Masking identification of discrete choice models under simulation methods

We present examples based on actual and synthetic datasets to illustrate how simulation methods can mask identification problems in the estimation of discrete choice models such as mixed logit. Simulation methods approximate an integral (without a closed form) by taking draws from the underlying distribution of the random variable of integration. Our examples reveal how a low number of draws can generate estimates that appear identified, but in fact, are either not theoretically identified by the model or not empirically identified by the data. For the particular case of maximum simulated likelihood estimation, we investigate the underlying source of the problem by focusing on the shape of the simulated log-likelihood function under different conditions.     

Reliability of Assessment of Reading Ability in Three Countries

Over the last few decades, there have been a number ofnational and international programmes to assessperformance in key subjects. For example theInternational Association for Evaluation ofEducational Attainment (IEA), International Assessmentof Educational Progress (IAEP), and the ThirdInternational Mathematics and Science Survey (TIMSS)programs have aimed to compare performance betweenparticipating countries.Such exercises have different requirements from thosetests, such as public examinations, which aim toassess the attainment of individuals. National andinternational assessments aim to cover a very widerange of materials, and often use the technique ofmultiple matrix sampling to do so.The investigation is based on generalizability-typeanalyses of three national data sets at two ages, andlooks not only at the variance components arising fromsampling of schools, and of pupils within them, butalso at the variation between different assessmentinstruments, and between items within assessmentinstruments. If interpreted with care, such resultscan be of value in the design of future studies. Thispaper concentrates largely on the precision ofestimates of overall means, but analyses of the typedescribed could also be used to compare theperformance of subpopulations.     

Advertising and recruitment: Marketing to minorities

Growing diversity, increasing multinationalism, and a need to better understand minority consumers make recruiting minority professionals to the workplace exceedingly important. This study examines the influences of a frequently used tool in organizational recruitment, the recruitment advertisement. Specifically, this research examines the influence of the racial composition of employees portrayed in these advertisements on a diverse sample of job‐seeker reactions. These reactions include perceptions of organizational attractiveness, perceived compatibility to the organization, and evaluations of organizational image. In addition, it was expected that the race of the perceiver, the job seeker, would moderate these relationships. Mixed support for the hypotheses was found. Implications for advertising, marketing, and recruitment research given emerging domestic diversity are offered. © 2000 John Wiley & Sons, Inc.     

Using the Family Self‐Sufficiency Program to Help Families with Housing Assistance Improve Earnings,Credit Score,and Debt Levels: A Quasi‐Experimental Analysis

HUD's Family Self‐Sufficiency (FSS) program aims to help housing assistance recipients increase their earnings and build savings to make progress toward economic security. This study examines an asset‐building nonprofit's innovative financial coaching‐based approach to FSS that adds an additional focus on helping clients build assets and financial capability. We use a quasi‐experimental approach to estimate the program's impact on earnings and cash assistance receipt and analyze credit and debt outcomes against a benchmark group. The findings show substantial, significant gains in households' earnings and significant decreases in receipt of some cash benefits. Participants also saw increases in credit score, and success in paying down credit card and derogatory debt. Our results indicate that FSS can be an effective platform for helping participants in subsidized housing make real progress toward economic security and financial health.     

The Economics of Sustainable Land Management Practices in the Ethiopian Highlands

This article uses data from household‐ and plot‐level surveys conducted in the highlands of the Tigray and Amhara regions of Ethiopia. We examine the contribution of sustainable land management (SLM) practices to net value of agricultural production in areas with low vs. high agricultural potential. A combination of parametric and non‐parametric estimation techniques is used to check result robustness. Both techniques consistently predict that minimum tillage (MT) is superior to commercial fertilisers (CFs), as are farmers’ traditional practices (FTPs) without CFs, in enhancing crop productivity in the low agricultural potential areas. In the high agricultural potential areas, in contrast, use of CFs is superior to both MT and FTPs without CFs. The results are found to be insensitive to hidden bias. Our findings imply a need for careful agro‐ecological targeting when developing, promoting and scaling up SLM practices.     

Estimating returns to soil conservation adoption in the northern Ethiopian highlands

Land degradation in the form of soil erosion and nutrient depletion presents a threat to food security and sustainability of agricultural production in many developing countries. Governments and development agencies have invested substantial resources to promote soil conservation practices as part of an effort to improve environmental conditions and reduce poverty. However, limited rigorous empirical work has been done on the economics of soil conservation technology adoption. This article investigates the impact of stone bunds on value of crop production per hectare in low and high rainfall areas of the Ethiopian highlands using cross-sectional data from more than 900 households, with multiple plots per household. We have used modified random effects models, stochastic dominance analysis (SDA), and matching methods to ensure robustness. The parametric regression and SDA estimates are based on matched observations obtained from the nearest neighbor matching using propensity score estimates. This is important, because conventional regression and SDA estimates are obtained without ensuring that there actually exist comparable conserved and nonconserved plots on the distribution of covariates. We use matching methods, random effects, and Mundlak's approach to control selection and endogeneity bias that may arise due to correlation of unobserved heterogeneity and observed explanatory variables.
We find that the three methods tell a consistent story. Plots with stone bunds are more productive than those without such technologies in semi-arid areas but not in higher rainfall areas, apparently because the moisture conserving benefits of this technology are more beneficial in drier areas. This implies that the performance of stone bunds varies by agro-ecology type, suggesting the need for designing and implementing appropriate site-specific technologies.     

Oral history and information technology: Human voices of assessment

Dramatic advances in technology for acquiring, managing, and sharing information promise to reshape the workplace by eliciting new behaviors and introducing new organizational patterns. For academic medical centers, the integration of information technology into programs of education, research, and patient care is essential for increased organizational effectiveness.

At Baylor College of Medicine, we have developed information‐sharing and management tools, collectively called the Virtual Notebook System (VNS). The VNS is a multiuser, workstation‐based hypermedia system that serves as a technologically extended analog of the laboratory notebook used in biomedical research. We are deploying the VNS in scientific groups at Baylor, and are applying oral history techniques to assess its impact.

This article shows how oral history captures the “human voices”; of Baylor's experience and helps us understand the effects of information technology on the processes of biomedical research.