Seeing through the Stars : A Journey through Human Fake Review Detection

Consumers heavily rely on online consumer reviews to make purchase decisions, while being poor at identifying fake reviews written to manipulate consumers’ opinion. Therefore, consumers’ fake review detection ability needs to be improved. This doctoral thesis used a mixed methods approach to answer the research questions (1) which cues do consumers use to detect fake online consumer reviews in real life shopping contexts, (2) how and when do they use the cues and (3) how can their detection skills be improved? First, a systematic literature revealed that only few research papers on human fake review detection are published and that the theories and methods used are mostly deductive and vary greatly. A unifying theoretical framework on consumers’ fake review detection was missing from existing research. This was a problem because interventions to improve behaviour, that are built upon theoretical frameworks, are often more effective than without. Using grounded theory approach and thinking-aloud approach I developed the Consumer Review Evaluation Model (CREM). The CREM explains that fake review detection was not a main objective for consumers when shopping online. Consumers’ focus lies on identifying information that describes the product, for this they often use consumer reviews. Reviews were evaluated using different detection cues in three steps: the relevance of the review, the reviewers’ credibility and the veracity of the review. If at any point the judgment was negative, the review did not weigh into the purchase decision. I then tested the model by developing a training usings its insights. The results revealed that the intervention based on the CREM significantly improved the fake review detection skills of participants. In conclusion, this thesis gives insights into the fake review process, embeds it within the wider context of the shopping process and proposes a training that can improve the detection of fake reviews.