Factor structure of the Test of English for Academic Purposes (TEAP®) test in relation to the TOEFL iBT® test

Language Testing in Asia

Table 6 Fit Indices for the Models for Item-parceling Data

Model	χ ²	df	CFI	TLI	RMSEA [CI]	WRMR	SRMR
4A: TEAP unitary	11.101*	2	.951	.854	.213 [.103, .343]	--	.033
4B: TEAP correlated	0.000	0	1.000	1.000	.000 [.000, .000]	--	.000
4C: TEAP receptive-productive	9.766*	1	.953	.718	.296 [.148, .477]	--	.030
5: TEAP unitary and TOEFL iBT unitary	37.692*	15	.963	.930	.123 [.074, .173]	--	.040

Note. df = degrees of freedom. CI = 90% confidence interval. *p < .05. Before constructing Models 4A, 4B, and 4C, we conducted item-level factor analysis to examine whether each skill was unidimensional, to judge the adequacy of aggregating the item-level score to form a total section score for each skill. For example, we conducted item-level factor analysis on 60 items for the Reading Section, to examine if they were unidimensional, and thereby warrant being combined into a single score. The results supported the unidimensionality of each skill (i.e., for reading, χ ² = 1803.004, df = 1710, p = .0578, CFI = .930, TLI = .928, RMSEA [CI] = .023 [.000, .034], and WRMR = .941)