Please fill out this short user survey of only 3 questions in order to help us improve the site. We appreciate your feedback!
Double preference voting conditioned on gender
Last registered on June 05, 2018


Trial Information
General Information
Double preference voting conditioned on gender
Initial registration date
June 03, 2018
Last updated
June 05, 2018 2:03 PM EDT
Primary Investigator
Other Primary Investigator(s)
PI Affiliation
Aalto University
Additional Trial Information
In development
Start date
End date
Secondary IDs
The purpose of this study is to investigate the impact of a double preference voting system conditioned on gender on the voting behavior of voters. The primary outcome of interest is the share of votes received by female candidates in three different scenarios: (i) when participants can cast only one vote, (ii) when participants have the option to vote for two candidates without constraints on gender, and (iii) when participants can cast one vote for a male candidate and one vote for a female candidate.
External Link(s)
Registration Citation
Bagues, Manuel and Carolina Kansikas. 2018. "Double preference voting conditioned on gender." AEA RCT Registry. June 05. https://doi.org/10.1257/rct.3039-1.0.
Former Citation
Bagues, Manuel, Manuel Bagues and Carolina Kansikas. 2018. "Double preference voting conditioned on gender." AEA RCT Registry. June 05. http://www.socialscienceregistry.org/trials/3039/history/30349.
Experimental Details
Intervention Start Date
Intervention End Date
Primary Outcomes
Primary Outcomes (end points)
share of votes received by male and female candidates
Primary Outcomes (explanation)
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
Intervention dates
4 June 2018 - 9 June 2018
Experimental Design Details
Intervention dates 4 June 2018 - 9 June 2018 Description The investigation will be conducted through a field experiment with potential participants to the local elections of 10th June, 2018 in Italy. The methodology used to conduct the experiment is the computer assisted web interviewing system (CAWI), where the panelists are required to answer a set of 4 questions, in which the question on voting behavior varies depending on whether the individual panelist is assigned to the treatment or the control group. The questions the panelists will be asked are the following: 1. (Qualifying question on municipality) - the questionnaire will only be available for completion if the participant is a resident of one of the municipalities for which we have been able to collect detailed candidate and list information; 2. Which is your preferred list for the local elections of 10th June 2018? 3a. Now, assuming you can vote for two candidates, independently of their gender, who would you vote for? (question 3 for the control group) 3b. Now, assuming you can vote for two candidates, one female and one male, who would you vote for? (question 3 for the treatment group) 4. Please tick the candidates you know personally, of the ones running for election in the municipal council. 5. If you have a partner, is your partner working in a remunerated employment outside the household? In the questionnaire, participants are asked to tick their preference for a candidate from a list of candidates for the position of municipal council. To proceed in answering the questionnaire, they are obliged to select an option (or the relevant amount of options, in case they are asked for the two votes). Questions are arranged in decreasing order of importance. The survey is completed by an italian company specialized in market research. The complete questionnaire related to the survey based on the questions above (in italian) will be made public as soon as the field experiment is completed. Panel participants have a monetary incentive to complete the questionnaire, and they are rewarded based on the number of questionnaires they successfully complete. In addition to their age and gender, participants to the experiment are also asked: (a) Whether they are personally acquainted with the candidate; and (b) Given they have a partner, whether their partner works in a remunerated employment (outside the household). These characteristics could influence the answer of the participants, and therefore, a characterization based on these variables will be added to the paper as a descriptive complement. The first factor influences the choice of candidates independent of gender, while the second factor might influence the preference for diversity within the candidates chosen. Recent evidence on the impact of personal interactions with candidates suggest that personal interaction with candidates affects the support received by individual candidates, although modestly (Cantoni & Pons, 2016). In addition, labour market participation of women may also affect voting behavior (Iversen & Rosenbluth, 2008) and is therefore included in the questionnaire. The prior is that personal knowledge of the candidate will shift support towards this candidate, and that and labor force participation of the spouse will increase preference for diversity. Finally, another variable taken into account for the interpretation of results will be the difference between votes cast by participants in small municipalities (population <5000) compared to participants in large municipalities. The reason for this classification is participants from larger municipalities might already have framed the voting decision in terms of two candidates of a different gender, while participants in smaller municipalities, where the double preference voting conditioned on gender is not in place, might not have framed the voting decision in these terms. Cantoni, E., & Pons, V. (2016). Do Interactions with Candidates Increase Voter Support and Participation?: Experimental Evidence from Italy (pp. 1-56). Harvard Business School. Iversen, T., & Rosenbluth, F. (2008). Work and power: The connection between female labor force participation and female political representation. Annu. Rev. Polit. Sci., 11, 479-495.
Randomization Method
The randomization is performed by assigning a number to each of the panelists involved in the trial in the order they appear in the database of the company. Based on the assigned numbers, two groups are formed: the first, composed of participants assigned with an even number, and the second collecting participants who received an odd number. The even group will receive one version of the questionnaire, while the odd group will receive the other version. Randomization is performed by the company giving the questionnaire to the panelists based on the above instructions.
Randomization Unit
Individual level
Was the treatment clustered?
Experiment Characteristics
Sample size: planned number of clusters
Expected size: 2500 participants
Sample size: planned number of observations
Expected size: 2500 participants
Sample size (or number of clusters) by treatment arms
1,250 in the treatment group and 1,250 in the control group
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
5 p.p.
IRB Name
IRB Approval Date
IRB Approval Number
Post Trial Information
Study Withdrawal
Is the intervention completed?
Is data collection complete?
Data Publication
Data Publication
Is public data available?
Program Files
Program Files
Reports, Papers & Other Materials
Relevant Paper(s)