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.     

Race, economic inequality, and violent crime

The current study used data drawn from the National Incident-Based Reporting System (NIBRS) and the census to investigate the relationship between indicators of interracial and intraracial economic inequality and violent crime rates, including White-on-Black, White-on-White, Black-on-White, and Black-on-Black offenses. Multivariate regression results for ninety-one cities showed that while total inequality and intraracial inequality had no significant association with offending rates, interracial inequality was a strong predictor of the overall violent crime rate and the Black-on-Black crime rate. Overall, these results were interpreted as consistent with J.R. Blau and Blau's (1982) relative deprivation thesis, with secondary support for P.M. Blau's (1977) macrostructural theory of intergroup relations. The findings also helped to clarify the unresolved theoretical issue regarding which reference group was most important in triggering relative deprivation among Blacks. It appeared that prior studies were unable to find support for the relative deprivation thesis for Black crime rates because of data and methodological limitations.     

Postmortem urine immunoassay showing false-positive phencyclidine reactivity in a case of fatal tramadol overdose

This is a report of postmortem false-positive reactivity using an enzyme-multiplied urine phencyclidine (PCP) immunoassay (EMIT II+) due to a single-agent fatal tramadol overdose. An autopsy of a 42-year-old male who died alone at home revealed no identifiable lethal anatomic abnormalities, thus leading to toxicologic analysis. Femoral blood was obtained for drug testing by high-performance liquid chromatography (HPLC) and showed a tramadol level of 14.0 mg/L, 2 orders of magnitude greater than the therapeutic range (0.1 to 0.3 mg/L). Urine was also obtained and EMIT II+ immunoassay revealed positivity for PCP at 88 mAU/min. However, confirmatory testing by HPLC failed to identify PCP in either the urine or serum. To verify the suspicion that this was a false-positive PCP result, stock solutions of tramadol and its major metabolite (O-desmethyltramadol) at concentrations of 100 mg/L in 10% methanol/H2O were compared with a blank solution (10% methanol/H2O) for EMIT II+ PCP reactivity and demonstrated reactivities of 44 mAU/min and 27 mAU/min, respectively. While these individual results were below the cutoff reactivity for a positive EMIT II+ PCP result (ca. 85 mAU/min), they were much more reactive than the blank calibrator (set at 0 mAU/min). Therefore, we conclude that the immunoreactivity of tramadol and its metabolites in aggregate is responsible for the PCP immunoassay interference and false-positive result.     

Cultural and Spiritual Constructions of Mental Distress and Associated Coping Mechanisms of Tibetans in Exile: Implications for Western Interventions

The Tibet-TPO project in Dharamsala, North India aims to provideculturally sensitive psychosocial support to Tibetan refugees.In this study we have examined the cultural and spiritual constructionsof mental distress of Tibetan exiles from a secondary analysisof previously published data. Tibetans refugees' constructionsof mental distress were intimately linked to cultural, religiousand political factors. Family and religious support were regardedas key coping strategies, yet many new refugees lacked bothfamily support and detailed knowledge and understanding of TibetanBuddhism. Not all of those interviewed were positive about ‘westernapproaches’ to dealing with mental distress, but thoseusing the service seemed to do so in a pragmatic and integrativeway. We conclude that culturally sensitive psychosocial supportcan usefully ‘fill a gap’, especially for new refugeeswho lack both family support and access to, or understandingof, traditional religious coping mechanisms.     

Do Jurors “Know” What Isn’t So About Child Witnesses?

Are expert witnesses needed in child sexual abuse cases to educate jurors about children’s memory, suggestibility, and reactions to abuse, or do jurors already know what such experts could tell them? To cast light on this question, we surveyed jurors and jury-eligible college students and compared their beliefs with what is known via scientific research regarding children’s memory and ability to testify, reactions to interrogation, and reactions to sexual abuse. We also asked participants to infer results of four widely cited studies of children’s suggestibility. Participants’ beliefs were consistent with findings from research on some issues (e.g., that children can be led to claim that false events occurred) but diverged from the scientific consensus on other issues (e.g., whether children can remember painful events in infancy). Similarly, participants sometimes overestimated and sometimes underestimated the level of suggestibility observed in empirical studies. Individual differences in accuracy were related to participants’ gender, education and ethnicity, and there was considerable disagreement among participants on many questions. Implications of findings for the admissibility of expert testimony in child abuse cases are discussed.     

Facing rejection: new relationships, broken relationships, shame, and stalking

This study provides a preliminary examination of the relationship among shame-proneness, emotions, and persistent, unwanted courting or pursuit behaviour. A total of 222 undergraduates completed a questionnaire measuring responses to the termination of a relationship or the declining of a date. The Test of Self-Conscious Affect was used to measure shame. Although shame was unrelated to types of intrusive behaviour, individuals who engaged more repetitively in covert pursuit tended to ruminate more over their love interest than did those who rarely engaged in such behaviour. Rumination was positively correlated with shame. Self-harm behaviours and harm toward others were associated with feelings of sadness and depression, and those who engaged more often in harm toward others also reported feelings of anger and jealousy. Conclusions are made regarding the function of shame and other emotions in intrusive behaviour and their potential relevance to treatment of stalkers.