I rarely own the copyright of published papers: it has been transferred to the publisher of the journal/book. I provide the files below for individual use only. Please obtain legal permission from the publisher for reproduction, if it is necessary. My recent papers are available as PDFs (Adobe Acrobat Portable Document Files), so they can be read on any platform and printed easily. (Since my papers are freely available on the web, I do not ordinarily mail reprints.) If you do not have the Adobe Acrobat reader installed, it can be obtained for free from the Adobe web page. Over 50 zipped data sets for my recent research are available below (“recent” is a relative term; I’ve been making my data and output accessible since at least 1990). These are usually available as STATA data sets, but sometimes as Excel worksheets, E-Views work files or in ASCII “dictionary” files (with variable descriptions in the headers). If you use my data sets, please acknowledge this in print where appropriate. More importantly, if you find an error in one of my data sets or programs, please let me know! There are also some sample programs and much sample output (typically covering the most important empirics; I try to comment so as to make output self-explanatory). Before complaining about a data set or requesting something, please read this. Graduate Students and others: I provide data sets and sample output as a professional courtesy so that others can replicate my findings (I try to adhere to King’s Replication Standard), not (primarily) for the convenience of others doing their own research. Please do not abuse my courtesy (e.g., by asking me for something that you can obtain yourself easily enough, complaining about formatting, or asking for access to a data set when I haven’t posted the paper yet). Lots of people think my data sets are easily accessible. I usually do not respond to questions that can be answered easily with information on my website. Please try to look before you ask! If you do not use STATA (I now use version 14), you can transfer between different formats with Stat/Transfer. Details are available at the Stat/Transfer web site. Sorry, I don’t provide their software! (And if you can’t handle ASCII [dictionary] files, I won’t be able to help). If you do not have a program to unzip files, you’ll need one. One (among many) choice is Winzip; you may obtain it from their web site. (And if you can’t handle zipped files, I won’t be able to help you). I provide my data sets in their final “massaged” format, not original source data, except in very special circumstances. If you’re interested in source data, you’ll be able to find it … at the sources (always discussed in my papers). I try not to use data sets that are proprietary or confidential. Please do not ask me to provide data sets that are easily accessible. I don’t typically provide codebooks, since they’re embedded: variable descriptions are provided in the STATA data sets and in STATA dictionary files. Variable descriptions and descriptive statistics are also typically provided in each output. I also often embed information into STATA data sets with “notes.” If you do have an issue with a data set (or whatever), please ensure I can understand it (e.g., mention which project and please be specific with e.g., URLs). I don’t promise to respond to frivolous or unclear e-mail about data sets. I provide output rather than raw program code. The code is embedded within output and clearly visible; I do this so that the skeptical can easily see actual results (useful for replication) as well as the code generating it. (It also makes it much easier for me to check estimates myself.) Feel free to use the embedded code if you like, but remember that it’s then yours. Please don’t ask me for code (or anything else) simply to make your own research easier; that’s not the reason I provide materials. All my recent research that is publicly available is … publicly available here. Please don’t ask me for something that isn’t here unless you’re either a friend or very important; I almost surely won’t be able to help you. Graduate students and others: output files have program code embedded within. Occasionally a co-author asks me not to post a data set; in such circumstances, please just ask the co-author. And … to repeat … please do look before you ask me for something already posted. Most states have statutes of limitations that are a decade or less; it’s unlikely you’ll get help for a very old project. If you’re interested in how I work with data, here are some of my guidelines. I sometimes also provide: press coverage; overhead slides (as PDF files); presentations (as PowerPoint files); original programs; links and/or responses to my critics; and so forth.
