================================================================================ Replication Files ================================================================================ Diffené, L., Leopold, T., Raab, M. & Buyukkececi, Z. (2026): Click, Collect, Compare: Evaluating a Nonprobability Web Survey for Family Demography Demographic Research 54(42): 1375-1412 -------- OVERVIEW -------- This replication package contains all code required to reproduce the comparison between the KINMATRIX survey (2022-24) and multiple external reference data sources across four dimensions: sociodemographic characteristics, family network size, family relationships, and family complexity. NOTE: Several data sources need to be downloaded first before running the code (see below). ================================================================================ SYSTEM REQUIREMENTS ================================================================================ Stata - Stata 18 (or later) R - R 4.4 or later - Main analysis (00b_set-up_master.R): tested with R 4.5.3 - DemoKin estimation (06_Rscripts/00_create_demokin_estimates.R): tested with R 4.5.2 using Positron (2026.03.0) - Required packages are installed automatically when running 00b_set-up_master.R - For exact package versions used in the main analysis, see session_info.txt ================================================================================ FOLDER STRUCTURE ================================================================================ 01_data/ Original input data (not modified by any script) 02_dofiles/ Stata data preparation scripts 03_posted/ Processed data output by Stata and R, used as input for analysis 04_temp/ Temporary data files 05_tables/ Excel table outputs 06_Rscripts/ R analysis and visualization scripts 07_Rgraphs/ SVG figure outputs ================================================================================ DATA ================================================================================ The replication package does NOT include the original input data, with the exception of the register data (see below). All data must be downloaded from the respective sources and saved in the subfolders of 01_data/ as described below. 1. KINMATRIX 2022-24 -------------------- File: ZA8825_v1-0-0.dta Save to: 01_data/kinmatrix 2022-24/ Source: GESIS Data Archive, Study No. ZA8825 Access: https://doi.org/10.4232/1.14380 NOTE: Free registration at https://www.gesis.org is required to download. 2. European Social Survey (ESS) -------------------------------- Three rounds are required. All files are available from the ESS Data Portal (https://ess.sikt.no/en/). Due to the COVID-19 pandemic, ESS Round 10 data were collected via face-to-face interviews in 22 countries (integrated file, Edition 3.3) and via self-completion questionnaires in 9 countries (self-completion integrated file, Edition 3.2). For this study, we use Edition 3.3 data for Finland, the Netherlands, Norway, Great Britain, and Italy, and supplement these with self-completion data from Germany, Poland, and Sweden from Edition 3.2. File Round Save to --------------------------------------------------------------------------------- ESS9e03_1.dta Round 9 (2018), Edition 3.1 01_data/ess 2018/ ESS10e03_3.dta Round 10 (2020), Edition 3.3 01_data/ess 2020/ ESS10SCe03_2.dta Round 10 (2020), Edition 3.2 01_data/ess 2020/ ESS11.dta Round 11 (2023-24), Edition 1.0 01_data/ess 2023-24/ NOTE: Free registration at the ESS Data Portal is required to download. 3. General Social Survey (GSS) -------------------------------- Files: GSS2022.dta, gss7222_r3.dta Save to: 01_data/gss 2021-22/ Source: NORC at the University of Chicago Access: https://gss.norc.org/get-the-data NOTE: Freely available, no registration required. 4. Generations and Gender Survey (GGS) ---------------------------------------- Wave 1 data for seven countries are required. All files are available from the GGS Data Portal (https://www.ggp-i.org/data). File Country Save to --------------------------------------------------------------- GGSII_Wave1_DE_V_1_0.dta Germany 01_data/ggs/ GGSII_Wave1_DK_V_1_1.dta Denmark 01_data/ggs/ GGSII_Wave1_FI_V_1_1.dta Finland 01_data/ggs/ GGSII_Wave1_NL_V_1_0.dta Netherlands 01_data/ggs/ GGSII_Wave1_NO_V_2_0.dta Norway 01_data/ggs/ GGSII_Wave1_SW_V_1_2.dta Sweden 01_data/ggs/ GGSII_Wave1_UK_V_1_1.dta United Kingdom 01_data/ggs/ NOTE: Registration and a data access application are required. Apply via the GGS Data Portal. Processing may take several days. 5. International Social Survey Programme (ISSP) ------------------------------------------------- File: ZA6980_v2-0-0.dta Save to: 01_data/issp 2017/ Source: GESIS Data Archive, Study No. ZA6980 (Social Networks and Social Resources 2017) Access: https://doi.org/10.4232/1.13322 NOTE: Free registration at https://www.gesis.org is required to download. 6. Pairfam ----------- File: anchor14.dta Save to: 01_data/pairfam 2021-22/ Source: GESIS Data Archive (Pairfam Wave 14, 2021-22) Access: https://doi.org/10.4232/pairfam.5678.14.0.0 NOTE: Free registration at https://www.gesis.org is required to download. 7. UN World Population Prospects 2024 (for DemoKin estimates) -------------------------------------------------------------- Four compressed CSV files are required. Download from the UN Population Division and save to 01_data/un_wpp/. WPP2024_Fertility_by_Age1.csv.gz Female age-specific fertility rates https://population.un.org/wpp/assets/Excel%20Files/1_Indicator%20(Standard)/CSV_FILES/WPP2024_Fertility_by_Age1.csv.gz WPP2024_Life_Table_Complete_Medium_Female_1950-2023.csv.gz Sex-specific life tables (female) https://population.un.org/wpp/assets/Excel%20Files/1_Indicator%20(Standard)/CSV_FILES/WPP2024_Life_Table_Complete_Medium_Female_1950-2023.csv.gz WPP2024_Life_Table_Complete_Medium_Male_1950-2023.csv.gz Sex-specific life tables (male) https://population.un.org/wpp/assets/Excel%20Files/1_Indicator%20(Standard)/CSV_FILES/WPP2024_Life_Table_Complete_Medium_Male_1950-2023.csv.gz WPP2024_Population1JanuaryBySingleAgeSex_Medium_1950-2023.csv.gz Population size by single age and sex https://population.un.org/wpp/assets/Excel%20Files/1_Indicator%20(Standard)/CSV_FILES/WPP2024_Population1JanuaryBySingleAgeSex_Medium_1950-2023.csv.gz NOTE: Freely available, no registration required. License: Creative Commons CC BY 3.0 IGO. Copyright: 2024 United Nations, DESA, Population Division. Licensed under Creative Commons license CC BY 3.0 IGO. Source: United Nations, DESA, Population Division. World Population Prospects 2024. Reference: United Nations (2024). World Population Prospects 2024: Summary of Results. (UN DESA/POP/2024/TR/NO. 9). United Nations. https://population.un.org/wpp/ 8. Register Data (included) ----------------------------- The following files are already included in the replication package and do NOT need to be downloaded: 01_data/register/De Bel et al. (2024). Online Supplementary Material D.xlsx 01_data/register/Kolk et al. (2023). Online Appendix 2.xlsx ================================================================================ EXECUTION ================================================================================ Once all data files have been downloaded and placed in the correct subfolders, run the two master scripts in the following order. Step 1 -- Data preparation (Stata) ------------------------------------ Open 00a_set-up_master.do in Stata 18 and run the entire script. The script automatically sets the working directory and sequentially runs all data preparation do-files. Processed data are saved to 03_posted/. Step 2 -- Analysis and figures (R) ------------------------------------ Open Replication.Rproj in RStudio and run 00b_set-up_master.R. The script will: 1. Install any missing R packages automatically 2. Generate DemoKin kinship estimates from the UN WPP 2024 data 3. Load all processed data from 03_posted/ 4. Produce all figures and save them as SVG to 07_Rgraphs/ 5. Write session_info.txt to the project root A random seed of 123454321 is set in the Stata master script. The R bootstrap script (06_Rscripts/02_network size_sMAPE_boot.R) uses a seed set within that script. ================================================================================ CODE FILES ================================================================================ Master scripts (project root) 00a_set-up_master.do Stata master script: Sets working directory and global path macros, sets random seed, and runs all data preparation do-files in order. 00b_set-up_master.R R master script: Installs missing packages, runs 00_create_demokin_estimates.R, loads all processed data, sources all analysis scripts, and saves all figures to 07_Rgraphs/. Stata do-files (02_dofiles/) 01_sociodemo_ess-gss_dataprep.do Prepares sociodemographic characteristics (gender, partnership status, parenthood, immigration background, employment, paternal education) from KINMATRIX, ESS, GSS, and GGS. Produces 01_*_merged.dta files. 02_network_register_dataprep.do Prepares family network size data (mean number of kin alive) from KINMATRIX and Nordic/Dutch register data. Produces 02_kinnum_kinma.dta and 02_kinnum_register.dta. 03a_relation_ess_dataprep.do Prepares family relationship indicators (contact frequency, emotional closeness, geographical distance, coresidence with parents) from KINMATRIX and ESS. Produces 03a_*_merged.dta files. 03b_relation_issp-pairfam_dataprep.do Prepares contact frequency with parents and siblings from KINMATRIX, ISSP, and Pairfam. Produces 03b_*_merged.dta files. 03c_relation_pairfam_sexdiff.do Prepares dyad-specific relationship indicators from KINMATRIX and Pairfam. Produces 03c_* files. 03d_relation_pairfam_sexdiff_aggr.do Aggregates dyad-specific relationship indicators across parent-anchor sex combinations. Produces 03d_*_merged_sexdiff_uw.dta files. 04_complex_ggs_dataprep.do Prepares family complexity indicators (parental separation, number of half-siblings) from KINMATRIX and GGS. Produces 04_*_merged.dta files. 05_samples.do Documents sample sizes for all indicators and data sources. Produces 05_tables/Table1.xlsx and 05_tables/S2. Sample sizes.xlsx. R scripts (06_Rscripts/) 00_create_demokin_estimates.R Computes kinship counts for 10 countries using the DemoKin package and UN WPP 2024 demographic data. Produces 03_posted/Demokin_counts_4yrs_shifted_un.csv. 01_sociodemo.R Produces Figure 1 (sociodemographic characteristics, main sample) and Figure A1 (including small reference samples). 02_network size.R Produces Figure 2 (mean number of kin alive by kinship category). 02_network size_scatter.R Produces Figure 3 (cross-national scatter plots of family network size). 02_network size_sMAPE_boot.R Computes sMAPE with bootstrapped confidence intervals. Produces Figure A2. 03_relation.R Produces Figure 4 (family relationships, main sample) and Figure A3 (including small reference samples). 03_relation_DE.R Produces Figure 5 (gender-specific adult child-parent relationships in Germany based on Pairfam data). 04_complex.R Produces Figure 6 (family complexity). ================================================================================ OUTPUT FILES ================================================================================ File Description ----------------------------------------------------------------------- 05_tables/Table1.xlsx Table 1: Numbers of egos and alters in KINMATRIX 05_tables/S2. Sample sizes.xlsx Table S2: Sample sizes 07_Rgraphs/Figure1_sociodemo.svg Figure 1: Sociodemographic characteristics 07_Rgraphs/Figure2_network.svg Figure 2: Family network size 07_Rgraphs/Figure3_network_scatter.svg Figure 3: Cross-national differences in family network size 07_Rgraphs/Figure4_relation.svg Figure 4: Family relationships 07_Rgraphs/Figure5_relation_DE.svg Figure 5: Gender-specific adult child-parent relationships in Germany 07_Rgraphs/Figure6_complex.svg Figure 6: Family complexity 07_Rgraphs/FigureA1_sociodemo.svg Figure A1: Sociodemographic characteristics, including small benchmark samples of n<200 07_Rgraphs/FigureA2_network_sMAPE_boot.svg Figure A2: sMAPE in % of the mean number of kin alive using DemoKin (2024) as reference 07_Rgraphs/FigureA3_relation.svg Figure A3: Family relationships, including small benchmark samples of n<200 ================================================================================ The authors May 2026 ================================================================================