Propensity Score Matching

Social funds have been defined as "agencies that finance small projects in several sectors targeted to benefit a country's poor and vulnerable groups based on a participatory manner of demand" . After the first social fund was set up in Bolivia in 1987, the idea attracted immediate support, and by May 2001 the World Bank had provided financing to more than 98 social funds in 58 countries. This is the context in which the Egyptian Social Fund for Development (SFD) was established in 1991, when the government sought to widen the social safety net in the wake of an economic reform and social adjustment program. Run by a semi-autonomous agency that reports to the office of the Prime Minister, the SFD is one of the three main safety net programs in Egypt, along with the system of subsidies of basic foodstuffs, and the Ministry of Social Solidarity's cash transfer program.

Biometrics is the science and technology of measuring and analyzing biological data of human body, extracting a feature set from the acquired data, and comparing this set against to the template set in the database. Experimental studies show that Unimodal biometric systems had many disadvantages regarding performance and accuracy. Multimodal biometric systems perform better than unimodal biometric systems and are popular even more complex also. We examine the accuracy and performance of multimodal biometric authentication systems using state of the art Commercial Off-The-Shelf (COTS) products. Here we discuss fingerprint and face biometric systems, decision and fusion techniques used in these systems. We also discuss their advantage over unimodal biometric systems.

Almost nine million children under 5 years of age die every year. Diarrhea is considered to be the second leading cause of under-five mortality in developing countries. About one out of five deaths is caused by diarrhea. In this paper, we use the newly available data set District Level Household Survey 3 to quantify the impact of access to improved sanitation on diarrheal morbidity for children less than 5 years of age in India. Using propensity score matching, we find that access to improved sanitation reduces the risk of contracting diarrhea by 2.2 percentage points. There is considerable heterogeneity in the impacts of improved sanitation. We find statistically insignificant treatment effects for children in low or middle socioeconomic status households and for girls; however, boys and children in high socioeconomic status households experienced economically significant treatment effects. The magnitude of the treatment effect differs largely by hygiene behavior.

Sustained mass media campaigns have been recommended to stem the tobacco epidemic in the United States. Propensity score matching (PSM) was used to estimate the effect of awareness of a national smoking cessation media campaign (EX 1 ) on quit attempts and cessation-related cognition. Participants were 4,067 smokers and recent quitters aged 18-49 in Downloaded from targeted U.S. media markets. Controlling for potential confounders through PSM and regression analysis, confirmed awareness of EX was not significantly associated with either outcome at 6-month follow-up. Matched analyses excluding 217 quitters resulted in a significant effect of EX on both outcomes.

We analyze the consequences of a teenage pregnancy event in the short- and long-run in Mexico. Using longitudinal and cross-section data, we match females who got pregnant and those that did not based on a propensity score. In the short-run, we find that a teenage pregnancy causes a decrease of 0.6-0.8 years of schooling, lower attendance to school, less hours of work and a higher marriage rate. In the long-run, we find a loss in years of education of 1-1.2, which implies a permanent effect on education, and that household income per capita is lower.

Using panel data econometric techniques and propensity score matching procedures, this study evaluates the impact of the MAUNLAD agricultural development program - a program designed to help alleviate poverty in Philippine coconut-producing communities. Our results ...

To reduce potential food hazards and increase the image of Thai horticultural products abroad, the Thai government introduced public standards of Good Agricultural Practices (Q-GAP). What makes orchid and mango producers in Thailand adopt Q-GAP standards and how do these affect their income and export shares? Primary data from 400 certified and non-certified orchid and mango producers was collected from main exporting provinces in Thailand. The binary probit model estimations show that it is the orchid and mango producers with higher education, and more physical and social capital who tend to comply with Q-GAP standards. Results from the Propensity Score Matching approach reveal that adoption of public GAP standards results in positive income effects for mango producers, but not for orchid producers. This can be explained by the fact that certified mango producers can sell their products to high-value retail chains which offer higher prices for their products, while certified and non-certified orchid producers cooperate with traders from the same value chain.

It is well-established that armed political conflict has a detrimental effect on food security and household welfare: conflict induces food insecurity by reducing own food production, access to food through the market, and various other resources to sustain healthy and productive lives. One way of mitigating these adverse effects is to provide food aid. In this study we evaluate the impact of a World Food Programme intervention on household food security and asset protection among conflict-affected households in Northern Uganda. We employ propensity score matching to estimate the average treatment effect on food expenditure, food consumption and preservation of assets using a sample of 1265 observations from a 2008 survey. Our results reveal that the operation's system of targeting beneficiaries was effective and in accordance with programme objectives. Food aid considerably reduced food expenditure of households, suggesting that recipients were net buyers of food, and that the food aid received was effectively consumed within the household. A corresponding positive effect on non-food expenditure was not found. Our results also indicate that food aid was effective in increasing meals consumed and in avoiding distress destocking of low value assets, but, surprisingly, only for male headed households.

Social Safety Nets (SSN) is a policy instrument that is largely used for household welfare improvement and poverty reduction. Bangladesh has implemented a number of SSN program for this purpose. The Vulnerable Group Development (VGD) program is one of the current and largest SSN two-year programs that provide in-kind transfer with development programs for selected women-headed poor family. This study is an attempt to provide evidence from a micro level household survey data on whether the VGD program can provide protection and promotion effect for the beneficiary households at the end of the benefit cycle. Based on pre-intervention criterion, similar non-beneficiary households were selected as a control group. Consumption expenditure is used as a protection, while savings, productive asset and investment behaviour, child schooling are treated as promotion indicators. The Cost of Basic Needs (CBN) method is used for poverty measurement for beneficiary households. Finally, the PSM method is employed to examine the program impact on the targeted variables. The study findings provide evidence that the VGD with integrated development program have protection and promotion roles on the beneficiary household well-beings and decrease the probability being poor for the beneficiary households compared to non-beneficiary households. Extending the number of program beneficiaries and duration of benefit cycle can help reduce the poverty rate significantly.

There has been research conducted which focused on the topic of juvenile transfer to adult court. However, less research has been done focusing on the effectiveness of juvenile transfer in reducing recidivism for transferred offenders. Of the research that has been done, most may not have adequately addressed selection bias. The current study investigates the impact of juvenile transfer on subsequent recidivism, using an eight-year follow-up period. The research is based on 308 violent youth legislatively waived to adult court in Pennsylvania. Propensity score matching is used to minimize the impact of selection bias, and sensitivity analysis is performed to determine the effect of hidden bias on the results. The findings indicate that youth processed in juvenile court have a higher recidivism rate than youth processed in adult court. Implications for subsequent research and policy are discussed.

La serie Documentos de Trabajo Sobre Economía Regional es una publicación del Banco de la República-Sucursal Cartagena. Los trabajos son de carácter provisional, las opiniones y posibles errores son de responsabilidad exclusiva de los autores y no comprometen al Banco de la República ni a su Junta Directiva.

En el presente artículo se realiza un análisis de los impactosque tiene el acceso y uso de los servicios de internet en los hogares peruanos considerando información desagregada. Así, a partir de la Encuesta Nacional de Hogares (ENAHO) en su versión de datos de panel para el período 2007-2009 y utilizando el método de Diferencias en Diferencias (DID) y una variante que permite controlar la heterogeneidad en las condiciones iniciales (PSM-DID), se muestra evidencia de los impactos que tiene el acceso a internet en un conjunto de variables que incluyen el ingreso, el gasto, la educación y el empleo.

With the financial support from various development partners, Ethiopia has designed and implemented several programs to improve household food security. Yet, food insecurity is still a major challenge to several millions of people in the country and it is questionable whether the different food security programs implemented over the past years have been successful. Using a propensity score matching method to control for pre-intervention differences, this study examined the impact on household food calorie intake of an integrated food security program (IFSP), which had been implemented in Northwestern Ethiopia by two non-governmental organizations as a case study. The estimated results provide evidence that IFSP has a positive and statistically significant effect on food calorie intake. In particular, IFSP has raised physical food calorie intake by 30% among the beneficiary households. However, we also found that IFSP has differential impact depending on family size, landownership and gender of head of household. Overall, the paper provides evidence that supporting integrated food security programs is important to improve food security in rural areas.

As U.S. correctional systems continue to rollout evidence-based programs, the utility of " complimentary " programs that do not address recidivism reduction remains in question. Many U.S. prisons have a variety of prison-based animal programs, yet the outcomes are largely unexplored. This research addresses a literature gap by evaluating the intermediate outcomes associated with a statewide prison-based dog handler program. Using propensity score matching, we compared 1,001 inmates in a pretest, postentry design, aimed at measuring change across four outcomes. Results indicate that dog handler program inmates experienced significant improvement in three of four areas. Implications and further research needs are explored.

The main focus of this article is to explore whether access to selected agricultural water management (AWM) technologies has led to significant reduction in poverty and, if they did so, to identify which technologies had higher impacts. In measuring impact we estimated the average treatment effect for the treated on crop income and measured the differences in consumption expenditures per adult equivalent of those with access and without access using matched data. The estimated average treatment effect was significant and amounted to USD 82 per season. Moreover, there was 24 less poverty incidence among users of AWM technologies compared to nonusers. All technologies were found to have significant poverty reducing impacts with micro dams, deep wells, river diversions, and ponds leading to 37%, 26%, 11%, and 9% reduction in poverty incidence compared to rainfed system. Finally, our study identified the most important correlates of poverty on the basis this we made the policy recommendations to build assets (AWM technologies, livestock, etc); to enhance human resource development and improve the functioning of labor markets for enhanced impact of AWM technologies on poverty.

This article evaluates the impacts of a controlled irrigation technique in rice production called alternate wetting and drying (AWD). Propensity score matching (PSM) and regression-based approaches applied to farm-level survey data are used to achieve the objective of the study. The PSM and regression-based approach accounts for the potential bias due to selection problems from observable variables. Results of the impact analysis using both empirical approaches indicate that AWD, particularly the ''Safe AWD'' variant, reduces the hours of irrigation use (by about 38%), without a statistically significant reduction in yields and profits. This reduction in irrigation time translates to corresponding savings in the amount of irrigation water and pumping energy used. However, further analysis of the impact estimates suggests that the potential magnitude of the selection bias based on unobservable variables may still be able to eliminate the measured impact from the PSM and regression-based techniques that only control for selection based on observable variables. Hence, the current impact results have to be interpreted with caution and further data collection is needed to construct a panel data that would allow one to account for selection problems due to unobservable variables and, consequently, better estimate the AWD impact.

Group homes fall into the broad category of residential care, a category that also includes half-way homes, campus based homes, emergency shelters, self-contained settings, and staff secured setting. In general, residential care services represent an option of last resort. In the current study we use administrative records from a large urban county and propensity score matching to investigate the relationship between group home placements in child welfare and the risk of delinquency (n = 8226). The results indicate that the relative risk of delinquency is approximately two and one half times greater for adolescents with at least one group home placement as compared with youth in foster care settings. This finding raises serious questions about the use of group homes for victims of physical abuse and neglect.

We examine the long term post acquisition performance of 1,385 firms over 1980-2006. Prior merger long run return studies show that acquiring firms experience a negative stock price drift after performing an acquisition. One of the methods used in this area of research involves matching a similar non-acquiring firm to the event firm to create a benchmark for expected returns. However, the matching methods used are often forced to employ only a small number of firm characteristics to perform the matching procedure. In this study, we use propensity score matching to solve this dimensionality problem. We find that acquiring firms do not underperform matching firms in the one year following an acquisition, while companies that complete two or more deals within one year from the first acquisition outperform their peers. Our analysis of the two-year acquirers' return performance suggests a significant underperformance of the acquirers vs. their peers, with firms conducting multiple acquisitions experiencing only slightly higher returns than their matching firms. In addition, our findings suggest that hostile takeovers and acquisitions by firms with higher leverage experience higher long term returns.

The Philippine government implemented the Pantawid Pamilyang Pilipino Program (4Ps) and patterned it after Latin American conditional cash transfer (CCT) programs, with the goals of poverty reduction and social development. With the insufficiency of research focusing on the 4Ps’ indirect effects, the program’s social impact remains unknown. Since the beneficiaries receive cash transfers from the government, they might be obliged to reciprocate their actions through civic engagement activities. In this study, civic engagement is defined as the overall activeness of citizens to work and positively impact their lives through participation in political and communal activities. Using propensity score matching, the researchers found that beneficiaries of the 4Ps tend to vote and participate in organizations more than the non-beneficiaries. However, this is accounted mostly to Luzon. Moreover, the study concluded that the 4Ps alone is not sufficient to improve civic engagement outcomes as they are also dependent on other factors which include, but are not limited to: geographical location, educational attainment and employment status, cultural diversity and uniqueness of organizations.

The purpose of this study was to identify determinants of motor pump adoption and its impact on Smallholders Farmers' income, in Southern Ethiopia. Primary data were collected from 140 sample respondents drawn from both motor pump adopting and non-adopting households. Secondary data were also collected from different sources. Descriptive statistics and econometric model were employed for data analysis. The logistic regression estimation of factors determining adoption of motor pump revealed that age of household head, gender, education level, availability of shallow ground water, number of adult family members, farm land size, participation in local organization and access to credit significantly influenced adoption of motor pump. Propensity score matching method was applied to analyze impact of motor pump adoption on outcome variable. The result revealed that motor pump adopting households had increased annual agricultural income on average 88047.49(Birr) per year compared to non-adopting households. Result showed that adoption of motor pump has significant positive impact on household annual agricultural income. Therefore, the study recommends establishing private manual well drilling enterprises at local and regional level to provide low cost access to ground water for pro-poor farmers that would expand the irrigated farm production area, and encouraging enterprises that combine the supply of standard pumps, technical support and spare parts to farmers and market for the produce would greatly improve the use of water lifting technologies (motor pump) and brings economic benefits to farmers. In addition, provision of well-functioning credit facility to improve adoption rate by female farmers to address gender imbalances.

Theory suggests that securitization provides financial institutions with an opportunity to lower the cost of funding; improve credit risk management and increase profitability. In practice, however, it might lead to adverse consequences through a number of indirect channels. Therefore, the net impact of securitization on bank performance is ambiguous. This study aims to evaluate whether banks improve their performance through the use of the securitization market by applying a propensity score matching approach. In other words, we build a counterfactual group of banks to assess what would have happened to the securitizing banks had they not securitized. Using US commercial banking data from 2001 to 2008, we first test these hypotheses using univariate analysis and find that securitizing banks are, on average, more profitable institutions, with higher credit risk exposure and higher cost of funding. However, the propensity score matching analysis does not provide evidence of significant causal effects of securitization on the performance of banks. Therefore, securitization does not seem to outperform alternative funding, risk management and profitability improvement techniques used by banks that have ex-ante similar characteristics to those securitizing. Our evidence raises important questions about the motives for banks' increasing securitization activities and consequent implications for the banking system.

This study examines whether the decision to voluntarily (i.e. without a statutory obligation) employ two audit firms to conduct a joint audit is related to audit quality. We use separate samples and empirical designs for public and privately held companies in Sweden, where a sufficient number of companies have a joint audit on a voluntary basis. Our empirical findings suggest that companies opting to employ joint audits have a higher degree of earnings conservatism, lower abnormal accruals, better credit ratings and lower perceived risk of becoming insolvent within the next year than other firms. These findings are robust to the use of a propensity score matching technique to control for the differences in client characteristics between firms that employ joint audits and those that use single Big 4 auditors (i.e. auditor self-selection). We also find evidence that the choice of a joint audit is associated with substantial increases in the fees paid by the client firm, suggesting a higher perceived level of quality. Collectively, our analyses support the view that voluntary joint audits are positively associated with audit quality in a relatively low litigious setting both for public and private firms.

Notices 1 Effective January 2007, the Discussion Paper series within each division and the Director General's Office of IFPRI were merged into one IFPRI-wide Discussion Paper series. The new series begins with number 00689, reflecting the prior publication of 688 discussion papers within the dispersed series. The earlier series are available on IFPRI's website at www.ifpri.org/pubs/otherpubs.htm#dp.

This study examines the adoption of improved maize germplasm in Oaxaca and Chiapas in Mexico. It employs a propensity score-matching approach to analyze the impact of the adoption of improved maize varieties on household income and poverty reduction, using cross-sectional data of 325 farmers from the two regions. The findings reveal a robust positive and significant impact of improved maize variety adoption on farm household welfare measured by per capita expenditure and poverty reduction. Specifically, the empirical results suggest that adoption of improved maize varieties helped raise household per capita expenditure by an average of 136-173 Mexican pesos, thereby reducing their probability of falling below the poverty line by roughly 19-31%.

We measure the relationships among income, poverty and charcoal production in three charcoal-producing districts of western Uganda. Using household survey data and propensity score matching techniques we find positive and statistically significant correlations between participation in charcoal-related activities and subsequent household income and poverty levels. Charcoal production is found to be especially important for households with low agricultural capacity and limited stocks of human and physical capital. However, in contrast to popular views and results from other studies, we find that those engaging in charcoal production are not necessarily the poorest cohorts in our sample. Our findings have implications for policies aimed at alleviating rural poverty and protecting forests. 2 In Uganda, approximately 95% of wood (excluding sawn wood) is consumed as fuel. In Kampala, consumption of charcoal increased 50% over the decade to 2004, to 1.3 million tons annually (Knopfle, 2004). Uganda's energy sector is characterized by a high degree of dependence on biomass energy. Access to electricity is limited to an estimated 9% of the population; the remaining 91% use biomass (mainly in the form of firewood or charcoal) for their energy needs (Ministry of Energy and Mineral Development (MEMD), 2007). At the beginning of this decade, annual consumption of wood was estimated to be about 25 million tons (or about 1.1 tons per capita), out of which 4 million tons was consumed as charcoal. The estimated contribution of firewood and charcoal to Uganda's GDP stood at Ush 120 billion and 67 billion, respectively; and charcoal consumption was postulated to increase by 6% per annum-the rate of urban population growth (Ministry of Energy and Mineral Development (MEMD), 2002). Total household expenditures on charcoal increased from 4.076 billion Ush in 1995/96 to 9.345 billion in 2005/06 (Ministry of Energy and Mineral Development (MEMD), 2007).

Contract farming is a form of vertical coordination largely aimed at correcting the market failure associated with spot markets that arise due to imperfect information. However the impact of contract farming on the welfare of smallholder farmers in Kenya is not well understood. While some authors have argued that contact farming improves access to ready markets by smallholder farmers, other studies have suggested that contract farming lowers the incomes of smallholder farmers because the contractors wield greater market power over the farmers. Consequently, it is seen as a blessing by some and a necessary evil by others. This study uses a propensity score matching technique to shed light on the impact of contract farming on smallholder farmers. The study also examines the conditioners of participation in contract farming. It uses data collected from 180 smallholder poultry farmers stratified by participation in contract production. The study finds that, on average, contracted farmers earned more net revenue per bird compared to the independent farmers, by approximately 27 percent, and as such participating in contract farming could improve the welfare of these small holder poultry farmers. This finding suggests that getting smallholder commercial poultry farmers to participate in contract farming can help improve their welfare through increasing the net revenues from these birds and thereof incomes.

MatchIt: Nonparametric Preprocessing for Parametric Casual Inference, 2005b. URL http://gking.harvard.edu/matchit. R package version – Ho, Imai, et al. ... Examining the impact of missing data on propensity score estimation in determining the effectiveness of self-...

This study analyzes the impact of remittances on poverty in Nigeria, using data from the 2004 Nigerian National Living Standard Survey (NNLSS). The paper used a multinomial logit model with instrumental variables and the propensity score matching (PSM) method to estimate the impact of remittances on poverty. The use of these methods was based on two reasons. The first is to control for the problems of selectivity and endogeneity. The second is the fact that the implicit hypothesis of estimating the expenditures of the counterfactual group, as done in some previous studies, is in similarity between the group that receives remittances (treated) and the other that does not (untreated). The study finds that both internal and international remittances reduce the incidence, depth and severity of poverty. The statistical tests show a significant Average Treatment Effect on the Treated (ATT), due to internal and external remittances. The receipt of internal remittances reduces the poverty headcount by 11.14% and poverty gap by 9.7% while the receipt of international remittances makes poverty indices almost nil.

We evaluated the impacts of the Ethiopian Productive Safety Net Program (PSNP) on rural households' holdings of livestock and forest assets including trees. Using panel data, we applied both regression analysis and propensity score matching. We found no indication that participation in PSNP induces households to disinvest in livestock or trees. In fact, households that participated in the program increased the number of trees planted, but there was no increase in their livestock holdings. We found no evidence that the PSNP protects livestock in times of shock. Shocks appear to lead households to disinvest in livestock, but not in trees. Our results suggest that there is increased forestry activity as a result of PSNP, and that improved credit access encourages households to increase their livestock holdings.

This paper examines the impact of being an innovator on firm productivity in Taiwan. Using a panel of 48794 firms observed over the 1997-2003 period, and distributed across 23 industries, we compute TFP by estimating Translog production functions with conventional K, L, E, M inputs. We evaluate the impact of being an innovator on TFP using propensity score matching. The rationale is that, over the period, innovating firms are likely to have benefited from one of many innovation policy measures known as IUS (until 1999) or "New IUS" (after 1999). The analysis was conducted separately for 1997-1999 and for 2000-2003. Our results show a significantly negative effect of being an innovator on TFP in every 2-digit industry and in both sub-periods. If TFP is defined as the part of productivity unexplained by conventional inputs, then this means that innovation activities explain a significant part of productivity at the firm level.

Decades of studies identify personality traits as prospectively associated with life outcomes. However, previous investigations of personality characteristic-outcome associations have not taken a principled approach to covariate use or other sampling strategies to ensure the robustness of personality-outcome associations. The result is that it is unclear (1) whether personality characteristics are associated with important outcomes after accounting for a range of background variables, (2) for whom and when personality-outcome associations hold, and 3) which background variables are most important to account for. The present study examines the robustness and boundary conditions of personality-outcome associations using prospective Big Five associations with 14 health, social, education/work, and societal outcomes across eight different person- and study-level moderators using individual participant data from 171,395 individuals across 10 longitudinal panel studies in a mega-analytic framework. Robustness and boundary conditions were systematically tested using two approaches: propensity score matching and specification curve analysis. Three findings emerged: First, personality characteristics remain robustly associated with later life outcomes. Second, the effects generalize, as there are few moderators of personality-outcome associations. Third, robustness was differential across covariate choice in nearly half of the tested models, with the inclusion or exclusion of some of these flipping the direction of association. In sum, personality characteristics are robustly associated with later life outcomes with few moderated associations. However, researchers still need to be careful in their choices of covariates. We discuss how these findings can inform studies of personality-outcome associations, as well as recommendations for covariate inclusion.

En el presente trabajo se evalúa el valor agregado de la universidad pública respecto al de las universidades privadas en razonamiento cuantitativo y lectura crítica. Para ello, se utiliza la metodología de Propensity Score Matching, la cual permite evaluar el impacto que tiene en los estudiantes estudiar en una universidad pública respecto a una privada. Dadas las diferencias en las universidades colombianas, se efectúa el ejercicio de diferencias en diferencias con distintas clasificaciones de universidades y diversos programas académicos. Los resultados obtenidos muestran que el valor agregado por las universidades públicas es heterogéneo y está en función del enfoque de la universidad. Puntualmente, las universidades públicas con enfoque doctoral generan un mayor valor agregado con respecto a las privadas, sin embargo, en otras categorías el valor agregado por las universidades públicas no es significativo.

When evaluating programme impact in a context where a randomised control trial is
either infeasible or not appropriate, the quasi-experimental approach of Propensity
Score Matching (PSM) is often used to construct a counterfactual. However, if there
are imbalances remaining after PSM, selection bias may persist. Increasingly,
researchers combine PSM and Difference-in-Differences (DID) to counter such
imbalances. While there is guidance on applying this combined approach using panel
data, applications of this approach in repeated cross-section settings are less frequent.
In this paper, we present an innovative approach to combining PSM and DID when only
cross-sections of data are available. We illustrate the methodology in the evaluation of
EQUIP-T, a UK Department for International Development-funded education
intervention in Tanzania. EQUIP-T is a four-year programme focused on improving
teacher performance, school leadership, and community participation, aiming to
increase the quality of primary education and improve pupil learning outcomes. This
study is likely to represent the first practical application of this PSM with DID procedure
for a repeated cross-section in an education evaluation. This paper will review the
implementation of the methodology in the context of the EQUIP-T programme. It will
also discuss strengths, appropriate contexts, and caveats to the approach, considering
unobservable characteristics, time-variant imbalances, implementation of concurrent
programmes, and challenges in calculating standard errors. In the first approach, the
Average Treatment Effect on the Treated (ATT) was compared across time, between
baseline and midline. In the second, PSM was used to match treatment units (pupils
and teachers in EQUIP-T schools) over time to construct a pseudo panel from repeated
cross-sections to estimate overall ATT. In the absence of panel data, the conventional
PSM approach of matching individuals at baseline and then calculating impact at
endline is not possible. The innovative pseudo panel approach addresses this,
following a suggestion by Blundell and Costa Dias (2000, p. 451). Impact estimates on
pupil tests are presented. Pupils’ test results were classified into one of five
curriculum-linked performance bands in Swahili and in Mathematics. The PSM-DID
analysis finds strong evidence that EQUIP-T has reduced the proportion of pupils in the
bottom performance band for Swahili in programme schools. These results remain
strong and highly significant across both our PSM with DID strategies.

We estimate the e ects of college quality using propensity score matching methods and the National Longitudinal Survey of Youth 1979 cohort. Matching allows us to relax the linear functional form assumption implicit in regression-based estimates. We also examine the support problem by determining whether there are individuals attending low-quality colleges similar to those attending high-quality colleges, and ÿnd that the support condition holds only weakly. Thus, the linear functional form plays an important role in regression-based estimates (and matching estimates have large standard errors). Point estimates from regression and matching are similar for men but not women.

This article examines the impact of marketing cooperatives on smallholder commercialization of cereals using detailed household data in rural Ethiopia. We use the strong government role in promoting the establishment of cooperatives to justify the use of propensity score matching to compare households that are cooperative members to similar households in comparable areas without cooperatives. The analysis reveals that although cooperatives obtain higher prices for their members, they are not associated with a significant increase in the overall share of cereal production sold commercially by their members. However, these average results hide considerable heterogeneity across households. In particular, we find that smaller farmers tend to reduce their marketed output as a result of higher prices, whereas the opposite is true for larger farmers.