Dataset for: Effects of Handle Diameter and Center of Gravity on Tremor Suppression

1. Research Hypothesis.
This study hypothesizes that modifying the physical morphology of a spoon—specifically, increasing the handle diameter and shifting the center of gravity (CG) proximally—will biomechanically suppress hand tremors during feeding tasks in older adults. Furthermore, we hypothesize that these objective biomechanical benefits may conflict with users' subjective preferences due to their long-term cognitive reliance on traditional slender utensils.

2. Data Gathering Methodology.
The data was collected from 20 older adults (aged 75 and above) who performed a simulated 8-second static feeding holding task. A 2 × 2 repeated-measures design was employed, with handle diameter (12 mm vs. 30 mm) and CG position (proximal vs. distal) as independent variables.
(a) Objective Kinematic Data: Captured using a wireless 6-axis Inertial Measurement Unit (IMU) mounted on the spoon handle, recording tri-axial acceleration and angular velocity at 200 Hz.
(b) Subjective Data: Collected via structured interviews immediately following the tasks to assess operational preference, perceived effort, and willingness to use, using a 3-point forced-choice scale.

3. Notable Findings & What the Data Shows.
(a) Objective Biomechanics: The kinematic data indicates that the 30 mm thick handle significantly improved linear stability (reduced acceleration CV, p < 0.05) and demonstrated greater rotational robustness during CG shifts compared to the 12 mm thin handle.
(b) Subjective Preference-Performance Paradox: Despite the objective physical advantages of the thick handle, the subjective data reveals that 65% of participants significantly preferred the thin handle (p < 0.01), driven by established mental models (e.g., chopstick usage habits). For CG, 80% preferred the proximal weighting (p < 0.001), aligning with mechanical intuition.

4. How to Interpret and Use the Data.
This dataset is structured to guarantee full reproducibility. It includes raw IMU time-series signals, processed/filtered data segments, and the final statistical feature dataset (containing Root Mean Square and Coefficient of Variation values, along with their Log10 transformations). A Jupyter Notebook (.ipynb) containing all Python processing scripts is also provided.Other researchers can use this dataset to:
(a) Replicate the repeated-measures ANOVA to verify the kinematic and statistical findings.
(b) Explore alternative signal processing and filtering algorithms for tremor quantification.
(c) Investigate the correlation and discrepancies between objective biomechanical metrics and subjective ergonomic preferences in age-friendly design.

For a detailed explanation of folder structures, condition codes (e.g., A1, B2), and variable definitions, please refer to the included README.txt file.