README
======

Dataset title:
Pre-registration: Passenger Density on Stockholm Subway Platforms With and Without Blue-Light Exposure: A Quasi-experimental Observational Study

Recommended Citation:
Berglund, D., Hadlaczky, G., & Alvarsson-Hjort, J. (2026). Pre-registration: Passenger Density on Stockholm Subway Platforms With and Without Blue-Light Exposure: A Quasi-experimental Observational Study (Version 1) [Data set]. Karolinska Institutet. https://doi.org/10.48723/028g-3y46 

See the Swedish National Data Service catalogue entry, https://researchdata.se/en, for this dataset for complete metadata, contact information, and information about how to obtain the dataset.

Responsible organization:
Karolinska Institutet

Description:
This dataset is connected to a  quasi-experimental observational field study examining passenger presence and density on subway platforms in the Stockholm metro system under conditions with and without exposure to blue LED lighting (BLED). The study uses naturally occurring variation across platforms.

Data were collected at four subway platforms across three stations in Stockholm County between 2025-09-15 and 2026-04-26. Two platforms were equipped with blue LED armatures and two platforms served as non-exposed control platforms with standard lighting. On exposed platforms, lighting conditions alternated between BLED switched on and switched off, primarily in two-week periods with one four-week period, allowing both within-platform and between-platform comparisons over time.

The dataset consists of three linked sources, 1) aggregated camera-based detections of human presence on platforms, 2) aggregated station passage counts derived from ticket gate systems, and 3) environmental measurements of solar irradiance from an outdoor reference station. Camera detections are provided as minute-by-minute counts within predefined spatial zones on each platform. The dataset does not include images, video recordings, or any individual-level tracking data.

For the purpose of preregistration and documentation, the files provided at this stage contain simulated data only. These simulated files mirror the structure, variables, formats, and temporal alignment of the real data that will be analysed once data collection is complete.

License:
The dataset is shared under a Creative Commons Attribution 4.0 (CC BY 4.0) licence, allowing reuse and adaptation with appropriate attribution. Access to the non-simulated dataset is restricted.

Access restrictions:
The data collection is ongoing, and the real dataset is therefore not available for distribution. The files included in this release contain simulated data and are intended to illustrate the expected structure of the final dataset. Access to real data will be possible after publication of analyses and subject to ethical and legal approval.

Essential information for dataset reuse:
The data files are provided as structured tabular files that can be read using standard statistical software. No proprietary software is required. The original analyses are planned to be conducted in R, although reuse is not restricted to any specific software environment. Analysis scripts and documentation are expected to be made available after publication of primary results.

Dataset organization:
Not applicable.

Datafile descriptions:
1) BLISP_Passenger density_Hypotheses_260428.pdf: Document outlining the preregistered hypotheses and planned analyses for the study.

2) Platform density events - SIMULATED.csv: Simulated minute-level counts of detected human presence within predefined spatial zones on each subway platform, indexed by time, station, platform, zone, and lighting condition.

3) Solar irradiance at Bredäng station 250901-260427.csv: Simulated time stamped measurements of solar irradiance from an outdoor reference station, intended to represent ambient natural light conditions.

4) Station passages - SIMULATED.csv: Simulated aggregated counts of entries and exits at ticket gates for each station at minute resolution, used as indicators of overall station usage.

Relevant references:
Matsubayashi, T., Sawada, Y., & Ueda, M. (2013). Does the installation of blue lights on train platforms prevent suicide? A before-and-after observational study from Japan. Journal of Affective Disorders, 147(1–3), 385–388. 	https://doi.org/10.1016/j.jad.2012.08.018

Matsubayashi, T., Sawada, Y., & Ueda, M. (2014). Does the installation of blue lights on train platforms shift suicide to another station?: Evidence from Japan. Journal of Affective Disorders, 169, 57–60. 	https://doi.org/10.1016/j.jad.2014.07.036

Ichikawa, M., Inada, H., & Kumeji, M. (2014). Reconsidering the effects of blue-light installation for prevention of railway suicides. Journal of Affective Disorders, 	152–154, 183–185. https://doi.org/10.1016/j.jad.2013.09.006

Abbreviations:
Not applicable.

Further description of some variable names and abbreviations used in the data files:
Platform density events - SIMULATED.csv
	Timestamp: yyyy-mm-dd hh:mm timestamp of one person in zone detection
	Period: control_period = B-LED OFF / exposure_period = B-LED ON
	Platform: ZID exposure = Zinkensdamm subway platform 1 with B-LED armatures
		  ZID control  = Zinkensdamm subway platform 2 with normal lighting
		  BRÄ exposure = Bredäng subway platform 1 and 2 with B-LED armatures
		  ÖBG control  = Örnsberg subway platform 1 and 2 with normal lighting
	Zone: bled_zone the area closest to the end of the platform 
	      control_zone the adjacent area to the bled_zone towards the middle of the platform

Solar irradiance at Bredäng station 250901-260427.csv
The data file contains meta data to recreate the dataset from the source from rows 1-6. The data headings starts at row 8. All irradiance variables are provided at hourly temporal resolution and represent estimated solar radiation levels at the outdoor reference location in Bredäng, Stockholm County. Zero values typically indicate night-time hours or periods with no measurable solar irradiance. The value 999.0 is placeholder value indicating missing, invalid, or unavailable data. 
	Timestamp: yyyy-mm-dd hh:mm timestamp of the hourly irradiance measurement
	CIE UV irradiance: ultraviolet irradiance weighted according to the CIE action spectrum, expressed in milliwatts per square meter (mW/m²)
	Global irradiance: global horizontal solar irradiance, representing the total incoming solar radiation on a horizontal surface (direct plus diffuse), expressed in watts per square meter (W/m²)
        Direct normal irradiance: solar irradiance received per unit area by a surface normal to the direction of the sun’s rays, expressed in watts per square meter (W/m²)
        PAR: photosynthetically active radiation, representing solar radiation in the 400–700 nm wavelength range, expressed in watts per square meter (W/m²)
    	Direct horizontal irradiance: direct component of solar irradiance projected onto a horizontal surface, expressed in watts per square meter (W/m²)
	Diffuse irradiance: diffuse (scattered) component of solar irradiance on a horizontal surface, expressed in watts per square meter (W/m²)

Data sources:
The main dataset is not aggregated from existing datasets. All data within have been generated within the context of the present study. The simulated data provided here are designed solely to document data structure and planned analyses.

The passages into the stations is part of the normal logistics of AB Stockholm Lokaltrafik and have been extracted for the three stations during the collection period

The solar irradiance data is part of an open source data set from the Swedish Meteorological and Hydrological Institute based on the STRÅNG model (Radiation using the ångström model), https://www.smhi.se/en/research/research-units/meteorology/method-and-models/solar-radiation-research. Data access is provided by the link: https://strang.smhi.se/extraction/index.php 

Version control:
This dataset does not contain files under version control.

Source code information:
Statistical analyses will be conducted in R. Specific package versions and scripts will be documented and released alongside the final dataset.

Data Quality:
The data are observational and rely on automated camera-based detection systems and administrative ticket gate counts. Potential sources of measurement error include occlusion, lighting variation, and temporal aggregation. These limitations are expected to be relatively stable within platforms over time but should be considered when interpreting results. Data collection is ongoing, and further quality checks may be applied before release of the final dataset.
