We request that any publications that come from these data sets cite the paper that originated this data set in their work.

Simple tests for the extent of vote fraud with absentee and provisional ballots in the 2020 US presidential election

STATA data files

Do file for all the tables and other regressions.

The data provided here has precinct-level information on absentee, in-person, and provisional ballot voting for counties where alleged fraud occurred during the November 3, 2020 election and in adjacent precincts in counties where no fraud was alleged. County voter turnout is also provided for the ten swing states during the election.

How is crime spreading in Los Angeles County over the last couple of years.

Explaining LA crime data collection

LA Showing the changing shares of crime rates by type of area Excel file

LA Showing the absolute annualized changes in crime by type of area Excel file

Original raw data in STATA file

Regressions, do files and data to replicate the changes in relative shares

Regressions, do files and data to replicate the actual relative changes in crime

The charts cover the zip codes in Los Angeles County. The zip codes are categorized based on the median property values or race. So you identify those zip codes where the median property values are less than $400K, $400K to $500K, $500K to $750K, $750K to $1 million, and so on. I used property values and demographic information from 2000 to identify and categorize zip codes. You then count up the number of each type of crime in each month for each of those zip code categories. Once you have the total number of robberies in Los Angeles County in January 2019 and the number of robberies in zip codes where the median property values are less than $400K, you will know the share of all robberies in that type of area. That is then calculated for each month from January 2019 through January 2022, and I ran simple ordinary least squares regressions with linear time trends on that data to measure how the share of robberies in that type of neighborhood is changing at an annual rate over those 37 months. The ways that the zip codes were categorized by median property values followed the categories used by the Census. The same approach was applied to other demographic information on race and income.

The second excel file shows the absolute change in crime by type of zip code.


Detailed information on mass public shootings from 1998 through January 26, 2023

An Excel spreadsheet that lists all the mass public shootings over this period is available here: Mass Public Shooting cases 1998 through January 1, 2023.


John R. Lott, Jr. and Carl Moody, “Do White Police Officers Unfairly Target Black Suspects?” Economics, Law, and Policy, December 2021.

The “do” and “log” files as well as the data and instructions are available here.

mSchool Shooting data 1998 through August 2018 — List of all school shootings of any type over those years

The Excel file with the data is available here.


Red Flag/Extreme Risk Order Laws

The paper is available here. Information to replicate tables and robustness checks for paper is available here.


School Shooting Data January 2000 to November 2021

CPRC’s school shooting data is derived from diverse sources including the K-12 School Shooting Database (K-12 SSDB)[1] by the Center for Homeland Defense and Security (CHDS), Gunfire on School Grounds database by Everytown, law enforcement reports, news coverage, correspondence with school supervisors, administrators or related personnel, and other publicly available information.

The data is available here.


How a Botched Study Fooled the World About the U.S. Share of Mass Public Shootings: U.S. Rate is Lower than Global Average” 2019

The paper is available to be downloaded here.

“Brought Into the Open: How the U.S. Compares to Other Countries in the Rate of Public Mass Shooters” 2020

This paper is available to be downloaded here.

Appendix 1 Foreign Cases

Appendix 2 US Cases

Excel file for International Mass Public Shootings (more information on the exact calculations is available on request, the rest of the file was too large to place on the website for download)

The data for mass public shootings around the world from 1998 to 2017 is available here.

“Undocumented Immigrants, U.S. Citizens, and Convicted Criminals in Arizona,” by John R. Lott, Jr.

The paper is available here.  The data as a STATA file is available here.  See also the codebook and the “do file” (also a STATA file).


Data showing murders are becoming even more concentrated in a handful of urban counties.

Our study is available here.

An Excel file with the data is available here.


Documented information on attacks on gun-free zones (see here).

“Do White Police Officers Unfairly Target Black Suspects?” by John R. Lott, Jr. and Carlisle E. Moody

The paper is available here.

Data and do file can be downloaded here.

Australian Data on homicides and suicides from 1915 to 2014 is available here.

More Guns, Less Crime (3rd Ed., University of Chicago Press, 2010)

MGLCRevision graphing before and after by year.smcl

MGLCRevision Table 4b reverse order.smcl


MGLCRevision Table 4a.smcl

MGLC Rev State 4b.do

By Decade Regressions.do

MGLC Rev State 4a.do

Most of this data involves STATA 7.0 data sets. The reason for using this is that the county level data involves a much larger set of control variables than can readily be handled by other statistical packages.  To help with the variable names:

for the demographics

pp = percent of the population

w= white

b= black

n= neither black or white

m= male

f= female

1019= 10 to 19 years of age, etc

so ppbm1019 = percent of the population that is black male 10 to 19 years of age

rpcpi = real per capita personal income

rpcim = real per capita income maintenance (welfare type payments)

rpcui = real per capita unemployment insurance

rpcrpo = real per capita retirement gov payments (SS, medicare)

CCWage = age eligible to have a concealed handgun permit

assaultweaponsban2 = assault weapon ban, either state or federal

the crime data is the natural log of the crime rate (though .01 is added to the crime rate when it is equal to zero)

ao stands for the arrest rate data by type of crime

The data and its sources are described in the data appendix for my book More Guns, Less Crime (University of Chicago Press, 3rd edition, 2010).

More data sets can be obtained by clicking on the following links which will take you to the download page:

The data and “do” file for the original paper by Lott and Mustard, “Crime, Deterrence, and Right-to-Carry Concealed Handguns,” Journal of Legal Studies, Vol. 26, no. 1, January 1997: 1-68 is available here.
STATA data set: sendcounty1997.dta
Do file: simpleoldverification.do

Terms of Use: All material posted on this website has been copyrighted since 2013 by John R. Lott, Jr. All rights reserved.