求助各位大神!
IRT理论的Graded Response Model中有一个描述Latent Trait Score的Extremity Parameters应该如何理解?看文献说的:
The extremity parameters show the latent trait score at which people have a 50/50 chance of selecting certain responses.
对于这个结果:
## Coefficients:
## Extrmt1 Extrmt2 Extrmt3 Extrmt4 Extrmt5 Dscrmn
## A1 -1.790 1.230 2.974 5.211 8.695 0.407
有如下解释:
The Extrmt1 is -1.790 suggesting that those with a latent trait score of -1.790 have a 50/50 chance of selecting a 1 for that item. The Extrmt2 is 1.230 and so those with a latent trait score of 1.230 have a 50/50 chance of selecting a 1 or a 2 for that item. Extrmt3 is the latent trait score at which people have a 50/50 chance of selecting a 1, a 2 or a 3. Extrmt4 is the latent trait score at which people have a 50/50 chance of selecting a 1, a 2, a 3 or a 4. Finally, Extrmt5 is the latent trait score at which people have a 50/50 chance of selecting a 1, 2, 3, 4 or 5.
主要是这里的潜在特质得分和Extremity参数该如何理解和应用,不清楚。
请大神指点!