Probit to percentage coversion

60 views
Skip to first unread message

Javaid Iqbal

unread,
May 24, 2025, 2:15:43 PM5/24/25
to lavaan
Dear Lavaan Team,
I am fairly new to lavaan. As I understand, lavaan computes the probit coefficients for direct, indirect and total effects. I wondered if it is possible to convert a model coefficient into percentage, which is easier to interpret than the probit coefficient. For example, let's suppose the coefficients of the direct, indirect and total effects are -0.04333667, 0.02129660 and  -0.02204007, respectively, can we convert these into percentages?
Thanks in advance for your input,
Javaid

Stas Kolenikov

unread,
May 24, 2025, 4:44:44 PM5/24/25
to lav...@googlegroups.com
What you are looking for is called “marginal effect”. Since probit is nonlinear, percents will be different depending on whether you start at 5% or 50%.


-- Stas Kolenikov, PhD, PStat (ASA, SSC) 
-- Principal Statistician, NORC @NORCnews
-- Opinions stated in this email are mine only, and do not reflect the position of my employer
-- Social media: @StatStas [ Twitter | mastodon.online ]
-- http://stas.kolenikov.name



--
You received this message because you are subscribed to the Google Groups "lavaan" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lavaan+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/lavaan/23fc8735-b948-4f51-af55-e0b8ea67798fn%40googlegroups.com.
Message has been deleted

Javaid Iqbal

unread,
May 30, 2025, 10:07:01 AM5/30/25
to lavaan
Thank you so much, Stas!

Yes, I did end up estimating the marginal effects as an additional analysis.

Best regards,
Javaid
Reply all
Reply to author
Forward
0 new messages