search

paul-buerkner / thurstonianIRT

Fit Thurstonian IRT models in R using Stan, lavaan, or Mplus
GNU General Public License v3.0
28 stars 12 forks source link

Further explanation fit_lavaan parameters extraction #37

Open nadimalfana opened 1 year ago

nadimalfana commented 1 year ago

I run fit_lavaan for estimating model and parameters on the sample "triplet" data on R.

I need to obtain the intercept/slope irt parameterization "... dividing threshold and the factor loading of each pairwise comparison {i,k}, by the square root of its error variance" (Brown & Maydeu Olivares, 2018 in Irwing (2018) p.548).

But there are threshold differences between the "i1i2" and "i1i3" pairs, and every pairs in the block that correspond to the same utility/traits. The result is somehow confusing because in the same book (Brown & Maydeu Olivares, 2018) produce the same thresholds between pairs.

Hope you can help with further detailed explanations or instructions to address this issue. Big thanks paul!

paul-buerkner commented 1 year ago

thurstonianIRT estimates different thresholds per item pair as is also done in brown I think. I am sorry I am not fully understanding your question.

nadimalfana @.***> schrieb am So., 11. Juni 2023, 08:08:

I run fit_lavaan for estimating model and parameters on the sample "triplet" data on R.

I need to obtain the intercept/slope irt parameterization "... dividing threshold and the factor loading of each pairwise comparison {i,k}, by the square root of its error variance" (Brown & Maydeu Olivares, 2018 p.548).

But there are threshold differences between the "i1i2" and "i1i3" pairs, and every pairs in the block that correspond to the same utility/traits. The result is somehow confusing because in the same book (Brown & Maydeu Olivares, 2018) produce the same thresholds between pairs.

Hope you can help with further detailed explanations or instructions to address this issue. Big thanks paul!

— Reply to this email directly, view it on GitHub https://github.com/paul-buerkner/thurstonianIRT/issues/37, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADCW2AA2YF53L3P46HTSOXTXKVHGBANCNFSM6AAAAAAZCDH7QY . You are receiving this because you are subscribed to this thread.Message ID: @.***>

nadimalfana commented 1 year ago

Thanks for your concern in my question. Let me change my question, could you explain how to extract parameters from the fit_lavaan to the parameters included in Brown & Maydeu Olivares (2018) and also the intercept/slope in Brown & Maydeu Olivares (2011)?

Table on Brown & Maydeu Olivares (2018): image

Table on Brown & Maydeu Olivares (2011): image