Helmut Farbmacher, Raphael Guber, Johan Vikstroem, "Increasing the Credibility of the Twin Instrument", Journal of Applied Econometrics, Vol. 33, No. 3, 2018, pp. 457-472. The zip file fgv-replication.zip contains two folders. The folder US_census_data contains the extract from the US 1980 census of married and unmarried mothers used in Angrist, Joshua D., and William N. Evans. "Children and Their Parents' Labor Supply: Evidence from Exogenous Variation in Family Size.", American Economic Review 88.3 (1998): 450-477. The file is compressed using 7zip. The do file AngristEvans1980_jae replicates all results related to these data. The data from Angrist and Evans (1998) were downloaded from Joshua Angrist's homepage: http://economics.mit.edu/faculty/angrist/data1/data/angev98 It is provided in SAS format. We have used SPSS to open the data (get sas data='C:\m_d_806.sas7bdat') and then stored it as a Stata dataset (AngristEvans1980.dta). The folder Swedish_administrative_data contains separate folders dedicated to tables and figures related to the Swedish administrative data. These contain do-files and notes explaining in which order to run them. The Swedish administrative data are confidential and not included. The data needed to replicate our results are in the following registers: Name Source Years SYS (population register) Statistics Sweden (SCB) 1987-1989 LOUISE (population register) Statistics Sweden (SCB) 1990-2007 Multi-generational register Statistics Sweden (SCB) 1932-2006 (cohorts born) National Hospital Patient Register National Board of Health and Welfare 1985-2007 The main data set is the population register data set LOUISE for the period 1987-2007. These data are collected by Statistics Sweden (SCB). We also use information from the National Causes of Death Register and the National Hospital Patient Register collected by The National Board of Health and Welfare (Socialstyrelsen), and the Multi-generational register collected by Statistics Sweden to create outcome variables. The parental identifier is matched to LOUISE and the Multi-generational register which then facilitates the matching with other administrative registers containing information on parental characteristics. To obtain data access for replication purposes, a procedure should be followed, and we are happy to emphasize that at all stages we are delighted to help out to facilitate the process. We now explain this procedure. IFAU (Institute for Labour Market Policy Evaluation) is a research institute of the Swedish Ministry of Employment, situated in Uppsala. Affiliates of IFAU have access to data files containing individual records from certain merged Swedish administrative population registers. Non-affiliates of IFAU can obtain access to the data files we used, in order to carry out replication analyses. This requires them to submit a request to the Director (see below) of IFAU. Subsequently, they must sign the relevant data protection and confidentiality statements. After that, the researchers can work with the data in a secure way on a local computer at IFAU (and in some cases through remote access to servers). This is a non-bureaucratic procedure. IFAU is delighted to help non-affiliated researchers with data access. Subsequently, in addition, we as authors are delighted to offer direct access to the constructed data files used in the econometric analysis for the article. Alternatively, non-affiliates may obtain access to the original data registers following a different procedure. First, permission should be obtained from an Ethical Review Board ("Regionala Etikprövningsnämnden". In our case, this was the regional board in Uppsala). Details are on http://www.epn.se/en/start/startpage/. Next, after having obtained an ERB approval to use the data, the relevant registers and/or variables from registers for the relevant years can be ordered from SCB and The National Board of Health and Welfare (Socialstyrelsen). In the data that are delivered, SCB assigns to each individual an identification number that is unrelated to the actual personal identification numbers used in Sweden. The former identification number enables the matching of variables across the various administrative registers. The data obtained from SCB do not contain personal information such as name and address. The relevant contact details are: IFAU Kyrkogårdsgatan 6 Box 513 751 20 Uppsala Tel: (+46) 18 - 471 70 70 Fax: (+46) 18 - 471 70 71 E-post: ifau@ifau.uu.se corresponding author for access to Swedish data: Johan Vikström E-mail: johan.vikstrom [at] ifau.uu.se Corresponding author of paper: Helmut Farbmacher E-mail: farbmacher [at] mea.mpisoc.mpg.de