By analyzing psychological and linguistic features of the reviews, this paper aims at getting an understanding of how emotional writing works: whether there is a correlation between emotional reviews and the choice of words that has been used in them. As Edo Marzá (2013) stated, evaluative adjectives play a key role in the analysis of subjective expected values. In marketing prospective, better understanding of the emotions represented in reviews will help the online sellers to enhance the benefits of the service.

Thus, the research question of this paper is as follows: Are there differences in the number of adjectives in online reviews with high and low ratings in comparison to neutral ratings? Based on the findings of the studies above, the hypothesis is that the number of evaluative adjectives in polar ratings is bigger than in the average ones.

The corpus selection was made based on the site www.tripadvisor.com. It is a site where people can share personal experiences about provided services and places they visited during their travel. This site was chosen as source of data as it is considered trusted by consumers because of the information quality (Filieri, Alguezaui & McLeay, 2015). The review section on www.tripadvisor.com is designed in the way that a consumer is invited to rate the service or place of interest with grades from one to five points which are labelled with adjectives (1- terrible, 2 – poor, 3 – average, 4 – very good, 5 – excellent), to title the review and to share the personal experience with people in a special comment section. In order to test the hypothesis a corpus of eighty online reviews was assembled.

The corpus of samples was divided into two categories: forty reviews with polar ratings (“terrible” with a rating of one point and “excellent” with a rating of five points) and forty reviews with neutral “average” ratings of three points. To avoid the probable difference in the level of language proficiency and, thus, the possibility of limited vocabulary of the review writers only reviews in English language written by native English speakers were selected. The following essential clarification concerns the terminology of evaluative adjectives.

Huston and Thompson (2000) define evaluation applied to adjectives as: “the broad cover term for the expression of the speaker or writer’s attitude or stance toward, viewpoint on, or feelings about the entities or propositions that he or she is talking about” (p. 5). This paper is based on the finding that evaluative adjectives (e.g. unworthy, friendly, high-risk, costly, favourable) indicate the subjectivity in sentences (Bruce & Wiebe, 1999).

Therefore, the analysis of exclusively evaluative adjectives in the reviews has been made based on the classification of Kerbrat-Orecchioni (1980). According to this classification there are two categories of adjectives: “objective” – e.g. single, married, divorced; and “subjective” which she divides in emotional and evaluative adjectives.