Endogeneity is often regarded as a key barrier in establishing the causal relationship between the third sector and its societal impact in empirical research. Through a systematic literature review of the quantitative studies on the third sector’s impact in the last two decades, we find that most quantitative studies of the third sector’s societal impact are published in journals outside main third sector journals. We also offer specific examples of how recent methodological advancements in addressing endogeneity help third sector researchers better solve this problem. Based on the analysis of this literature review, we recommend that third sector scholars should (1) refocus on the big question of the third sector’s impact on society, (2) catch up with the methodological advancement in addressing endogeneity, (3) be creative and transparent about addressing endogeneity, and (4) build better theories to link the third sector to broad societal outcomes.

We set out to test whether the effect of promises on trustworthiness derives from the fact that they are made (internal consistency) or that they are received (social obligation). The results of an experimental trust game appeared at first to support the former mechanism. Even when trustee messages are not delivered to trustors, trustees who make a promise are more likely to act trustworthy than those who do not make a promise. However, we subsequently ran a control treatment with restricted (non-promise) communication to examine whether the correlation between promises and trustworthiness is causal. The results show that the absence of promises does not decrease average cooperation rates. This indicates that promises do not induce trustworthiness, they are just more likely to be sent by cooperators than by non-cooperators.

The inverse probability of treatment weighted (IPTW) estimator can be used to make causal inferences under two assumptions: (1) no unobserved confounders (ignorability) and (2) positive probability of treatment and of control at every level of the confounders (positivity), but is vulnerable to bias if by chance, the proportion of the sample assigned to treatment, or proportion of control, is zero at certain levels of the confounders. We propose to deal with this sampling zero problem, also known as practical violation of the positivity assumption, in a setting where the observed confounder is cluster identity, i.e., treatment assignment is ignorable within clusters. Specifically, based on a random coefficient model assumed for the potential outcome, we augment the IPTW estimating function with the estimated potential outcomes of treatment (or of control) for clusters that have no observation of treatment (or control). If the cluster-specific potential outcomes are estimated correctly, the augmented estimating function can be shown to converge in expectation to zero and therefore yield consistent causal estimates. The proposed method can be implemented in the existing software, and it performs well in simulated data as well as with real-world data from a teacher preparation evaluation study.

Value added is a common tool in educational research on effectiveness. It is often modeled as a (prediction of a) random effect in a specific hierarchical linear model. This paper shows that this modeling strategy is not valid when endogeneity is present. Endogeneity stems, for instance, from a correlation between the random effect in the hierarchical model and some of its covariates. This paper shows that this phenomenon is far from exceptional and can even be a generic problem when the covariates contain the prior score attainments, a typical situation in value added modeling. Starting from a general, model-free definition of value added, the paper derives an explicit expression of the value added in an endogeneous hierarchical linear Gaussian model. Inference on value added is proposed using an instrumental variable approach. The impact of endogeneity on the value added and the estimated value added is calculated accurately. This is also illustrated on a large data set of individual scores of about 200,000 students in Chile.

In psychometrics, the canonical use of conditional likelihoods is for the Rasch model in measurement. Whilst not disputing the utility of conditional likelihoods in measurement, we examine a broader class of problems in psychometrics that can be addressed via conditional likelihoods. Specifically, we consider cluster-level endogeneity where the standard assumption that observed explanatory variables are independent from latent variables is violated. Here, “cluster” refers to the entity characterized by latent variables or random effects, such as individuals in measurement models or schools in multilevel models and “unit” refers to the elementary entity such as an item in measurement. Cluster-level endogeneity problems can arise in a number of settings, including unobserved confounding of causal effects, measurement error, retrospective sampling, informative cluster sizes, missing data, and heteroskedasticity. Severely inconsistent estimation can result if these challenges are ignored.

This study addresses endogenous factors related to the strategic planning of corporate social responsibility (CSR). Our findings help explain the paradox: If better CSR always leads to better firm performance, why do so many companies either choose not to engage in CSR or act irresponsibly? Managers may make decisions regarding CSR based on the environment. Some companies may be better served through a proactive CSR strategy; however, others may be unable to achieve better performance through this strategy for a variety of endogenous causes. Our sample included 594 U.S. publicly traded companies with 2,019 firm-year observations. We empirically simulated scenarios where companies selected inappropriate CSR strategies and found that the companies were unable to achieve better firm performance if they did not select appropriate CSR strategies based on their internal and external environment. Practical and theoretical implications are discussed.

This paper shows how econometric identification can be improved in studies making use of crop insurance participation as either an independent or dependent variable. The paper provides the reader with a succinct overview of how crop insurance contracts are priced and how to use publicly available data to derive a novel composite crop insurance design parameter that emulates existing crop insurance rating parameters using a procedure that is based on current actuarial practices. The derived design parameter performs well at predicting historic crop insurance loss-cost ratios and satisfies the requirements for an instrumental variable for a variety of empirical applications related to crop insurance. Representative empirical examples are presented where it is shown that the proposed instrument has favorable two-staged least squares diagnostic tests and is effective at eliminating endogeneity bias.

How has public healthcare spending prepared countries for tackling the COVID-19 pandemic? Arguably, spending is the primary policy tool of governments for providing effective health. We argue that the effectiveness of spending for reducing COVID deaths is conditional on the existence of healthcare equity and lower political corruption because the health sector is particularly susceptible to political spending. Our results, obtained using ordinary least squares and two-stage least squares estimations, suggest that higher spending targeted at reducing inequitable access to health has reduced COVID deaths. Consistent with the findings of others, our results indirectly suggest that health spending is necessary, but not sufficient unless accompanied by good governance and equitable access. Equitable health systems ease the effects of COVID presumably because they allow states to reach and treat people more effectively. Spending aimed at increasing health system capacity by increasing access thus seems a sound strategy for fighting the spread of disease, ultimately benefiting us all.