Results from this analysis are presented in the manuscript "Social contact patterns following the COVID-19 pandemic: a snapshot of post-pandemic behaviour from the CoMix study" available on MedrXiv.

The repository is organized in the following folders:

• data: Folder that contains the datasets.
• r: Folder with the main code.
• stan: Folder with Stan related functions.
• output: Folder where the output of the analysis (tables and figures of the manuscript) is saved.

The data analysed in the manuscript is available in the folder data/zenodo, together with the corresponding dictionary (CoMix_last_wave_dictionary.xlsx). The dataset is also available independently on Zenodo at the following link. The data is formatted according to the standard of the socialcontactdata.org initiative and is therefore interoperable with the socialmixr R package. The participant information is reported (one participant per line) in the files participants_common and participants_extra, while the contacts information are reported (one contact per line) in the contacts_common and contacts_extra files.

To perform the analysis:

• Update the "r/00_setup_filepaths.R" file to include the working path of the local machine.
• Run the "r/master.R" file. The code relies on specific implementa

The code generates as output the three tables of the main paper:

• Table 1 ("tab1_characteristics.docx"): Participants characteristics. The code generates a table of the participants features, listing both absolute numbers and percentages. To compute correctly the percentages the custom funcion get_perc is used.
• Table 2 ("tab2_contacts.docx"): Mean daily contacts per participant by country and setting, together with bootstrap confidence intervals.
• Table 3 ("tab3_contacts.docx"): Mean daily contacts per participant by characteristics, together with bootstrap confidence intervals.

And figure 1,2 and 3 of the main paper. In particular, the code to plot the contact matrices can be found in an_fig3_cms.R and could be adapted to explore the data.

We refer to the manuscript for the interpretation of the results.

• data.table
• lubridate
• flextable
• magrittr
• dplyr
• socialmixr
• ggplot2
• patchwork
• forcats
• gtsummary
• boot
• wpp2022
• viridis
• cowplot
• tidyverse
• stringr
• readxl
• reshape2