26:T42b,

Nowadays, many people refer to online customer reviews that are available on most shopping websites to make a better purchase decision. An automated review helpfulness prediction model can help the websites to rank reviews based on their level of helpfulness. This study examines the effect of review title features on predicting the helpfulness of online reviews. Moreover, a new method is proposed to categorize action verbs in a review text. Text, reviewer, readability, and title features are the four main categories that are used in this article. We examine our proposed prediction model on two real-life Amazon datasets using machine learning techniques. The results show a promising performance of the model. However, feature importance analysis reveals the low importance of title features in the predictive model. It means that the title characteristics cannot be a powerful determinant of online review helpfulness. The results of this study can be beneficial to both buyers and website owners to have a deep insight into online reviews helpfulness.

27:T49a,

Mediation and conditional process analyses have become popular approaches for examining the mechanisms by which effects operate and the factors that influence them. To estimate mediation models, researchers often augment their structural equation modeling (SEM) analyses with additional regression analyses using the PROCESS macro. This duality is surprising considering that research has long acknowledged the limitations of regression analyses when estimating models with latent variables. In this article, we argue that much of the confusion regarding SEM’s efficacy for mediation analyses results from a singular focus on factor-based methods, and there is no need for a tandem use of SEM and PROCESS. Specifically, we highlight that composite-based SEM methods overcome the limitations of both regression and factor-based SEM analyses when estimating even highly complex mediation models. We further conclude that composite-based SEM methods such as partial least squares (PLS-SEM) are the preferred and superior approach when estimating mediation and conditional process models, and that the PROCESS approach is not needed when mediation is examined with PLS-SEM.

28:T444,

A firm’s public disclosure, such as its annual report, usually carries information of what it has done and/or will do. Since brand equity is a result of a firm’s branding efforts resulting from its long-term culture, strategy, and execution, an interesting but understudied issue is whether the information revealed in the text of a firm’s public disclosure predicts its brand equity. In this study, we propose that how managers think and therefore what the firm behind a brand does will help predict brand equity, with the thoughts and deeds of concern reflected in the Management Discussion and Analysis (MD&A) component of the firm’s annual reports (i.e., 10-K filings). Through text analysis and regression analysis on annual reports, we empirically demonstrate that a firm’s degree of customer orientation and a set of word categories as inferred from its MD&A component in the annual reports significantly predict its brand equity (as measured by the four dimensions of Young & Rubicam’s Brand Asset Valuator, i.e., BAV).

29:T405,

A qualitative study was conducted to explore Chinese advertising practitioners’ perceptions and interpretations of engagement in the digital era. Twenty-three in-depth interviews were conducted to collect data. Findings revealed that Chinese advertising practitioners emphasize interactive experience in defining engagement, which acknowledges multiple dimensions of engagement; they believe that the execution of engagement should emphasize content creation, media selection, and proper interaction; they tend to discuss engagement in a very holistic way, while they consider the effort versus effect dual-approach as an appropriate criterion to measure engagement, confusions, and discrepancies exist surrounding the eng