Search

This project examines the impact of the number of outcome questions and the type of outcome questions on data quality in conjoint survey experiments.

Organizational Attributes, Organizational Affinity, and Collective Action

This project aims to identify how organizational attributes can signal congruence -- similarity in values --- and such congruence can impact their willingness to engage with an organization. It also contains a novel methodological approach to analyzing both of these questions. I argue that we can place organizations and individuals in the same latent values space. Individuals will be more likely to want to engage with organizations to which they are closer in that space. Individuals will generally prefer organizations that are closer to them in values. There are many different organizational traits that can cue shared values. In this project I focus on descriptive representation and localness: people will be more likely to think an organization shares their value if its membership reflects them in descriptive ways and if its roots are close to them geographically. I propose a novel two-step modeling approach that allows me to estimate an individual's proximity to a conjoint profile, and then test the effect of that proximity on other outcomes of interest. My model has two parts: an IRT-like component that helps estimate where organizations and individuals are in the same space, and a logistic regression that uses the results of the IRT model when estimating latent distances.

This registration is for a project looking at how organizations in Cambodia, Serbia, and Uganda react to government repression. It uses a conjoint survey experiment to study this question of interest.

See original pre-registration: https://osf.io/u87me/?view_only=f7e6ed533895429ebc1238c21bf1724a (link is anonymized)

In the first paper, I define the QualMix model, a mixture modeling approach to derive estimates of survey data quality in situations in which two sets of responses exist for all or certain subsets of respondents. I apply this model to the context of survey backchecks. Through simulation based on real-world data, I demonstrate that the model successfully identifies incorrect observations and recovers latent enumerator and survey quality. I further demonstrate the model's utility by applying it to data from a large survey in Malawi, using it to identify significant variation in data quality across observations generated by different enumerators. In the second paper, I investigate the how a match in values impacts individual decisions to engage with organization. I develop a new way to use conjoint survey experiments to study such questions. The proposed model has two parts: a component that helps estimate where organizations and individuals are in the same values space, and an outcome model that uses latent distances from the first part as inputs. I argue that individuals will be more likely to want to engage with organizations to which they are closer in a latent values space, and that they look at organizational traits for cues about an organization's values. I find that individuals are more likely to want to engage with organizations that are more descriptively representative of them. In addition, I find that congruence more consistently impacts considerations of benefits rather than costs. In the third paper, my co-authors and I discuss the impact that enumerators can have in survey and lab experiments. We conceptualize enumerators as treatment versions and clarify the implications of treatment versions for inference and for external validity. Researchers need to pay much more attention to the enumerators used in their studies and should assess the impact of enumerators on their results. We propose a hierarchical model that researchers can use to estimate enumerator treatment effects, which also allows them to incorporate information on enumerator characteristics. We present several recommendations for researchers doing experimental work with enumerators and illustrate them on a survey experiment carried out in Uganda.

What is Civil Society? Can we assess how strong it is? Using the problems present conceptualizations of civil society entail as a point of departure, this work develops a definition that strips civil society of its normative assumptions and functional form and fits better with the reality we observe. Civil society can be thought of as a space between the state, the market, and the family that can be divided into different sectors based on the goals of the civil organizations that inhabit it. The strength of each sector can be assessed by gauging how cohesive civil society organizations within that sector are, how embedded they are in the social fabric of society, and how developed their bureaucratic capital is. This work then sketches out how this approach can be used to analyze civil society in the United States and Armenia. In sum, it presents the basis for a new research agenda aims to investigate the relationship between civil society and democracy.