Teachers' readiness for integrating artificial intelligence into K-12 schools
Description
The study examines variables to assess teachers' preparedness for integrating AI into South African schools. The dataset on the Excel sheet consists of 42 columns. The first ten columns comprise demographic variables such as Gender, Years of Teaching Experience (TE), Age Group, Specialisation (SPE), School Type (ST), School Location (SL), School Description (SD), Level of Technology Usage for Teaching and Learning (LTUTL), Undergone Training/Workshop/Seminar on AI Integration into Teaching and Learning Before (TRAIN), and if Yes, Have You Used Any AI Tools to Teach Before (TEACHAI). Columns 11 to 42 contain constructs measuring teachers' preparedness for integrating AI into the school system. These variables are measured on a scale of 1 = strongly disagree to 6 = strongly agree. AI Ethics (AE): This variable captures teachers' perspectives on incorporating discussions about AI ethics into the curriculum. Attitude Towards Using AI (AT): This variable reflects teachers' beliefs about the benefits of using AI in their teaching practices. It includes their expectations of having a positive experience with AI, improving their teaching experience, and enhancing their participation in critical discussions through AI applications. Technology Integration (TI): This variable measures teachers' comfort in integrating AI tools and technologies into lesson plans. It also assesses their belief that AI enhances the learning experience for students, their proactive efforts to learn about new AI tools, and the importance they place on technology integration for effective AI education. Social Influence (SI): This variable examines the impact of colleagues, administrative support, peer discussions, and parental expectations on teachers' preparedness to incorporate AI into their teaching practices. Technological Pedagogical Content Knowledge (TPACK): This variable assesses teachers' ability to use technology to facilitate AI learning. It includes their capability to select appropriate technology for teaching specific AI content, and bring real-life examples into lessons. AI Professional Development (AIPD): This variable evaluates the impact of professional development training on teachers' ability to teach AI effectively. It includes the adequacy of these programs, teachers' proactive pursuit of further professional development opportunities, and schools' provision of such opportunities. AI Teaching Preparedness (AITP): This variable measures teachers' feelings of preparedness to teach AI. It includes their belief that their teaching methods are engaging, their confidence in adapting AI content for different student needs, and their proactive efforts to improve their teaching skills for AI education. Perceived Self-Efficacy to Teaching AI (PSE): This variable captures teachers' confidence in their ability to teach AI concepts, address challenges in teaching AI, and create innovative AI-related teaching materials.
Files
Steps to reproduce
This study gathered data from teachers employed in government and privately owned schools in South Africa. Ethical principles outlined in the Declaration of Helsinki were strictly followed to protect respondents' rights. Teachers were informed about the study's purpose, requirements, potential risks, and benefits. Participation was voluntary, and informed consent was obtained from each respondent. Measures were implemented to ensure privacy, and the study underwent a thorough review by the ethics committee of the researcher's institution. To encourage widespread participation, the survey link was shared through various channels, including teachers' association platforms and commonly used WhatsApp groups for professional collaboration. The online questionnaire, designed to take approximately 10 to 15 minutes to complete, was available for two and a half months, from October 14th to December 31st, 2023, providing ample time for teachers to participate. We used stratified random sampling to ensure representation across various specializations, including sciences, social sciences, mathematics, and languages. Initially, 581 responses were collected through the online survey. However, 151 responses from in-service teachers who reported no prior participation in AI training, workshops, or seminars were excluded, resulting in a final sample size of 430 respondents. The data collection instrument was a structured questionnaire based on the extended Unified Theory of Acceptance and Use of Technology (UTAUT2) framework. The questionnaire included eight key variables: AI ethics, attitudes toward using AI, technology integration, social influence, Technological Pedagogical Content Knowledge (TPACK), AI professional development, AI teaching preparedness, and perceived self-efficacy in teaching AI. These items were adapted from validated sources to ensure reliability and validity. The data was analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM) implemented through SmartPLS software version 4.1.0.1 (Ringle et al., 2024). PLS-SEM is a multivariate analysis technique that estimates cause-effect relationship models using established theories. It is particularly suitable for handling complex models that require relationship estimation and prediction without strict data requirements. The adequacy of the sample size was established using G*Power analysis version 3.1.9.7, a tool widely used in behavioral and social science research (Faul et al., 2007). The key parameters considered were the number of predictor variables (seven constructs), the expected effect size (0.15 for a medium effect), the anticipated probability (0.05 significance level), and the statistical power level (0.80). Based on the structural complexity of the model, the software determined the minimum sample size required to detect the specified effect. Our study's sample size 430 exceeded the recommended minimum of 103, confirming its sufficiency.