Improving Collaboration in Online Learning Groups via Automated Prompting.

Literaturnachweis - Detailanzeige

Trefferliste

Autor/in	Stoyanova, Filipa
Titel	Improving Collaboration in Online Learning Groups via Automated Prompting.
Quelle	Duisburg; Essen: Universität Duisburg-Essen (2021), 182 S. PDF als Volltext kostenfreie Datei (1); PDF als Volltext kostenfreie Datei (2); PDF als Volltext kostenfreie Datei (3); PDF als Volltext kostenfreie Datei (4) Link als defekt meldenVerfügbarkeit Dissertation, Universität Duisburg-Essen, 2020.
Sprache	englisch
Dokumenttyp	online; Monographie
DOI	10.17185/duepublico/74927
URN	urn:nbn:de:hbz:464-20211110-155401-0
Schlagwörter	Lernen; Lerngruppe; Benutzerführung; Kollaboration; Dissertation; Kommunikationsdesign; E-Learning; Unterstützung; Online
Abstract	The rapidly developing field of e-learning still faces many challenges such as high dropout rates, continuously promoting the need for better understanding of the processes beyond it. Group work is hailed as a solution of typical participation issues, however it is also prone to them too and can easily turn to be a rather frustrating element. To unfold its potential and to be satisfying, group work needs to function properly but barely does by default. Further assistance in terms of nudging others to do their share of the work has demanding conflict potential among group members, thus it should be undertaken by technology. However, the perception of automated support can vary depending on its characteristics and is not systematically examined with regard to the nudging phenomenon so far. Thus, re-examining the media equation theory, this dissertation investigated how support and especially nudging should be provided to be most beneficial and improve collaboration in terms of participation and satisfaction. Specifically, this work focused on automated prompting - the significance of its source (human vs. system) as well as on its determining characteristics such as appearance and communication manner. To this aim, three empirical studies including mixed methodological approaches were conducted. The foundations were laid with an online study which investigated the relevance of the prompting source by comparing automated and human prompting, their past proficiency and publicness as a basic message characteristic (Study 1). Results of the online experiment with a between-subject design 2 (prompt publicness: public vs. private) x 2 (source: system vs. human) (x 2 (only human´s past proficiency: high vs. low)) showed that public prompts were assessed as more persuasive but also more negative. Moreover, compared to messages from average and low proficient team members, messages from an automated system were assessed as more persuasive and positive but also came along with a higher feeling of guilt for the negative feedback and (potentially linked) with a higher negative emotional affect. Furthermore, the characteristics of an automated prompting system were addressed within two field studies, combining behavioral and survey data. First, personification of an automated system with the aid of an appearance and a name was investigated (Study 2). The one-factorial between-subjects design in a field study within 6 weeks (abstract vs. personified system with appearance and name) revealed more persuasiveness and satisfaction with an abstract system and that the equal participation predicted higher satisfaction with group work independent of the conditions. Nevertheless, the personification had also a positive effect on satisfaction explicitly when mediated by equal participation. The third study focused on the communication manner of the more beneficial, abstract system in terms of severity. Therefore, within 4 weeks in a field study by means of a between-subjects design (2 no severity vs. severity (x 2 transparency only of the severe system: high vs. low)), negative consequences for inactive members were variously signaled through vicarious punishment. The transparency of the system was varied to additionally investigate its legitimizing role. Results did not show an effect on contribution as expected - neither with a severe nor with a transparent system, although the latter one induced more fear and higher tendency to imitate inactivity. Additionally, equality of participation was worse both with severity and transparency but changed positively over time with the severe, non-transparent system. Lastly, yet importantly, equal participation once again predicted the higher satisfaction with group work over all conditions. Having re-examined the media equation theory in terms of nudging in online groups provided by humans and technological entities and having investigated the social-psychological factors related to prompting, this dissertation broadens current knowledge about the nudging phenomenon and its determining characteristics especially within online group work. This systematical investigation of an automated prompting system over time and considering both behavioral and survey-based outcomes, lays important foundations for the endeavor to prevent dropout in e-learning. (Orig.).
Erfasst von	Deutsche Nationalbibliothek, Frankfurt am Main
Update	2022/2