Sketch With the Robot - data

This repository holds the data collected during the experimental sessions for the project "Sketch it for the robot! How child-like robots' joint attention affects humans' drawing strategies" accepted at ICDL (2024) Conference. 

The repository contains the drawings of the categories, the raw data file, the quantitative data file and elaborated data for the analysis.

The folder /Experimental_data  contains:

The folder /no_robot contains all the the drawings (.png format), produced in the individual condition, organized in subfolders. Each subfolder corresponds to a participant and the presence of the 'i' in the name means the partcipant was Italian , while the presence of the 's' means that the participant was Slovakian.

The folder /robot contains all the the drawings (.png format), produced in the robot condition, organized in subfolders. As the previous case, the letter 'i' and 's' stands for the nationality of the participant.

The file all_drawings.ndjson contains all the raw data (all the coordinates and timestamp of each drawing), where we also included the features extraction data. Thanks to the raw data (triplets (x, y, t)) it is possible to extract all the features needed.

The file quantitative_data.ndjson contains all the quantitative rankings data collected during the experiment (category, condition, latency time, total time, number of strokes, enjoyment_rank, difficulty_rank, likeability_rank).

The folder /Analysis_data contains the datasets used for the analysis. They are basically subsets specifically generated for the different analysis, containing all the features extraction and questionnaires data:

The file all_drawings_social_influence_all.csv is the 'mother' file containing all the relevant data.

The file all_drawings_social_influence_no_repetitions.csv contains the data relative to the categories that were drawn just 1 time. The file has been generated to study the effects of the social influence due to the robot's presence.

The file all_drawings_social_influence_only_repetitions.csv contains data relative just to the categories that were drawn more than 1 time. This file has been generated to compare results between categories repeated 2 and 3 times, to highlight the effect introduced by repeating a category for the third time.

The file all_drawings_repetition_influence.csv contains data relative to just the categories repeated 3 times, to study the repetition effect.

It is possible to finde the code used to collect the data at the DOI: 10.5281/zenodo.10944480