Experimental Design Details
The couple should be participating in the research at the same time. The research will be conducted through Qualtrics online survey and WeChat video chat. The couples should be using separate devices and located in separate spaces (for privacy).
Consent form: Participants will first the consent form before they participating in the research, which they will sign through Qualtrics.
Interview: An interview will be conducted to collect demographic information from the couple (age, place grow up, education, income, number of children, length of marriage).
Surveys (Household decision-making, RAS, CRT):
After the interview, the couple will both receive a survey containing:
- A question related to the way the couple make household financial decisions
- A seven-item "Relationship Assessment Scale" to measure marital assessment (Hendrick, 1988)
- A three-item "Cognitive Reflection Test" to measure cognitive ability (Frederick, 2005)
Main task (Individual decision-making):
- Instructions on portfolio allocation task (with examples and oral quiz with experimenter)
- Participants are then asked to make individual decisions in three risk allocation tasks (with three multipliers: P=1.5, P=2, P=2.5) . The tasks were designed by Gneezy and Potters (1997), and the three multipliers were also used by Ertac and Gurdal (2012) in their study.
Beliefs:
After both have finished the main task, individuals complete a survey that asks them to indicate how much they believe their partner has allocated in the previous three scenarios. This is incentivized.
Collaborative decision-making (verbal responses will be collected, and their discussion process will be video recorded):
After they have both finished the surveys, the couple will be brought together to make decisions on the same three scenarios together.
- In the INFO treatment, couples will be informed what their partner chose before they decide together. Only after they have made their decision in a particular scenario will we inform them of their choices in the next scenario.
- In the NOINFO treatment, couples will not be informed what their partner chose before they decide together.
Payment:
At the end of the experiment, only one scenario (one decision) will be randomly selected for payment by rolling a 6-sided die. The selected scenario (decision) will determine both the husband's and the wife's payoffs. That is, every decision a participant makes is made on behalf of him/her and his/her spouse.
For groups with NOINFO treatment, if individual scenario (choice) is selected for payment, there is a 25% chance that we will use the choice of a previous couple for payment to ensure that the selected individual's previous choices will not be revealed to the other person. We will ask participants in NOINFO to privately assign die rolls to outcomes at the end of the main task (individual).