ICPSR Bulletin, 2019 Special Edition - Prize Competition Winners

ICPSR Bulletin logo

Photo of ICPSR Director Margaret Levenstein


ICPSR Announces Winners of 2019 Research Paper Competitions!

ICPSR is pleased to announce the First- and Second-Place winners of our 2019 Research Paper Competitions.  

Arnold Johnsen, (Statistics) of Northwestern University earned First Place in the Undergraduate Competition with a paper titled “Modeling Parole & Conditional Release: An Application of Predictive Analytics Techniques.”

Emalie Rell, (Sociology/Anthropology) of Elizabethtown College earned Second Place in the Undergraduate Competition with a paper titled “Mother Doesn't Always Know Best: The Effects of Sex and Support for Sex Education on Views of Teen Access to Birth Control without Parental Consent.”

Angela Lee, (Government) of Harvard University earned First Place in the Master’s Competition with a paper titled “A Time-Sensitive Analysis of the Work-Crime Relationship for Young People.”

Congratulations to all of the winners! Their work is showcased below. Thanks to everyone who submitted an entry or spread the word about the competitions!


Margaret C. Levenstein
Director, Inter-university Consortium for Political and Social Research

Undergraduate Competition Winners

Arnold Johnsen
Northwestern University

Winning entry: Modeling Parole & Conditional Release: An Application of Predictive Analytics Techniques 

Abstract: The list of released prisoners disseminated by the National Corrections Reporting Program (NCRP) documents fifteen reasons for release. Eight of them—after discounting special cases such as executions or escapes—pertain to the categories conditional release or non-conditional release. Prisoners in the latter category serve the entirety of their sentence in prison, whereas conditional releasees are allowed to serve the remainder of their sentences under community supervision. Accurately predicting conditional release is of great consequence to studying social justice and defendants’ rights, so in this paper I aim to illustrate how and to what degree different methods can improve prediction of conditional release. By analyzing missing values, state-to-state variations in parole rate, sampling methods, and different predictive models, I arrived at a useful practical guide for dealing with the NCRP data and a methodological outline for better predictive performance, both of which can serve as a foundation for more sophisticated analysis in the future.

Abstract: Many individuals are unaware of the legislation dictating the availability of contraceptives for minors. Each state’s policy for teen access to birth control varies, but a total of 46 states and the District of Columbia allow minors to receive birth control without parental consent. Previous literature has found minors who have access to birth control and parents who support the use of birth control are more likely to use contraceptives than teens with parents who are unsupportive of contraceptives. This research examines the effects of support for sex education and sex on views of teen access to birth control without parental consent. The data for this research were obtained from the 2016 wave of the General Social Survey conducted by the National Opinion Research Center at the University of Chicago. The sample size consisted of 1,521 individuals. The majority of respondents supported teen access to birth control without parental consent.


Master's Competition Winner

Abstract: While entrance into the labor market and desistance from crime both typically occur during the transition to adulthood, it is unclear whether employment causes reductions in crime for young people. Employment may reduce crime by offering routines, income, and supervision. However, selection may also occur: people may start working when they are already making positive changes in their lives and stop working when they are already making harmful decisions. To evaluate these possibilities, I model month-to-month, within-person changes in offending during the periods surrounding job transitions. Using data from Pathways to Desistance, a longitudinal study of young offenders, I find large reductions in income-related offending prior to job entry, but no further reductions after job entry. I also find that offending spikes before job exit. These patterns suggest that job transitions do not instigate changes in offending but rather occur in response to other changes in young people’s lives.
Call for entries for 2020


The ICPSR Research Paper Competitions accept entries of papers for analyses on any topic using data from ICPSR. The purpose of the competitions is to highlight exemplary research papers based on quantitative analysis that uses ICPSR data. We invite submissions from students and recent graduates at ICPSR member institutions.