Moderator Assistant: A Natural Language Generation-Based Intervention to Support Mental Health via Social Media

As online mental health support groups become increasingly popular, they require more support from volunteers and trained moderators who help their users through “interventions” (i.e., responding to questions and providing support). We present a system that supports such human interventions using Natural Language Generation (NLG) techniques. The system generates draft responses aimed at reducing moderators’ workload, and improving their efficacy. NLG and human interventions were compared through the ratings of 35 psychology interns. The NLG-based system was capable of generating messages that are grammatically correct with clear language. The system needs improvement, however, moderators can already use it as draft responses.     

Exploring the Predictors of Organizational Preparedness for Natural Disasters

There is an extensive body of research on the determinants of disaster preparedness at the individual and household levels. The same cannot be said for the organizational level. Hence, the purpose of this study is to shed light on the predictors of organizational preparedness for natural disasters. Since leaders of organizations have an incentive to overstate their level of preparedness and because surveys of organizational leaders suffer from selection bias and low response rates, we take the novel approach of interviewing employees about the organizations that employ them. Using an online survey, we collected information from a national sample of 2,008 U.S. employees and estimated the predictors of preparedness at the organizational level. We find, among other results, that organization size (facility level) is a consistent predictor of preparedness at the organizational level. We conclude with policy recommendations and outline an agenda for future research on organizational preparedness for natural disasters.     

First Order Dominance Techniques and Multidimensional Poverty Indices: An Empirical Comparison of Different Approaches

In this empirically driven paper we compare the performance of two techniques in the literature of poverty measurement with ordinal data: multidimensional poverty indices and first order dominance techniques (FOD). Combining multiple scenario simulated data with observed data from 48 Demographic and Health Surveys around the developing world, our empirical findings suggest that the FOD approach can be implemented as a useful robustness check for ordinal poverty indices like the multidimensional poverty index (MPI; the United Nations Development Program’s flagship poverty indicator) to distinguish between those country comparisons that are sensitive to alternative specifications of basic measurement assumptions and those which are not. To the extent that the FOD approach is able to uncover the socio-economic gradient that exists between countries, it can be proposed as a viable complement to the MPI with the advantage of not having to rely on many of the normatively binding assumptions that underpin the construction of the index.     

Social resources and Arab women’s perinatal mental health: A systematic review

Background

Women’s mental health in the perinatal period is understudied worldwide and in Arab countries especially.

An application of negative binomial distribution of order k in sensitive surveys

This article is an attempt to generalize some of the recent papers on randomized response techniques by using the negative binomial distribution of order k to randomize the responses in the randomization design where respondents can report outcome of one of two binary devices depending upon their actual status. The relative efficiency results are observed to be better than those of many recent and relevant randomized response techniques. The results are also better than those of the base line model used in this study, providing the sensitive attribute is rare. An extra advantage of the proposed technique is that it does not require any additional sampling and administrative cost.     

Poverty and Income Replacement Profile Among EPF Retiree in Malaysia

Social security policies often focus on replacement rates, which indicate retirement income or social security benefits in relation to preretirement income. The higher replacement rate among the pensioners would ensure that the ageing society would have sufficient income to lead a normal life after retirement. This study examined factors that influencing the income replacement rate of Malaysian Employee Provident Fund (EPF) retirees. The analysis was based on a nationwide survey conducted in 2013–2014 among EPF retiree. A logit model was used to evaluate the likelihood of selected socio-demographic and economic factors contributing to income of the retirees. The results indicated that around 62 % of elderly has lower retirement income compared to their preretirement income and it makes them more vulnerable to unpredictable events and financial conditions. The study suggested that the income replacement rate of elderly could be strengthened by investing in ageing workforce, raising retirement age, enhancing educational achievements of low income groups and restructuring employment. These may in turn increase the availability of skilled workers, enhancing the national productivity, increase the income security of retirees, reduce poverty, and develop economic growth of the country.     

A new index for measuring evenness

Diversity and evenness play a significant role to satisfy the basic needs like food, drinking water, fuel, shelter, and medicine. Hence, these phenomena need to be measured with great precision. We propose a new index for measuring the evenness phenomenon quantitatively based on the frequencies (number of individuals) of different communities (species, families etc.). The new index is compared with fourteen other indices based on the different criteria laid down by Smith and Wilson (1996). Our proposed index, named as Z-A Index, is easy to calculate and easy to understand. Its standard limits between 0 and 1 make this new index interpretable logically. The comparison with the existing indices indicates that the newly proposed index performs well in the sense that it satisfies 11 of the 14 requirements and desirable features given by Smith and Wilson (1996).     

Enrolling girls without learning: Evidence from public schools in Afghanistan

While more girls are now attending school in Afghanistan than prior to 2010, there is a lack of evidence on how school attendance is helping their cognitive development. We use data from a large sample of all‐girls state‐funded school students to estimate for the first time the “learning profile” for Afghanistan. Students enrolled in grades 4 to 9 were assessed using the Early Grade Reading Assessment (EGRA) and the Early Grade Mathematics Assessment (EGMA) test items. We find that higher grade progression (particularly in grades 6–9) leads to almost no gain in numeracy (addition and subtraction) skills that should have been achieved in early grades. Similar results follow from the analysis of student performance in the EGRA test, particularly in oral reading fluency (correct words read per minute) and oral reading comprehension (number of correct answers out of 12). These findings warn that simply enrolling girls in school without improving the relationship between grade completion and learning is unlikely to transform the lives of women in Afghanistan.     

Unemployment in Urban China: an Analysis of Survey Data from Shanghai

Abstract. We analyse the unemployment problem in the largest Chinese city, Shanghai, based on a survey of the unemployed conducted in late 1989. We find, as expected, that unemployment is particularly serious among school-leavers and contract and temporary workers, but also find a surprisingly high incidence of unemployment amongst permanent state employees and relatively long unemployment durations. The Shanghai unemployed rely heavily on occasional work for support, and only a small proportion-receive state benefit. We identify problems of interpretation stemming from the stock-sampling nature of the survey, and use non-parametric methods to reveal an underlying strong downward movement in individual re-employment probabilities.