Originaltitel	Materialien und Daten zu Research Paper "Bots Have to Be Fast: The Detrimental Effects of Response Delays in Service Chatbots and the Moderating Role of Anthropomorphism"
Handle	20.500.14123/1080
Datenart / Typ	Umfragedaten Statistische Auswertungen / Tabellen Erhebungs- / Messinstrumente Kontext- / Begleitmaterialien
Ressourcentyp	Datensatz
Autor* in / Erzeuger* in	Seitz, Lennart  0000-0003-0070-0309 (Institut für Management & Organisation (IMO), Leuphana Universität Lüneburg  02w2y2t16)
Beschreibung des Datensatzes	The dataset contains data and materials that have been created and collected as empirical basis for the third paper of the cumulative dissertation "Social Actor or Technology? Experimental Studies on the Perception of Chatbots Versus Humans and Their Implications for Anthropomorphic Chatbot Design". The objective of the present research is to enhance our understanding of when and why specific social cues in service chatbots may have adverse effects as there is still a lot to learn. Precisely, the research examines if a social cue backfires when it contradicts one of the main purposes and advantages of chatbots over human agents – enhancing the efficiency of service delivery. The research group, therefore, conducted five experimental studies in which participants either watched videos of an interaction between a customer and a service chatbot (Study 1 and 2) or interacted with responsive chatbots (Study 3–5). The researchers mainly manipulated the chatbot's response behavior (dynamic response delays vs. no such delays) and examined the impact on usage intentions (Study 1–4) and service provider evaluation (Study 5). To approach the underlying mechanisms, they tested the mediating role of a reduction in perceived usefulness in all studies and the moderating role of the application of computer-like vs. human-like schemas. The authors used both internal indicators for the participants' tendency to apply computer- vs. human-like schemas (i.e., their tendency to anthropomorphize chatbots, Study 1 and 3) or external manipulations by comparing (1) a chatbot with a human agent (Study 2) and (2) a computer- vs. human-like service task (Study 4).
Angewandte Methoden	Experiment (Webbasiert)
Schlagwörter	Künstliche Intelligenz; Chatbot; Mensch-Computer-Interaktion; Anthropomorphismus; Antwortverhalten; Natural Language Processing; Maschinelles Lernen; Soziale Reaktion; Artificial Intelligence; Chatbot; Human-Computer-Interaction; Anthropomorphism; Response Behavior; Natural Language Processing; Machine Learning; Social Response
Thematische Einordnung	Mensch-Computer-Interaktion
Untersuchungsgebiet (Geo)	Land: Deutschland
Anmerkungen zum Datensatz	Die Datensätze wurden mithilfe von SPSS ausgewertet. Die entsprechenden Ausgabe- und Ergebnisdateien sind ebenfalls im Datensatz abgelegt.
Veröffentlicht durch	Medien- und Informationszentrum, Leuphana Universität Lüneburg
Übergeordneter Datenbestand	× Übergeordneter Datenbestand Schließen Daten PhD Seitz
Zugehörige Ressourcen	× Beziehungen des Datensatzes Schließen Beziehungen des Datensatzes