Investigating the Invariance of the Factor Structure and the Items of the National Math Test of the Mathematics and Technology and Experimental Group 2016-2017

The aim of this study was to investigate the invariance of the factor structure of the national test and the item parameters (factor and threshold) in the provinces. The research method is descriptive correlation. In order to investigate the purpose of the research, a sample of 1000 participants in the math test of the Mathematics and Technology experimental group of 1396 was selected from each province. The results showed that in most items, the performance of the examinees in Tehran, Isfahan, Khorasan Razavi, Fars, Mazandaran, Yazd and Alborz is better than most provinces. The number of non-invarince items in Tehran and Ilam is higher than other provinces. The alignment optimization method showed that 37.5% of the items in the first factor and 16% in the second factor for all provinces had approximate invariance of thresholds and 83% of the items in the first factor and 71% in the second factor had invariance in factor loadings. The number of non-invariance items in the second factor is higher than the first factor, so it is better to use the first factor items to compare the examinees in the provinces. Investigating DIF by using the alignment method alone is not enough in a large number of groups. If a specific group is involved in the study, it is necessary to use two-group DIF methods.