Open nadimalfana opened 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: @.***>
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):
Table on Brown & Maydeu Olivares (2011):
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!