Gender Influences on Preference of Pedagogical Interface Agents
Ntima Mabanza
Abstract
Ntima Mabanza
Abstract
The purpose of this paper was to examine gender influences on the preference of Pedagogical Interface Agents (PIAs) amongst a group of male and female adult computer literacy trainees. This was a comprehensive study with participants drawn from a township community in Bloemfontein, South Africa and matched with descriptors such as little or no formal post-school education, little or limited previous experience in computers and able to understand and speak English. Therefore, this study sought to investigate whether there is an association between the gender of adult computer literacy trainees and the preferred gender of PIAs. A total of 72 adults, specifically 41 females and 31 males participated. They were trained and given tasks to perform using Simulator Word Processor System (SMOS) which incorporated 10 PIAs that differ in terms of their human-like features such as of appearance, gender, voice and reality. A post-test questionnaire was used to measure participants' personal opinions about their preferences regarding each PIA that they interacted with. Similarly, Fischer's exact test was done to determine if there was an association between the gender of the participants and the preferred gender of the PIA. The post-test questionnaire findings indicated that there were only slight differences regarding the number of participants (females and males) who preferred one type over another one. It was also noted that human-like features of PIAs could influence the manner in which participants related to a particular PIA and their preference level of a particular PIA. Additionally, a Fischer's exact test found no statistically significant effect between participants' gender (male or female) and the preferred gender of PIA. It is, therefore concluded that the gender of participants (i.e. male or female) has no role on their preferences regarding the types, features, or gender of a PIA, neither on their preference level of a PIA.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The purpose of this paper was to examine gender influences on the preference of Pedagogical Interface Agents (PIAs) amongst a group of male and female adult computer literacy trainees. This was a comprehensive study with participants drawn from a township community in Bloemfontein, South Africa and matched with descriptors such as little or no formal post-school education, little or limited previous experience in computers and able to understand and speak English. Therefore, this study sought to investigate whether there is an association between the gender of adult computer literacy trainees and the preferred gender of PIAs. A total of 72 adults, specifically 41 females and 31 males participated. They were trained and given tasks to perform using Simulator Word Processor System (SMOS) which incorporated 10 PIAs that differ in terms of their human-like features such as of appearance, gender, voice and reality. A post-test questionnaire was used to measure participants' personal opinions about their preferences regarding each PIA that they interacted with. Similarly, Fischer's exact test was done to determine if there was an association between the gender of the participants and the preferred gender of the PIA. The post-test questionnaire findings indicated that there were only slight differences regarding the number of participants (females and males) who preferred one type over another one. It was also noted that human-like features of PIAs could influence the manner in which participants related to a particular PIA and their preference level of a particular PIA. Additionally, a Fischer's exact test found no statistically significant effect between participants' gender (male or female) and the preferred gender of PIA. It is, therefore concluded that the gender of participants (i.e. male or female) has no role on their preferences regarding the types, features, or gender of a PIA, neither on their preference level of a PIA.
Key concepts: Preference, Test (biology), Psychology, Association (psychology), Literacy, Social psychology, Pedagogy, Mathematics