Inference in differences-in-differences with few treated groups and heteroskedasticity

Differences-in-Differences (DID) is one of the most widely used identification strategies in applied economics. However, how to draw inferences in DID models when there are few treated groups remains an open question. We show that the usual inference methods used in DID models might not perform well when there are few treated groups and errors are heteroskedastic. In particular, we show that when there is variation in the number of observations per group, inference methods designed to work when there are few treated groups tend to (under-) over-reject the null hypothesis when the treated groups are (large) small relative to the control groups. This happens because larger groups tend to have lower variance, generating heteroskedasticity in the group x time aggregate DID model. We provide evidence from Monte Carlo simulations and from placebo DID regressions with the American Community Survey (ACS) and the Current Population Survey (CPS) datasets to show that this problem is relevant even in datasets with large numbers of observations per group. We then derive an alternative inference method that provides accurate hypothesis testing in situations where there are few treated groups (or even just one) and many control groups in the presence of heteroskedasticity. Our method assumes that we can model the heteroskedasticity of a linear combination of the errors. We show that this assumption can be satisfied without imposing strong assumptions on the errors in common DID applications. With many pre-treatment periods, we show that this assumption can be relaxed. Instead, we provide an alternative inference method that relies on strict stationarity and ergodicity of the time series. Finally, we consider two recent alternatives to DID when there are many pre-treatment periods. We extend our inference methods to linear factor models when there are few treated groups. We also derive conditions under which a permutation test for the synthetic control estimator proposed by Abadie et al. (2010) is robust to heteroskedasticity and propose a modification on the test statistic that provided a better heteroskedasticity correction in our simulations.

Inference in differences-in-differences with few treated groups and heteroskedasticity

Differences-in-Differences (DID) is one of the most widely used identification strategies in applied economics. However, how to draw inferences in DID models when there are few treated groups remains an open question. We show that the usual inference methods used in DID models might not perform well when there are few treated groups and errors are heteroskedastic. In particular, we show that when there is variation in the number of observations per group, inference methods designed to work when there are few treated groups tend to (under-) over-reject the null hypothesis when the treated groups are (large) small relative to the control groups. This happens because larger groups tend to have lower variance, generating heteroskedasticity in the group x time aggregate DID model. We provide evidence from Monte Carlo simulations and from placebo DID regressions with the American Community Survey (ACS) and the Current Population Survey (CPS) datasets to show that this problem is relevant even in datasets with large numbers of observations per group. We then derive an alternative inference method that provides accurate hypothesis testing in situations where there are few treated groups (or even just one) and many control groups in the presence of heteroskedasticity. Our method assumes that we know how the heteroskedasticity is generated, which is the case when it is generated by variation in the number of observations per group. With many pre-treatment periods, we show that this assumption can be relaxed. Instead, we provide an alternative application of our method that relies on assumptions about stationarity and convergence of the moments of the time series. Finally, we consider two recent alternatives to DID when there are many pre-treatment groups. We extend our inference method to linear factor models when there are few treated groups. We also propose a permutation test for the synthetic control estimator that provided a better heteroskedasticity correction in our simulations than the test suggested by Abadie et al. (2010).

There is no difference in nitric oxide metabolites and neonatal outcome between premature infants born to pre-eclamptic and those born to normotensive women

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

A sex difference in seasonal timing of birth in a livebearing fish

Sex differences in seasonal timing include differences in hatch- or birth-date distribution and differences in the timing of migration or maturation such as protandrous arrival timing (PAT), which is early male arrival at breeding sites. I describe a novel form of protandrous arrival timing, as a sex difference in birth-date distribution in a live-bearing fish (Dwarf Perch, Micrometrus minimus). In this species, birth coincides with arrival at breeding sites because newborn males are sexually active. A series of samples of pregnant females and young of year was collected in Tomales Bay, CA. I analyzed the daily age record in otoliths to estimate the conception date of broods and the age that young-of-year individuals were born. Males were born at a younger age than females, as indicated by the daily age record and also by the predominance of females in broods from which some young had already been born, which was a common occurrence in pregnant females with older embryos. Sex ratio of broods varied with conception date such that early-season broods were predominantly male, possibly as a result of temperature-dependent sex determination. The combined effects of the sex difference in age at birth and seasonal shift in sex ratio were to shift the mean birth date of males relative to females by five days. The most likely ultimate explanation for PAT in the Dwarf Perch is that it arises from exploitation (scramble) competition for mating opportunities among recently-born young-of-year males.

Gender Difference in Emotional and Behavioral Responses of Being Rendered Invisible

This study examined gender differences in emotional and behavioral responses to an experience of being invisible to others. Invisibility was defined as being ignored, slighted and overlooked by others. Participants recalled their own experience and answered questions about it and their responses on an anonymous web-based survey. Although such experiences could be very unpleasant, people may respond to such negative experiences very differently. It was hypothesized that in a patriarchal society like the United States in which men hold more power than women, that men would show emotion that was more aggressive such as anger, and respond more violently to incidents they were not respected. Women, on the other hand, were expected to be more subservient in their behavior and responses, show submissive emotions such as sadness, and respond less violently when they were not respected.

Cooperative or collaborative learning: Is there a difference in university students’ perceptions?

The hypothesis that the same educational objective, raised as cooperative or collaborative learning in university teaching does not affect students’ perceptions of the learning model, leads this study. It analyses the reflections of two students groups of engineering that shared the same educational goals implemented through two different methodological active learning strategies: Simulation as cooperative learning strategy and Problem-based Learning as a collaborative one. The different number of participants per group (eighty-five and sixty-five, respectively) as well as the use of two active learning strategies, either collaborative or cooperative, did not show differences in the results from a qualitative perspective.

Rethinking difference in the iWorld: possibilities, challenges and ‘unexpected consequences’ of digital tools in literacy education

Within contemporary literacy classrooms, mobile touchscreen devices are occupying a more prominent place. For children who have disabilities or learning differences, such devices can offer increased participation and access and may also provide social capital to users. We share examples of how iPads and iPods were successfully used in classrooms by children who might be categorized as experiencing various challenges, as well as autobiographical examples we have experienced as parents of children with disabilities. Through these illustrations, we examine the possibilities of ‘new tools’ as well as challenges encountered in changing existing literacy practices.

Gender difference in the use of coronary interventions for patients with acute coronary syndrome: experience from a major metropolitan hospital in Melbourne, Australia

BACKGROUND: Literature suggests an ongoing gender disparity in the use of coronary angiography and subsequent interventions among patients with acute coronary syndrome (ACS). OBJECTIVES: The study aimed to examine gender differences in the use of coronary interventions amongst patients with acute coronary syndrome (ACS) admitted to a major metropolitan hospital in Melbourne during the period 2009-2012. METHODS: We undertook a retrospective analysis of a hospital database of 2096 ACS patients. ACS included unstable angina (UA), ST-segment-elevation myocardial infarction (STEMI) and non-STEMI (NSTEMI). RESULTS: The mean age of the patients was 64.3 years and 624 (30%) were women. Half of them were diagnosed as NSTEMI, 23% as STEMI and 25% as UA. Compared to men, women were older at admission, less likely to be diagnosed with STEMI and less likely to smoke. No gender difference was observed for severe co-morbidities or use of coronary angiography. Women diagnosed with STEMI were 39% less likely to receive an angioplasty stent (adjusted odds ratio 0.61, 95% confidence intervals 0.39-0.96) and 66% less likely to receive grafts (adjusted OR 0.34, 95% CIs 0.13-0.93). Women diagnosed with NSTEMI were 44% less likely to receive grafts (adjusted OR 0.56, 95% CIs 0.37-0.83). Younger women aged 35-49 years were less likely to receive an angioplasty stent, and older women >50 years were less likely to receive grafts. CONCLUSIONS: Adherence to guideline based treatment will help to ensure knowledge translation from guideline to practice. Further research investigating symptom presentation, use of non-invasive tests and medical management of ACS by gender may further explain gender difference for coronary interventions.

Through Eurocentric logics : The construction of difference in foreign news discourse on Venezuela

This study aims to explore the construction of difference in foreign news discourse on culturally similar but politically different non-Western subjects. Applying critical discourse analysis (CDA) together with a critique of Eurocentrism, the study examines difference in newspaper constructions of government supporters and oppositional groups in Venezuela. Discursive differences are evident in the strategies used for constructing the two groups with regard to political rationality and violence. Government supporters are associated with social justice, Venezuela’s poor, dogmatic behavior, and the use of political violence. The opposition, in contrast, is constructed as following a Western democratic rationale that stresses anti-authoritarianism. This group is primarily associated with victims of violence. While the opposition is conveyed as being compatible with Eurocentric values and practices, government supporters to great extent deviate from these norms. Such constructions serve to legitimize politico-ideological undercurrents of Eurocentrism, as the defense of liberalism.

Diversity and difference in young children’s academic experience : an issue for globalization

One view of early childhood education concerns the preparation of young children for their future lives. It is arguedin this paper that issues of preparing young children for future should be addressed in terms of living in the contextof increasing globalization. The paper is based on the academic experiences of young children in China, Americaand New Zealand. The idea is to identify congruence and differences in academic experiences of young children inthese countries and set them in the context of globalization. Drawing on the perspective of transworld experience inglobalization, the paper places diverse learning experiences of young children in an interconnected world.Contemporary early childhood education needs to change in ways that capture contemporary realties.The papertakes the view that a more globally interconnected approach to learning of young children is a contemporary realityin early childhood education. This has implications for attempts to bring out into the open diverse ways ofchildren’s learning and development. These attempts constitute an essential basis on which to provoke thinkingabout the significance of shifting early childhood education from a local orientation to a more encompassing viewof global practice.

Aboriginal public health and education : has it made a difference in practice?

Over the years, public health in relation to Australian Aboriginal people has involved many individuals and groups including health professionals, governments, politicians, special interest groups and corporate organisations. Since colonisation commenced until the1980s, public health relating to Aboriginal and Torres Strait Islander people was not necessarily in the best interests of Aboriginal and Torres Strait Islander people, but rather in the interests of the non-Aboriginal population. The attention that was paid focussed more generally around the subject of reproduction and issues of prostitution, exploitation, abuse and venereal diseases (Kidd, 1997). Since the late 1980s there has been a shift in the broader public health agenda (see Baum, 1998) along with public health in relation to Aboriginal and Torres Strait Islander people (NHMRC, 2003). This has been coupled with increasing calls to develop appropriate tertiary curriculum and to educate, train, and employ more Aboriginal and Torres Strait Islander and non-Aboriginal people in public health (Anderson et al., 2004; Genat, 2007; PHERP, 2008a, 2008b). Aboriginal and Torres Strait Islander people have been engaged in public health in ways in which they are in a position to influence the public health agenda (Anderson 2004; 2008; Anderson et al., 2004; NATSIHC, 2003). There have been numerous projects, programs and strategies that have sought to develop the Aboriginal and Torres Strait Islander Public Health workforce (AHMAC, 2002; Oldenburg et al., 2005; SCATSIH, 2002). In recent times the Aboriginal community controlled health sector has joined forces with other peak bodies and governments to find solutions and strategies to improve the health outcomes of Aboriginal and Torres Strait Islander peoples (NACCHO & Oxfam, 2007). This case study chapter will not address these broader activities. Instead it will explore the activities and roles of staff within the Public Health and Research Unit (PHRU) at the Victorian Aboriginal Community Controlled Health Organisation (VACCHO). It will focus on their experiences with education institutions, their work in public health and their thoughts on gaps and where improvements can be made in public health, research and education. What will be demonstrated is the diversity of education qualifications and experience. What will also be reflected is how people work within public health on a daily basis to enact change for equity in health and contribute to the improvement of future health outcomes of the Victorian Aboriginal community.

Difference in mosquito species (Diptera: Culicidae) and the transmission of Ross River Virus between coastline and inland areas in Brisbane, Australia

This study examined the distribution of major mosquito species and their roles in the transmission of Ross River virus (RRV) infection for coastline and inland areas in Brisbane, Australia (27°28′ S, 153°2′ E). We obtained data on the monthly counts of RRV cases in Brisbane between November 1998 and December 2001 by statistical local areas from the Queensland Department of Health and the monthly mosquito abundance from the Brisbane City Council. Correlation analysis was used to assess the pairwise relationships between mosquito density and the incidence of RRV disease. This study showed that the mosquito abundance of Aedes vigilax (Skuse), Culex annulirostris (Skuse), and Aedes vittiger (Skuse) were significantly associated with the monthly incidence of RRV in the coastline area, whereas Aedes vigilax, Culex annulirostris, and Aedes notoscriptus (Skuse) were significantly associated with the monthly incidence of RRV in the inland area. The results of the classification and regression tree (CART) analysis show that both occurrence and incidence of RRV were influenced by interactions between species in both coastal and inland regions. We found that there was an 89% chance for an occurrence of RRV if the abundance of Ae. vigifax was between 64 and 90 in the coastline region. There was an 80% chance for an occurrence of RRV if the density of Cx. annulirostris was between 53 and 74 in the inland area. The results of this study may have applications as a decision support tool in planning disease control of RRV and other mosquito-borne diseases.