|
Mplus vs Multilog in 2PL IRT
This is an example
result of IRT analysis using MPLUS and Multilog.
DATASET
1. MPLUS Syntax & Output
lsat-mplus.dat |
lsat-multilog.MLG
2. MULTILOG Syntax &
Output
lsat-multilog.dat |
lsatmplus.inp
MPLUS
MPLUS SYNTAX
TITLE: Analisis Teori Respons Butir MPLUS (GRM)
DATA: FILE IS lsat-mplus.dat;
VARIABLE: NAMES ARE a01-a05 freq;
FREQWEIGHT IS freq;
MISSING ARE ALL (-9);
USEVARIABLE ARE a01-a05;
CATEGORICAL ARE a01-a05;
ANALYSIS: ESTIMATOR IS ML;
LINK IS LOGIT;
MODEL:
F1 BY a01-a05*;
!F1 BY a01@1 a02-a05;
[a01$1-a05$1*]
[F1@0]; F1@1;
OUTPUT: STAND;
PLOT: TYPE IS PLOT1; ! Perintah Plot untuk deskriptif
TYPE IS PLOT2; ! Perintah Plot untuk kurva IRT
TYPE IS PLOT3; ! Perintah Plot untuk nilai theta
MPLUS OUTPUT.
MODEL RESULTS
Two-Tailed
Estimate S.E. Est./S.E. P-Value
F1 BY
A01 0.825 0.258 3.198 0.001
A02 0.722 0.186 3.872 0.000
A03 0.891 0.233 3.826 0.000
A04 0.688 0.185 3.718 0.000
A05 0.659 0.210 3.133 0.002
Means
F1 0.000 0.000 999.000 999.000
Thresholds
A01$1 -2.773 0.206 -13.482 0.000
A02$1 -0.990 0.090 -11.005 0.000
A03$1 -0.249 0.076 -3.266 0.001
A04$1 -1.285 0.099 -12.977 0.000
A05$1 -2.054 0.136 -15.148 0.000
Variances
F1 1.000 0.000 999.000 999.000
IRT PARAMETERIZATION IN TWO-PARAMETER LOGISTIC METRIC
WHERE THE LOGIT IS 1.7*DISCRIMINATION*(THETA - DIFFICULTY)
Item Discriminations
F1 BY
A01 0.486 0.152 3.198 0.001
A02 0.425 0.110 3.872 0.000
A03 0.524 0.137 3.826 0.000
A04 0.405 0.109 3.718 0.000
A05 0.387 0.124 3.133 0.002
Means
F1 0.000 0.000 0.000 1.000
Item Difficulties
A01$1 -3.360 0.867 -3.875 0.000
A02$1 -1.371 0.308 -4.455 0.000
A03$1 -0.280 0.100 -2.807 0.005
A04$1 -1.868 0.435 -4.296 0.000
A05$1 -3.119 0.868 -3.595 0.000
Variances
F1 1.000 0.000 0.000 1.000
MULTILOG
Syntax Analysis
MML PARAMETER ESTIMATION FOR THE 2PL MODEL, LSAT DATA
>PROBLEM RANDOM, PATTERN, NITEMS=5, NGROUPS=1, NPATTERNS=32,
DATA='lsat-multilog.DAT';
>TEST ALL, L2;
>SAVE;
>END;
2
01
11111
N
(4X,5A1,F4.0)
<
h4 align
= "left" >
Syntax Analysis
OUTPUT ANALYSIS
ITEM 1: 2 GRADED CATEGORIES
P(#) ESTIMATE (S.E.)
A 1 0.82 (0.18)
B( 1) 2 -3.36 (0.62)
@THETA: INFORMATION: (Theta values increase in steps of 0.2)
-3.0 - -1.6 0.166 0.161 0.154 0.146 0.136 0.126 0.115 0.104
-1.4 - 0.0 0.094 0.084 0.074 0.065 0.057 0.050 0.043 0.038
0.2 - 1.6 0.032 0.028 0.024 0.021 0.018 0.015 0.013 0.011
1.8 - 3.0 0.009 0.008 0.007 0.006 0.005 0.004 0.004
OBSERVED AND EXPECTED COUNTS/PROPORTIONS IN
CATEGORY(K): 1 2
OBS. FREQ. 76 924
OBS. PROP. 0.0760 0.9240
EXP. PROP. 0.0760 0.9240
ITEM 2: 2 GRADED CATEGORIES
P(#) ESTIMATE (S.E.)
A 3 0.72 (0.11)
B( 1) 4 -1.37 (0.21)
@THETA: INFORMATION: (Theta values increase in steps of 0.2)
-3.0 - -1.6 0.094 0.101 0.108 0.115 0.120 0.125 0.128 0.130
-1.4 - 0.0 0.131 0.131 0.129 0.126 0.122 0.116 0.110 0.104
0.2 - 1.6 0.097 0.089 0.082 0.075 0.068 0.061 0.055 0.049
1.8 - 3.0 0.044 0.039 0.034 0.030 0.027 0.023 0.020
OBSERVED AND EXPECTED COUNTS/PROPORTIONS IN
CATEGORY(K): 1 2
OBS. FREQ. 291 709
OBS. PROP. 0.2910 0.7090
EXP. PROP. 0.2910 0.7090
< /pre >
COMPARISON
< pre>
< /dt >
< pre>
< /dt >
< pre>
< /dt >
a=Item Discriminations
b=Item Difficulties
Wahyu Widhiarso | Fakultas Psikologi UGM
http://wahyupsy.blog.ugm.ac.id
|