I wanted to question the €1000 per month for the internships? (Note I appreciate the forum poster isn’t responsible) To me this amount seems exploitative and I’d like to know the The School of Moral Ambition’s reasoning behind this.
These internships are 6-12 months long, based in Amsterdam—which is an expensive place to live. From the government’s own website, minimum wage for 21+ is €2,317.83 per month, and the €1000 offered will barely cover living costs.
I recognise that the organisation’s ambitions are good and that internships offering this amount are not infrequent in the Netherlands. However, I don’t think this provides an excuse.
-
This is not an internship at the IMF or an investment bank, where the role has a large value in future employment. These are roles in operations and event management.
-
This pay level encourages elitism within EA—most people who take this job will need support from family or reliance on savings. Those from a low income background are heavily penalised given the wage is much lower than minimum wage.
-
More importantly, even if this behaviour is acceptable in other industries, we should ask if this is how we, the EA community, want to treat our young talent. In my eyes, we can do a lot better.
Hi, thanks for putting together a serious attempt to measure the effect of EAGx conferences.
Whilst I am highly supportive of your attempts to do this, I wanted to ask about your methodology. You are attempting a difference in differences, with a control and treatment group, but immediately state: ’In the data from the initial collection period (‘before’), the control group and treatment groups showed several significant differences.’
To my knowledge this effectively means you no longer have a suitable control and treatment groups and makes a difference in differences ill-advised?
Very happy to be wrong here! But whilst you made some attempts to find a control and treatment group, your initial analysis of the two groups suggests you are comparing apples to oranges?
Control and treatment groups can have some differences, but critically we want the parallel trends assumption to hold. From your descriptive analysis of the two groups this is unlikely to be true. Would you agree that your control group and treatment group would have gone on to engage with EA in different ways if the conference had never happened? I’d say so, given that you demonstrate they are rather different groups of people!
If I’m right—I may not be—This still leaves room for analysing the effects on conference attendance on attendees behaviour, but just losing the control and using differing methodology. I hope I’m wrong—but I think this means that we can’t use this part of your results.