FossilSketch was developed to address the lack of resources in undergraduate education for micropaleontology - the study of microfossils used to understand past environments and aid in
oil and gas exploration. Traditionally, training in micropaleontology requires specialized resources, such as microscopes and fossil materials. In such training, personalized instructor
feedback may require significant time and thus only be available in small classes. As a result, most institutions offer very limited micropaleontology exposure to students. FossilSketch was
developed to overcome this hurdle by providing accessible and engaging activities for learning micropaleontology. Its features include educational videos, step-by-step exercises, and
interactive mini-games that provide real-time automatic feedback, enabling independent learning while offering instructors tools to manage and track progress. In this study, we introduced
FossilSketch to high school students during an outreach program, SeaCamp 2024. We collected data that included students' outcomes (scores) in different activities. Also, the data were collected
using a survey that reflect students' perceptions towards studying micropaleontology using FossilSketch activities. We calculated the sentiment value of students' perception using the
Valence Aware Dictionary and sEntiment Reasoner (VADER) from the survey data. The following research question was addressed in the paper: What aspects of students' learning
outcomes from FossilSketch's activities shaped their sentiments toward studying micropaleontology? To answer this research question, we analyzed the data using Spearman's
correlation and XGBoost's feature importance values. Correlational analysis showed that three aspects of FossilSketch's activities showed a significant relationship with students' sentiments
toward learning more about micropaleontology: 1) Morphotype Matching Mini-Game Attempt Count (p=0.001), 2) Ostracod Orientation Mini-Game Average Score (p=0.041), and 3) Ostracod
Outline Mini-Game Average Score (p=0.035). The Total Average Score per Attempt of an Activity had the highest value in XGBoost's list of feature importance values while negatively
correlated with students' sentiments. Our analysis showed that students participating in activities more frequently had more positive sentiments towards learning more about micropaleontology,
even though they did not have higher average scores per activity attempt. The results of this study are novel and provide information on the effectiveness of FossilSketch for outreach
activities from the perspectives of high school students.
The full paper will be available to logged in and registered conference attendees once the conference starts on June 22, 2025, and to all visitors after the conference ends on June 25, 2025