Primary Outcomes (end points)
We have three primary outcomes of interest
1) Beliefs about the opinions of others in their neighborhood
* Using the same set of statements we presented in our main experiment, we will ask participants to guess how many people would agree to each statement if 30 residents of their neighborhood were randomly selected and polled. The statements we plan to present are as follows. We will first present a statement regarding the labor market in general: “In my opinion, the minimum wage for Saudis (SAR 3000) should be kept at its current level”. This will be followed by two statements regarding participation of women in the labor force (our primary outcomes): “In my opinion, women should be allowed to work outside of the home”; “In my opinion, a woman should have the right to work in semi-segregated environments”.
2) Wife's labor supply actions
* These questions will include whether their wife is currently employed, whether she was employed three months ago, whether she has interviewed for a job in the past three months, whether she has applied to a job in the past three months, and whether she has any interviews scheduled for the future. For each of these questions, we will also ask whether the job is at home or outside the home. If the job is outside the home, we will ask if the job is more or less than 30 hours a week. Finally, if she is currently employed or was employed three months ago, we will also ask for her occupation.
3) Sign-up for driving lessons
* In order to understand the effect of our initial information treatment on an outcome that is not necessarily strictly related to our job service sign-up intervention but is nonetheless relevant for FLFP and female empowerment in general, we will ask participants whether they would sign up their wife for driving lessons if given the opportunity.
It is worth noting that since this is a follow-up survey among the participants of the main experiment, subjects may have had social interactions and communication post-treatment. (This issue is particularly pertinent given that subjects were recruited from the same neighborhood with overlapping social connections.)
Thus, individuals in the treatment group may have shared information with those in the control group, effectively contaminating the experimental treatment. For this reason, if one cannot reject the null hypothesis of no differences in outcomes in the follow-up, it may either reflect a true null or non-trivial contamination. By contrast, if the null hypothesis is rejected it would imply contamination was limited.