|
Field
Abstract
|
Before
This pre-analysis plan outlines the design and proposed analysis of a survey experiment embedded in a multi-country study examining citizen discontent with political systems across seven Latin American countries. This large-scale study combines original survey data with national household surveys conducted by the National Statistics Offices in three countries, while in the other four countries it relies on our original survey with a new characterization module. The experiment described in this pre-analysis plan is embedded in both types of surveys—national household surveys in three countries and original surveys in the remaining four. It is designed to investigate the effects of two key factors: (i) security concerns and (ii) economic expectations and anxiety on different expressions of political discontent. Respondents are randomly assigned to receive one of three questionnaire versions, where the order of the modules varies. One version presents a module on perceptions of security first, followed by outcome measures. The second version presents questions on economic anxiety and expectations first, followed by outcomes. The third version asks outcome-related questions first, followed by the treatment modules. Key outcomes include general sentiments of discontent, such as anti-establishment attitudes, trust in institutions, and support for democratic principles, along with expressions of discontent through exit (e.g., opting out of public services) and voice (e.g., protests). The findings aim to shed light on the sources of citizen discontent and inform debates on how two core problems in Latin America—insecurity and economic volatility—shape public perceptions of and relationships with political systems.
|
After
This pre-analysis plan outlines the design and proposed analysis of a survey experiment embedded in a multi-country study examining citizen discontent with political systems across seven Latin American countries. This large-scale study combines original survey data with national household surveys conducted by the National Statistics Offices in three countries, while in the other four countries it relies on our original survey with a new characterization module. The experiment described in this pre-analysis plan is embedded in both types of surveys—national household surveys in three countries and original surveys in the remaining four. It is designed to investigate the effects of two key factors: (i) security concerns and (ii) economic expectations and anxiety on different expressions of political discontent. Respondents are randomly assigned to receive one of three questionnaire versions, where the order of the modules varies. One version presents a module on perceptions of security and victimization first, followed by outcome measures. The second version presents questions on economic anxiety and expectations first, followed by outcomes. The third version asks outcome-related questions first, followed by the treatment modules. Key outcomes include general sentiments of discontent, such as anti-establishment attitudes and support for democratic principles, along with expressions of discontent (e.g., support for protests). The findings aim to shed light on the sources of citizen discontent and inform debates on how two core problems in Latin America—insecurity and economic volatility—shape public perceptions of and relationships with political systems.
|
|
Field
Last Published
|
Before
September 09, 2025 11:42 AM
|
After
September 15, 2025 02:55 PM
|
|
Field
Power calculation: Minimum Detectable Effect Size for Main Outcomes
|
Before
We calculate the Minimum Detectable Effect (MDE) for each type of outcome variable. The MDE represents the smallest true effect size that can be detected with a specified level of power, given the study's sample size and design. We opted for this approach because we have already established an approximate fixed sample size of N=2,400 in each country, rather than determining the sample size that maximizes statistical power.
Our analysis encompasses 25 outcome variables, distributed as follows:
- 14 binary variables
- 5 ordinal variables (scale: 1 to 5)
- 5 ordinal variables (scale: 0 to 10)
- 1 count variable
To accommodate this diverse set of outcomes, we calculated the MDE using each of the 4 types of outcomes. Furthermore, we simulated various potential distributions for these four types of variables, exploring different combinations of means and standard deviations. This comprehensive approach allows us to assess the study's sensitivity to detect effects across a range of plausible scenarios.
Notably, our calculations indicate that our study design has sufficient statistical power to detect even relatively small effect sizes. Our MDEs resulting from the most conservative assumptions are: for binary outcomes is 0.04, for ordinal variables (1 to 5) is 0.12, for ordinal variables (1 to 10) is 0.28, and for count variables is 0.18.
|
After
We calculate the Minimum Detectable Effect (MDE) for each type of outcome variable. The MDE represents the smallest true effect size that can be detected with a specified level of power, given the study's sample size and design. We opted for this approach because we have already established an approximate fixed sample size of N=2,400 in each country, rather than determining the sample size that maximizes statistical power.
Our analysis encompasses 25 outcome variables, distributed as follows:
- 14 binary variables
- 5 ordinal variables (scale: 1 to 5)
- 5 ordinal variables (scale: 0 to 10)
- 1 count variable
To accommodate this diverse set of outcomes, we calculated the MDE using each of the 4 types of outcomes. Furthermore, we simulated various potential distributions for these four types of variables, exploring different combinations of means and standard deviations. This comprehensive approach allows us to assess the study's sensitivity to detect effects across a range of plausible scenarios.
Notably, our calculations indicate that our study design has sufficient statistical power to detect even relatively small effect sizes. Our MDEs resulting from the most conservative assumptions are: for binary outcomes is 0.02, for ordinal variables (1 to 5) is 0.06, for ordinal variables (0 to 10) is 0.12, and for count variables is 0.08.
|