GENERAL INFORMATION
-------------------

1. Title of dataset

Semantic fluency responses in Spanish (L1) and English (L2) from Spanish primary-school children

2. Authors

Name: María Paz Suárez-Coalla
Institution: Language Psychology Laboratory, Department of Psychology,
University of Oviedo, Oviedo, Spain
Email: suarezpaz@uniovi.es
ORCID: https://orcid.org/0000-0001-9772-2680

Name: Uxue Pérez-Litago
Institution: Language Psychology Laboratory, Department of Psychology,
University of Oviedo, Oviedo, Spain
Email: perezuxue@uniovi.es
ORCID: https://orcid.org/0000-0002-8317-8159

Name: Lucía García Castro
Institution: University of Oviedo, Oviedo, Spain

Name: Cristina Esteban Saster
Institution: Department of Organization Engineering, University of Burgos,
Burgos, Spain
Email: ces1006@alu.ubu.es
ORCID: https://orcid.org/0009-0000-8750-2043

Name: José Manuel Galán
Institution: Department of Organization Engineering, University of Burgos,
Burgos, Spain
Email: jmgalan@ubu.es
ORCID: https://orcid.org/0000-0003-3360-7602

3. Dataset description

This dataset contains the primary semantic-fluency response sequences used in
the analyses of the related article. The participants were 192 Spanish-L1
primary-school pupils in Grades 4-6 learning English as a foreign language.
Each child was assigned two semantic categories in Spanish (L1) and two
different categories in English (L2) in a crossed design. The categories were
animals, fruits, body parts, and clothing. Participants had two minutes per
task to write as many category members as possible.

The dataset contains 724 participant-language-category tasks and 9,070 written
responses. Response order is retained within each task. Original participant
codes and the mapping between those codes and the neutral identifiers in this
dataset are not deposited.

4. Keywords

semantic fluency; child EFL learners; lexical-semantic access; bilingual
lexical processing; Spanish; English; serial position; primary education

5. Funding

Spanish Ministry of Science, Innovation and Universities, predoctoral grant
FPU21/02740.

6. Geographic location of data collection

An urban mainstream state-subsidised primary school in a city in northern
Spain. The school and city are not identified in the public dataset.


ACCESS INFORMATION
------------------

1. Licence

Creative Commons Attribution 4.0 International (CC BY 4.0):
https://creativecommons.org/licenses/by/4.0/

2. Dataset DOI

To be assigned by RIUBU.

3. Related publication

Submitted manuscript.


METHODOLOGICAL INFORMATION
--------------------------

1. Participants

The dataset comprises 192 pupils: Grade 4, n=57; Grade 5, n=72; and Grade 6,
n=63. Recorded sex was 94 boys, 97 girls, and one value not reported. All
participants were Spanish-L1 speakers receiving approximately three hours of
English-as-a-foreign-language instruction per week.

2. Task and design

Each child completed two categories in Spanish and two different categories in
English. Category-language assignment was crossed across pupils so that a
participant did not repeat the same category in both languages. Language order
was counterbalanced. Responses were transcribed in production order.

3. Ethics

Guardians provided written consent, and the study procedure was approved by
the Research Ethics Committee of the Principality of Asturias.

4. Data structure

semantic_fluency.csv is the canonical data file. It has one row per semantic-
fluency task and retains the complete ordered response sequence in the
response_sequence field. Individual responses are separated by commas inside
that field; standard CSV quoting protects those internal commas.

A separate token-level table is not included because it would contain no new
primary observations. It can be reconstructed from semantic_fluency.csv by
splitting response_sequence at commas while retaining the within-row order;
the resulting position is the serial position used in word-level analyses.

Neutral participant identifiers (P001-P192) replace the original participant
codes. No re-identification key is included. Missing recorded sex is represented
as not_reported. Variable definitions and allowed values are provided in
codebook.csv.

5. Software and formats

All files use UTF-8 encoding. The data and codebook are supplied as CSV files
and can be read with R, Python, LibreOffice, Microsoft Excel, or a text editor.
No specialist software is required.


FILE OVERVIEW
-------------

readme.txt
    Dataset documentation.

semantic_fluency.csv
    Primary data used in the article analyses: 724 rows and 8 columns, with one
    row per participant-language-category task.

codebook.csv
    Definitions, data types, allowed values, and units for the eight variables
    in semantic_fluency.csv.

checksums.txt
    SHA-256 checksums for verifying the integrity of the three files above.


VARIABLE SUMMARY
----------------

task_id
    Neutral unique task identifier.

participant_id
    Neutral participant identifier.

grade
    Primary-school grade (4, 5, or 6).

sex
    Recorded sex (boy, girl, or not_reported).

language
    Task language (Spanish or English).

category
    Semantic category (animals, fruits, body_parts, or clothing).

response_count
    Number of responses in the task sequence.

response_sequence
    Comma-delimited responses in production order.