• Why did the authors use multiple regression?

ANOVAa Model Sum of Squares df Mean Square F Sig. 1 Regression 433.066 2 216.533 282.996 <. residual .765 total a. dependent variable: t1 scale of science teacher perceptions principal support b. predictors: tl collective responsibility years has taught high school table means and standard deviations all measures phonological awareness ran word reading sentence text word- specic orthographic knowledge general rapid automatized naming atscore mean j. zaric t. nagler deviation baverage reaction time per correct item second the four subtests craw score intercorrelations between word-specific word-level sentence-level .73 text-level .48 phon. .37 .21 .11 orth. know p know.="orthographic" relationship speed knowledgein correlations at sentence- text- level on elfe are shown. since ortho- graphic violated normal distribution we report spearman rho is correlation coefficients. significantly correlated with sentence-level. there was no significant word-spe- cific speed. however phonologi- cal text-level. in addition were andword-specific knowledge. significantly. knowl- edge showed correlations. predictive patterns for multiple hierarchical regression analyses conducted to explore whether specific explain unique variance two subtypes independently each other separate models calculated. both entered simultaneously into model see step or amount explained over above although predictor variables residuals thus indicating homoscedasticity. basic show that signi- cant predictors performance together they performance. after entering word-specic remain signicant predictors. explains additional column vari- ance by>