Title Party Building and Politician Defection in Zambia
Post date 06/22/2016
C1 Background and Explanation of Rationale

We conduct a survey experiment of parliamentary candidates in Zambia. In the context of countries with nascent political party systems, the survey seeks to understand the factors influencing individual politicians in their choices over party affiliation and defection, leadership endorsements, campaign entry, campaign financing, and policy positions. Embedded in the survey is an experiment that uses randomized conjoint analysis. Drawing on hypotheses from the existing literature on political parties and African politics, we assess candidate preferences by asking respondents to choose their preferred party from among hypothetically generated pairs of parties.

C2 What are the hypotheses to be tested?

H1 (Party Structure): Candidates are more likely to prefer political parties that have a high level of party organization and institutionalization, with participatory mechanisms for the selection of the party leadership.
H2 (National Voter Support): Candidates are more likely to prefer political parties that have a high level of voter support at the national level.
H3a (Local Voter Support): Candidates are more likely to prefer political parties that have a high level of voter support in their own constituencies.
H3b (Local Elite Support): Candidates are more likely to prefer political parties that are supported by local power elites such as tribal chiefs, religious leaders, and headmen.
H4 (Candidate Selection): Candidates are more likely to prefer political parties that have bottom-up processes of candidate selection such as primary elections.
H5a (Incentives - Campaign Financing): Candidates are more likely to prefer political parties that offer support for campaign finances.
H5b (Incentives - Government Appointments): Candidates are more likely to prefer political parties that promise future cabinet and subcabinet appointments.
H5c (Incentives - Guaranteed Nominations): Candidates are more likely to prefer political parties that guarantee party nomination for the constituency for which they are seeking office.
H6 (Relationship with party leaders): Candidates are more likely to prefer political parties in which they share social ties with the party leader.
H7: Candidates are more likely to prefer political parties that currently control the executive branch.

C3 How will these hypotheses be tested? *

We use conjoint analysis to test hypotheses derived from the existing literature. We present respondents with two profiles of hypothetical parties generated from eight attributes, and we ask them to make a choice between them.

C4 Country Zambia
C5 Scale (# of Units) 125
C6 Was a power analysis conducted prior to data collection? No
C7 Has this research received Insitutional Review Board (IRB) or ethics committee approval? Yes
C8 IRB Number 2015-12-8222
C9 Date of IRB Approval 01/13/2016
C10 Will the intervention be implemented by the researcher or a third party? Researchers
C11 Did any of the research team receive remuneration from the implementing agency for taking part in this research? No
C12 If relevant, is there an advance agreement with the implementation group that all results can be published? No
C13 JEL Classification(s) not provided by authors