**READ ME**
**Title:** Coastal proximity and visits are associated with better health but may not buffer health inequalities
**Description:** In this project, we analyses secondary data from the Seas, Oceans, and Public Health survey in Europe and Australia (SOPHIE/A; *n* = 11,916-14,702) to investigate: (a) relationships of home coastal proximity and recreational coastal visits with self-reported general health; (b) the potential of both to buffer income-related health inequalities; and (c) the generalizability of these propositions across 15 countries, using a Bayesian approach to quantify the relative support for or against any relationships.
**Authors:** blinded for review
**Correspondence:** For any questions, please contact blinded for review (blinded for review).
**Date:** 2022-10-06
**Digital Object Identifier (DOI):** not yet available
----------
**Overview of files in the repository:**
**1. Preregistration** (`0_preregistration`)
- `0_Preregistration.pdf`: contains the preregistration (13.12.2021) for this project (Hypothesis 1 and 2, Research Question 1) and a related project. The full preregistration is under "Registrations".
- `0_Preregistration_Addendum1.pdf`: contains the addendum to the preregistration (07.03.2022) for this project (Hypothesis 3 and 4, Research Question 4).
- `Additions_PriorKnowledge.pdf`: contains details about the prior knowledge of those authors who had access to the data prior to the preregistration.
----------
**2. Data and Materials** (`1_data` and `2_materials`)
The data, codebook, and materials of the SOPHIE (Europe) survey are available via the UK Data Service (https://beta.ukdataservice.ac.uk/datacatalogue/studies/study?id=8972).
For computational reproducibility, the cleaned data of both the SOPHIE and SOPHIA (Australia) survey that were used for the current analyses are available on this OSF page (`data.rds`, `data_H1.rds`, `data_H1_Cz.rds`, `data_H2.rds`, `data_H3.rds`, `data_H3_Cz.rds`, `data_H4.rds`).
When reusing the data for different analyses, please exclusively reuse and cite the data on the UK Data Service.
----------
**3. Data Analysis** (`3_analysis`)
- `project.RProj`: contains the R project
- `analysis.Rmd`: contains the R code to prepare and analyze the data
- **H1** `bmod_H1_c_Cz.rds`: contains the model with home coastal proximity (categorical) and control variables as predictors, health as an outcome, and a random intercept for country, without Czechia
- **H1** `bmod_H1_c_d_Cz.rds`: contains the same model as `bmod_H1_c_Cz.rds` but with additional random slopes for home coastal proximity
- **H1** `bmod_H1_m_Cz.rds`: contains the model with home coastal proximity (monotonic) and control variables as predictors, health as an outcome, and a random intercept for country
- **H1** `bmod_H1_m_d_Cz.rds`: contains the same model as `bmod_H1_m_Cz.rds` but with an additional random slope for home coastal proximity
- **H2** `bmod_H2_c.rds`: contains the model with recreational visits (categorical) and control variables as predictors, health as an outcome, and a random intercept for country
- **H2** `bmod_H2_c_v.rds`: contains the same model as `bmod_H2_c.rds` but with additional random slopes for recreational visits
- **H2** `bmod_H2_m.rds`: contains the model with recreational visits (monotonic) and control variables as predictors, health as an outcome, and a random intercept for country
- **H2** `bmod_H2_m_v.rds`: contains the same model as `bmod_H2_m.rds` but with an additional random slope for recreational visits
- **H3** `bmod_H3_c_Cz.rds`: contains the model with home coastal proximity (monotonic), household income (categorical), their interaction, and control variables as predictors, health as an outcome, a random intercept for country, and a random slope for home coastal proximity, without Czechia
- **H3** `bmod_H3_c_i_Cz.rds`: contains the same model as `bmod_H3_c_Cz.rds` but with additional random slopes for household income
- **H3** `bmod_H3_c_i_i_Cz.rds`: contains the same model as `bmod_H3_c_i_Cz.rds` but with additional random effects for the interaction of home coastal proximity and household income
- **H3** `bmod_H3_m_Cz.rds`: contains the model with home coastal proximity (monotonic), household income (monotonic), their interaction, and control variables as predictors, health as an outcome, a random intercept for country, and a random slope for home coastal proximity, without Czechia
- **H3** `bmod_H3_m_i_Cz.rds`: contains the same model as `bmod_H3_m_Cz.rds` but with additional random slopes for household income
- **H3** `bmod_H3_m_i_i_Cz.rds`: contains the same model as `bmod_H3_m_i_Cz.rds` but with additional random effects for the interaction of home coastal proximity and household income
- **H3** `loo_bmod_H3_c_Cz.rds`, `loo_bmod_H3_c_i_Cz.rds`, `loo_bmod_H3_c_i_i_Cz.rds`, `loo_bmod_H3_m_Cz.rds`, `loo_bmod_H3_m_i_Cz.rds`, and `loo_bmod_H3_m_i_i_Cz.rds` contain the loo objects for the respective models
- **H4** `bmod_H4_c_Cz.rds`: contains the model with recreational visits (monotonic), household income (categorical), their interaction, and control variables as predictors, health as an outcome, a random intercept for country, and a random slope for recreational visits, without Czechia
- **H4** `bmod_H4_c_i_Cz.rds`: contains the same model as `bmod_H4_c_Cz.rds` but with additional random slopes for household income
- **H4** `bmod_H4_c_i_i_Cz.rds`: contains the same model as `bmod_H4_c_i_Cz.rds` but with additional random effects for the interaction of recreational visits and household income
- **H4** `bmod_H4_m_Cz.rds`: contains the model with recreational visits (monotonic), household income (monotonic), their interaction, and control variables as predictors, health as an outcome, a random intercept for country, and a random slope for recreational visits, without Czechia
- **H4** `bmod_H4_m_i_Cz.rds`: contains the same model as `bmod_H4_m_Cz.rds` but with additional random slopes for household income
- **H4** `bmod_H4_m_i_i_Cz.rds`: contains the same model as `bmod_H4_m_i_Cz.rds` but with additional random effects for the interaction of recreational visits and household income
- **H4** `loo_bmod_H4_c_Cz.rds`, `loo_bmod_H4_c_i_Cz.rds`, `loo_bmod_H4_c_i_i_Cz.rds`, `loo_bmod_H4_m_Cz.rds`, `loo_bmod_H4_m_i_Cz.rds`, and `loo_bmod_H4_m_i_i_Cz.rds` contain the loo objects for the respective models
----------
**4. Figures** (`4_figures`)
- `cond_eff_H1_H2_combined.svg` displays the conditional effects for Hypothesis 1 (Panel A) and 2 (Panel B) based on `bmod_H1_m_d_Cz.rds` and `bmod_H2_m_v.rds`
- `RQ1a_b.svg` displays the posterior distribution of the regression coefficient of home coastal proximity (Panel A) and recreational visits (Panel B) per country
- `cond_eff_H3_H4_combined.svg`, `cond_eff_H3_ord_Cz.svg`, and `cond_eff_H4_ord.svg` display the conditional effects for Hypothesis 3 (Panel A) and 4 (Panel B) based on `bmod_H3_m_i_Cz.rds`and `bmod_H4_m_i.rds`, where health is treated as either continuous (combined) or ordinal (ord)
----------
**5. Sensitivity Analysis** (`5_sensitivity`)
contains the models of the sensitivity analysis with
- a narrower prior N(0,5) (folder `prior`)
- when excluding speeders (i.e., respondents who completed the survey faster than 5 min; folder `speeders`)
- when including Czechia (only H1 and H3; folder `incl_Cz`)
- additional control variables (education, work status, and political orientation; folder `covariates`)
----------
**6. Session Info** (`6_sessioninfo`)
contains the session info (`sessionInfo.txt`) and the packages used including citations (`grateful-report.Rmd`)
----------
**Licence:**
The content of this folder is licensed under the CC BY 4.0 license (https://creativecommons.org/licenses/by/4.0/). You may share and adapt all content of this folder, as long as appropriate credit is given.