Comparing Experimental and Matching Methods Using a Large-Scale Voter Mobilization Experiment

Alan S. Gerber and Donald P. Green Yale University, Institution for Social and Policy Studies, P.O. Box 208209, 77 Prospect Street, New Haven, CT 06520 e-mail: kevin.arceneaux{at}temple.edu (corresponding author) e-mail: alan.gerber{at}yale.edu e-mail: donald.green{at}yale.edu In the social sciences, randomized experimentation is the optimalresearch design for establishing causation. However, for a numberof practical reasons, researchers are sometimes unable to conductexperiments and must rely on observational data. In an effortto develop estimators that can approximate experimental resultsusing observational data, scholars have given increasing attentionto matching. In this article, we test the performance of matchingby gauging the success with which matching approximates experimentalresults. The voter mobilization experiment presented here comprisesa large number of observations (60,000 randomly assigned tothe treatment group and nearly two million assigned to the controlgroup) and a rich set of covariates. This study is analyzedin two ways. The first method, instrumental variables estimation,takes advantage of random assignment in order to produce consistentestimates. The second method, matching estimation, ignores randomassignment and analyzes the data as though they were nonexperimental.Matching is found to produce biased results in this applicationbecause even a rich set of covariates is insufficient to controlfor preexisting differences between the treatment and controlgroup. Matching, in fact, produces estimates that are no moreaccurate than those generated by ordinary least squares regression.The experimental findings show that brief paid get-out-the-votephone calls do not increase turnout, while matching and regressionshow a large and significant effect.     

Automated PCR setup for forensic casework samples using the Normalization Wizard and PCR Setup robotic methods

Human genome, pharmaceutical and research laboratories have long enjoyed the application of robotics to performing repetitive laboratory tasks. However, the utilization of robotics in forensic laboratories for processing casework samples is relatively new and poses particular challenges. Since the quantity and quality (a mixture versus a single source sample, the level of degradation, the presence of PCR inhibitors) of the DNA contained within a casework sample is unknown, particular attention must be paid to procedural susceptibility to contamination, as well as DNA yield, especially as it pertains to samples with little biological material. The Virginia Department of Forensic Science (VDFS) has successfully automated forensic casework DNA extraction utilizing the DNA IQ(trade mark) System in conjunction with the Biomek 2000 Automation Workstation. Human DNA quantitation is also performed in a near complete automated fashion utilizing the AluQuant Human DNA Quantitation System and the Biomek 2000 Automation Workstation. Recently, the PCR setup for casework samples has been automated, employing the Biomek 2000 Automation Workstation and Normalization Wizard, Genetic Identity version, which utilizes the quantitation data, imported into the software, to create a customized automated method for DNA dilution, unique to that plate of DNA samples. The PCR Setup software method, used in conjunction with the Normalization Wizard method and written for the Biomek 2000, functions to mix the diluted DNA samples, transfer the PCR master mix, and transfer the diluted DNA samples to PCR amplification tubes. Once the process is complete, the DNA extracts, still on the deck of the robot in PCR amplification strip tubes, are transferred to pre-labeled 1.5 mL tubes for long-term storage using an automated method. The automation of these steps in the process of forensic DNA casework analysis has been accomplished by performing extensive optimization, validation and testing of the software methods.     

Developmental validation of the quantifiler real-time PCR kits for the quantification of human nuclear DNA samples

The Quantifiler Human DNA Quantification Kit and the Quantifiler Y Human Male DNA Quantification Kit were designed for the quantification of human genomic DNA in forensic samples. The kits use a real-time PCR-based process to quantify, respectively, total human DNA or human male DNA only. We report the results of a developmental validation study that we performed with the Quantifiler Kits, following the official SWGDAM guidelines. The Quantifiler Kits were tested for performance criteria such as species specificity, sensitivity, stability, precision and accuracy, and in addition, were tested with forensic case-type samples and mixed (male:female) DNA samples. The Quantifiler Kit methods were highly specific for human DNA, and could detect as little as 32 picograms of DNA using 2 microL of sample per assay. The accuracy and precision of the Quantifiler Kit methods was comparable or superior to that of other quantification methods.     

Relationship between depressed mood and sex-typed personality characteristics in adolescents

The current study investigated the relationship between self-reports of depressed mood, self-esteem, and endorsement of sex-typed personality characteristics in a normal adolescent population. For positively valenced (socially desirable) sex-typed characteristics, masculinity was inversely related to depression while femininity was not significantly related. In contrast, for negatively valenced sex-typed characteristics the masculine and feminine scales showed similar effects: all were positively correlated with depression. Self-esteem was the best single predictor of depression and none of the sextyped characteristics contributed significantly beyond that accounted for by self-esteem. The results support previous findings with younger children and high school students by demonstrating that sex-typed characteristics in adolescents are significantly associated with depression, although this relationship is accounted for by the variance shared with self-esteem, a more general measure of mental health. However, the presence of negative feminine-typed characteristics was as highly correlated with depression as the absence of positive masculine characteristics and both relationships were stronger for females than males. Thus the linkages between sex-role socialization, sex-typed characteristics, self-esteem, depressed mood, and clinical depression bear further investigation, and may prove useful in understanding the development of depression in males and females.Received Ph.D. in clinical psychology from the Pennsylvania State University in 1976. Current research interests include the development of depression and eating disorders.Received Ph.D. in Counseling Psychology from the Pennsylvania State University in 1981. Current research interest is childhood depression.     

The theoretical aspects of the time dependent Z equation as a means of postmortem interval estimation using body temperature data only

It has been clearly demonstrated that the rectal cooling curve does not obey Newton's Law, which is exponential. The first success in modelling rectal cooling mathematically was achieved by Marshall and Hoare [1]. An amendment was made to the simple exponential curve which led to a good mathematical model, exhibiting the three main sections of rectal cooling, i.e. lag, linear and quasi-exponential. The resultant method of postmortem interval estimation required a knowledge of the body mass and height. The present study has led to a totally different amendment to Newton's Law, which provides a means of postmortem interval estimation from body temperature data only. The derivation of the method, with a background on Newton's Law follows.