Early Design of a Conversational AI Development Platform for Middle Schoolers

Aug 19, 2022

Early Design of a Conversational AI Development Platform for Middle Schoolers

Abstract

More young people are interacting with smart conversational agents such as Alexa and Google Assistant. These platforms are extensible, providing, in principle, a compelling opportunity for young users to create and tinker with their own conversational agents. However, to date the interfaces for conversational app development are adult-focused. This paper presents the early design process for AMBY (AI Made by You), which we are building to empower young learners to create their own conversational agents. We first conducted a contextual inquiry with 14 middle school students (aged 11-13) in an AI summer camp, followed by two other usability studies. The system design has been refined after each study. Key features of AMBY include a visual dialogue management panel, testing panel with a diverse avatar, and a voice input modality. AMBY is designed to serve as a pedagogically-robust resource for K-12 AI education and as an engaging and creative way for middle schoolers to explore AI.

Download Complete PDF

Authors

Amit Kumar
University of Florida
kumar.amit@ufl.edu

Xiaoyi Tian
University of Florida
tianx@ufl.edu

Mehmet Celepkolu
University of Florida
mckolu@ufl.edu

Maya Israel
University of Florida
misrael@coe.ufl.edu

Kristy Elizabeth Boyer
University of Florida
keboyer@ufl.edu

Trends and Priorities of Educational Technology Research: A Delphi Study

Aug 19, 2022

Return to Spotlight Research

Trends and Priorities of Educational Technology Research: A Delphi Study

Abstract

As journal editors play an important role in defining and shaping academic discourse, knowing their opinions could likely prove beneficial for both current and future academic journal stakeholders. Within this vein, this study used the Delphi method to help build a profile on the trends and priorities within educational technology, from the unique perspective of the journals’ editors-in-chief. This expert panel—initially built from 117 national and international research journals—concluded with 25 editors-in-chief who finished all three rounds of the survey. Results indicated five emerging themes for trends and priorities: computer-focused, teaching and learning, online and digital education, societal, and research and theory. By exploring these current trends and priorities within educational technology, this study may provide meaningful insights to better understand the field as a whole and may also help scholars in their goal of publishing relevant, high-quality academic scholarship.

Download Complete PDF

Authors

Xiaoman Wang
University of Florida

John Hampton
University of West Georgia

Albert D. Ritzhaupt
University of Florida

Kara Dawson
University of Florida

A Preliminary Data-driven Analysis of Common Errors Encountered by Novice SPARC Programmers

Aug 19, 2022

Return to Spotlight Research

A Preliminary Data-driven Analysis of Common Errors Encountered by Novice SPARC Programmers

Abstract

Answer Set Programming (ASP), a modern development of Logic Programming, enables a natural integration of Computing with STEM subjects. This integration addresses a widely acknowledged challenge in K-12 education, and early empirical results on ASP-based integration are promising. Although ASP is considered a simple language when compared with imperative programming languages, programming errors can still be a significant barrier for students. This is particularly true for K-12 students who are novice users of ASP. Categorizing errors and measuring their difficulty has yielded insights into imperative languages like Java. However, little is known about the types and difficulty of errors encountered by K-12 students using ASP. To address this, we collected high school student programs submitted during a 4-session seminar teaching an ASP language known as SPARC. From error messages in this dataset, we identify a collection of error classes, and measure how frequently each class occurs and how difficult it is to resolve.

Download Complete PDF

Authors

Zach Hansen
University of Nebraska Omaha
zachhansen@unomaha.edu

Hanxiang Du
University of Florida
h.du@ufl.edu

Wanli Xing
University of Florida
wanli.xing@coe.ufl.edu

Rory Eckel
Texas Tech University
rory.eckel@ttu.edu

Justin Lugo
MRC LLC
jlug331221@gmail.com

Yuanlin Zhang
Texas Tech University
y.zhang@ttu.edu

Modeling One-on-one Online Tutoring Discourse using an Accountable Talk Framework

Aug 19, 2022

Return to Spotlight Research

Modeling One-on-one Online Tutoring Discourse using an Accountable Talk Framework

Abstract

The National Council of Teachers of Mathematics (NCTM) has been emphasizing the importance of teachers’ pedagogical communication as part of mathematical teaching and learning for decades. Specifically, NCTM has provided guidance on how teachers can foster mathematical communication that positively impacts student learning. A teacher may have different academic goals towards what needs to be achieved in a classroom, which require a variety of discourse-based tools that allow students to engage fully in mathematical thinking and reasoning. Accountable or academically productive talk is one such approach for classroom discourse that may ensure that the discussions are coherent, purposeful and productive. This paper discusses the use of a transformer model for classifying classroom talk moves based on the accountable talk framework. We investigate the extent to which the classroom Accountable Talk framework can be successfully applied to one-onone online mathematics tutoring environments. We further propose a framework adapted from Accountable Talk, but more specifically aligned to one-on-one online tutoring. The model performance for the proposed framework is evaluated and compared with a small sample of expert coding. The results obtained from the proposed framework for one-on-one tutoring are promising and improve classification performance of the talk moves for our dataset.

Download Complete PDF

Authors

Renu Balyan
SUNY College at Old Westbury
balyanr@oldwestbury.edu

Tracy Arner
Arizona State University
tarner@asu.edu

Karen Taylor
Arizona State University
karnetaylor.sb@gmail.com

Jinnie Shin
University of Florida
jinnie.shin@coe.ufl.edu

Michelle Banawan
Arizona State University
mbanawan@asu.edu

Walter L. Leite
University of Florida
walter.leite@coe.ufl.edu

Danielle S. McNamara
Arizona State University
dsmcnamara1@gmail.com