[Clyde Schechter]

Your questions here are only tangentially relevant to the initial topic of this thread. It would have been better to start a new thread so that others who might want to find this discussion later (because they have the same question you do) would be able to do so using the Forum's search facilities.

Be that as it may.

In the margins output it is written "Pr(swb==0)". According to Greene "Assuming that β is positive (for this x), Prob(y = 0|x) must decline. Alternatively, from the previous expression, it is obvious that the derivative of Prob(y = 0 | x) has the opposite sign from β."
Is that the crucial point?

Yes, precisely.

As for your second issue, I take it you are using an older version of Stata. In version 14, -margins- automatically provides the marginal effects for all outcome levels by default. The behavior you show was, I believe, the default in version 13. In version 13, you will need to write a loop to get all of them.

// RUN YOUR oprobit

forvalues i = 0/10 {
    margins, dydx(*) predict(outcome(`i'))
}

[Roman Mostazir]

Welcome Mischa to the Statalist. Please read the FAQ section to familiarize yourself with posting rules and ettiquettes which will help you to get useful replies from other members. The fourm rule suggests, it is not a good idea to dig an old thread started by someone. Please start a new thread with relevant title.

[Mischa Kleis]

Clyde Schechter: Thanks for the helpful answer! :-)
Roman Mostazir: I just skimmed the FAQ. I'm gonna do that.

[Maarten Buis]

You should not include age_sq. Instead you should have included c.age##c.age to include both age and age square in a way that Stata knows how to deal with.

[Mischa Kleis]

@Maarten Buis: Thank you for the advice!