*** Supplementary materials for *** "Estimating male fertility in eastern and western Germany since 1991: A new lowest low?" *** by Christian Dudel and Sebastian Klüsener *** Contact: dudel@demogr.mpg.de or kluesener@demogr.mpg.de *** Contents *** 1. General remarks 2. Data files: Overview 3. Code files *** 1. General remarks *** This file contains a description of the code and data files acopmanying the article "Estimating male fertility in eastern and western Germany since 1991: A new lowest low?". Note that the data files we can supply are intermediate results based on the German birth register. We are not allowed to distribute the raw data. Interested researchers can access the German birth register through the Research Data Centres of the Federal Statistical Office and the statistical offices of the Länder, see http://www.forschungsdatenzentrum.de/en/ Overall, we supply 206 data files in csv format and 23 files of code for the statistical software R, which is available free of charge at https://www.r-project.org/ Data are supplied in the same format we extracted it at the Research Data Centre. *** 2. Data files: Overview *** The names of most (but not all) data files are in the format 'XXXX_name.csv', where 'XXXX' stands for the year, e.g, 1991, and 'name' indicates the content of the file. The following files are supplied (for a more thorough description see readme_data.txt): - male_fertility_Germany.csv: Age specific fertility rates for males and females by region (Germany total, eastern Germany, western Germany) for the years 1991 to 2013. For males ages 17 to 59, for females 16 to 45. If you are only interested in our main results this file is probably the most interesting for you. - DE_REG_MASTER_PAR_TOP90_2012_2013.csv: Population counts for the years 2012 and 2013 by gender, age, and federal state. - DE_REG_MASTER_PAR_ERR_ACC_TOP89_corrected.csv: Adjusted population counts for the years 1991 to 2011 (in one file). - XXXX_births_wo.m.csv: Age-specific birth counts for males by region (Germany total, eastern Germany, western Germany) for 1991 and 1993 to 2013 (XXXX). Ages 16 to 59. Based on the first imputation variant described in the paper. - 1992_births_wo.m_agered.csv: Age-specific birth counts for males by region (Germany total, eastern Germany, western Germany) for 1992. Ages 17 to 59. Based on the first imputation variant described in the paper. This file starts with age 17 because of data privacy issues. - XXXX_births_dr.m.csv: Age-specific birth counts for males by region (Germany total, eastern Germany, western Germany) for 2000 to 2013 (XXXX). Ages 16 to 59. Based on the second imputation variant described in the paper, i.e. imputation based on non-marital births. - XXXX_births_reg.m.csv: Age-specific birth counts for males by region (Germany total, eastern Germany, western Germany) for 2000 to 2013 (XXXX). Ages 16 to 59. Based on the third imputation variant described in the paper, i.e. imputation via regression. - XXXX_births_match.m.csv: Age-specific birth counts for males by region (Germany total, eastern Germany, western Germany) for 2000 to 2013 (XXXX). Ages 16 to 59. Based on the fourth imputation variant described in the paper, i.e. imputation via matching. - XXXX_births.f.csv: Age-specific birth counts for females by region (Germany total, eastern Germany, western Germany) for 1991 and 1993 to 2013 (XXXX). Ages 16 to 45. - 1992_births.f_agered.csv: Age-specific birth counts for females by region (Germany total, eastern Germany, western Germany) for 1992. Ages 16 to 45. This file is similar to XXXX_births.f.csv, but for males the age range is reduced for this year (because of which 'agered' was added to the file name). - XXXX_missings.csv: Total number of births by age of mother (16 to 45) and region (Germany total, eastern Germany, western Germany) for 1991 and 1993 to 2013 (XXXX). Moreover, the number of births with missing information on the age of the father is included. - 1992_missings_agered.csv: Total number of births by age of mother (16 to 45) and region (Germany total, eastern Germany, western Germany) for 1992. Moreover, the number of births with missing information on the age of the father is included. This file is similar to XXXX_missings.csv, but for males the age range is reduced for this year (because of which 'agered' was added to the file name). - XXXX_missings.nm.csv: Total number of non-marital births by age of mother (16 to 45) and region (Germany total, eastern Germany, western Germany) for 1991 and 1993 to 2013 (XXXX). Moreover, the number of births with missing information on the age of the father is included. - 1992_missings.nm_agered.csv: Total number of non-marital births by age of mother (16 to 45) and region (Germany total, eastern Germany, western Germany) for 1992. Moreover, the number of births with missing information on the age of the father is included. This file is similar to XXXX_missings.nm.csv, but for males the age range is reduced for this year (because of which 'agered' was added to the file name). - XXXX_phi_wo_cleaned_2nd_step_g.csv: Birth counts by age of mother and age of father for Germany in total for 1991 to 2013. Age of mother from 16 to 45, age of father from 16 to 59, except for 1992, for which age of father is from age 17 to age 59. - XXXX_phi_wo.west_cleaned_2nd_step_g.csv: Birth counts by age of mother and age of father for western Germany for 1991 to 2013. Age of mother from 16 to 45, age of father from 16 to 59, except for 1992, for which age of father is from age 17 to age 59. - XXXX_phi_wo.ost_cleaned_2nd_step_g.csv: Birth counts by age of mother and age of father for eastern Germany for 1991 to 2013. Age of mother from 16 to 45, age of father from 16 to 59, except for 1992, for which age of father is from age 17 to age 59. *** 3. Code files *** We supply four R syntax files stored in "Code", which together with the data described above can be used to replicate most of our analyses. The file names begin with numbers, roughly ordering the code by the appearance of its results in the paper. After the number, a short descriptive name follows. All files include comments, describing what each step in the code is doing. - 1-missings.r: This file includes the analyses of the poportions of missing values we present in section 2 of the paper. It also includes the code for figures 1 and 2. - 2-fertility_rates_males.r: This file produces age-specific fertility rates for males based on adjusted population counts and the first imputation method described in the paper. Code for figures 3(a) and 4 can also be found in this file. - 3-sensitivity.r: This file contains the code of our sensitivity analyses and for figure 3(b). - 4-fertility_rates_females.r: This file is used to calculate fertility rates for females and includes the code for figure 3(c). We also make available 23 additional files (in "Code FDZ"), which can be used with data of the "Statistik der Geburten" to generate the data sets given above. All syntax files are named as statistik_der_geburten_XXXX.R, where "XXXX" indicates the year. All files include comments, describing what each step in the code is doing. Additionally, the folder "Code FDZ" includes the file "2003.csv". This file has a similar structure than the original data, but its variables are either randomly generated or constant. Still, it can be used together with statistik_der_geburten_2003.R and allows to better understand how the code works. ***END OF FILE***