Adjusting for Confounding with Text Matching

We identify situations in which conditioning on text can address confounding in observational studies. We argue that a matching approach is particularly well-suited to this task, but existing matching methods are ill-equipped to handle high-dimensional text data. Our proposed solution is to estimate a low-dimensional summary of the text and condition on this summary via matching. We propose a method of text matching, topical inverse regression matching, that allows the analyst to match both on the topical content of confounding documents and the probability that each of these documents is treated. We validate our approach and illustrate the importance of conditioning on text to address confounding with two applications: the effect of perceptions of author gender on citation counts in the international relations literature and the effects of censorship on Chinese social media users.     

Children’s Use of a ‘Time Line’ to Indicate When Events Occurred

Children who allege abuse are often asked to provide temporal information such as when the events occurred. Yet, young children often have difficulty recalling temporal information due to their limited knowledge of temporal patterns and linguistic capabilities. As time is an abstract concept (we cannot see it), some investigators have begun to use ‘time-lines’ or pictorial representations of time to aid children. Yet, there is no published research testing whether children are able to use time-lines and whether they can provide adequate temporal information using them. We tested whether children could indicate the time-of-day of events using a pictorial time-line and then compared their responses to their parents’. Seven- to 8-year-olds were most consistent with parental estimates while 4-year-olds were least consistent. Responses from the 5- to 6-year-olds depended on the temporal task. Guessing and using general knowledge to estimate the time-of-day were ruled out, and so children were genuinely drawing on episodic memory when making time-line judgments. Thus, there was a developmental progression in children’s use of physical representations to communicate abstract information. These results are promising for the use of the time-line in forensic settings but much more research is needed.     

Isolation and Individualization of Conceptus and Maternal Tissues from Abortions and Placentas for Parentage Testing in Cases of Rape and Abandoned Newborns*

Abstract: Abortion specimens are often submitted to forensic laboratories as the only piece of physical evidence in rape and incest cases. The recovery of conceptus tissues from this evidence permits the use of paternity testing to evaluate suspects. In cases of abandoned newborns, the recovery of maternal tissue from the placenta allows for the direct comparison of genetic profiles between the suspected mother and the biological mother. We report on the identification and isolation of conceptus tissues from embryonic‐ and fetal‐period abortions, and maternal tissues from delivered placentas, by gross and low‐magnification examination with manual dissection. Hundreds of single‐source samples have been successfully recovered by this method and short tandem repeat typed using standard forensic procedures. We additionally describe extraembryonic tissues that can be recovered and typed in the absence of the embryo proper. We conclude that an expertise and protocols can be developed by forensic laboratories for the routine analysis of this evidence.     

Differential effects of general versus cued invitations on children’s reports of a repeated event episode

The ability to describe individual episodes of repeated events (such as ongoing abuse) can enhance children’s testimony and assist the progression of their cases through the legal system. Open-ended prompts have been advocated as a means to assist children in accurately retrieving information about individual episodes. In the current study, two subtypes of open-ended prompts (cued and general invitations) were compared for their effects on five- to nine-year-olds’ (n?=?203) reports about individual episodes of a repeated event. Interviews occurred 1–2 weeks after the last of 4 event sessions. Cued invitations assisted children to provide specific details about individual episodes of a repeated event, while general invitations were useful to elicit more broad happenings of the episodes. The accuracy of responses to general invitations was similar for children of all ages up to one week after the event, but at a longer interview delay younger children were less accurate than older children. There were no differences in the accuracy of responses to cued invitations as a function of age or interview delay. Results suggest that interviewers tasked with eliciting accounts of individual episodes from a repeated event, such as ongoing abuse, should consider the differential efficacy of each prompt-type on children’s reports.     

Proximal End Root Morphology Characteristics in Antemortem Anagen Head Hairs

The proximal end morphology of antemortem anagen head hair was compared with the characteristics documented to occur in postmortem hairs. Antemortem anagen and telogen head hairs (N = 967) were recovered following exposure to seven environments. Root morphology characteristics consistent with those reported in postmortem hairs were observed in 66 (14%) hairs exposed to a water, normal saline, outdoor soil, or indoor shower environment. Thirty‐three anagen hairs (7%) exhibited a root band at the proximal end. The mean distance from the root tip to the onset of the root band ranged from 0.23 to 0.7 mm, depending on the environment. The mean distance from the root tip to the onset of the root band was 0.46 mm, with a mean band length of 0.44 mm. The results illustrate the need to better characterize postmortem banding through quantitative measurements, including the range for root tip to band distance and the overall band length.     

The Effect of School Dropout on Verbal Ability in Adulthood: A Propensity Score Matching Approach

Compared to high school graduates, adolescents who drop out of school are more likely to have a range of negative outcomes, including lower verbal capacities; however, the true nature of this association is not well-understood. Dropping out of school could have an important effect on reducing verbal skills, or the link between dropping out of school and diminished verbal skills could be a spurious association that is the result of unmeasured confounding variables. The current study tested these two competing perspectives by using propensity-score-matching (PSM) to unpack the association between school dropout and verbal skills among 7,317 respondents from the National Longitudinal Study of Adolescent Health (51% female, 49% male; 62% Caucasian, 38% minority). The results of the PSM models indicated a small yet meaningful statistically significant effect of dropout on verbal skills in adulthood even after taking into account a range of confounders. We conclude by discussing the implications of our results.