results (F(1,118) = 5.47, p .05) indicated that the model as a whole is significant and that EFL learners’ expository text comprehension across three proficiency levels can be significantly predicted by their use of reading strategies. Accordingly, the ninth null hypothesis of the study was rejected.

In order to test the tenth research hypothesis of the study in finding whether beginner EFL learners’ use of reading strategies can be a significant predictor of their comprehension of expository text, another multiple regression analysis was performed.

Table 4.20 provides the details.

Table 4.20 Model Summary

Model Summary

Model

R

R Square

Adjusted R Square

Std. Error of the Estimate

1

.221a

.488

.493

.4814

a. Predictors: (Constant), Use of Reading Strategies

As shown in Table 4.20, in this case a value of R= 0.22 indicates a rather low level of prediction. The R2 value indicates that the learners’ use of reading strategies explains 48% of the variability of EFL beginner expository text comprehension.

In order to determine whether the provided model is a good fit for the data, a one-way ANOVA was performed. The results are shown in Table 4.21.

Table 4.21 ANOVA of regression model

ANOVAa

Model

Sum of Squares

df

Mean Square

F

Sig.

1

Regression

3.005

1

3.005

8.87

.021b

Residual

9.714

42

.231

Total

12.718

43

a. Dependent Variable: Expository text (Beginner)

b. Predictors: (Constant), SILL – Beginner

The F value in the Table 4.21 showed that (F(1,42) = 8.87, p .05). as can be seen p value is lower than assumed level of significance therefore, beginner EFL learners’ use of reading strategies can be considered as a significant predictor of their comprehension of expository text (i.e., the regression model is a suitable for the data). Therefore, the tenth null hypothesis of the study was rejected.

In order to test the 11th research hypothesis of the study in finding whether intermediate EFL learners’ use of reading strategies a significant predictor of their comprehension of expository text, another multiple regression analysis was performed.

Table 4.22 provides the extent to which variability in the dependent variable is accounted for by the independent variable.

Table 4.22 Model Summary

Model Summary

Model

R

R Square

Adjusted R Square

Std. Error of the Estimate

1

.651a

.423

.451

.540

a. Predictors: (Constant), SILL

As shown in Table 4.16, the coefficient of multiple correlations is presented in the “R” column. R is the measure of the prediction of the dependent variable. A value of 0.65 indicates a good level of prediction. The R2 value indicates that intermediate EFL learners’ use of reading strategies explains 42% of the variability of the expository text comprehension.

In order to determine whether the provided model is a good fit for the data, a one-way ANOVA was performed. The results are shown in Table 4.23.

Table 4.23 ANOVA of regression model

ANOVAa

Model

Sum of Squares

df

Mean Square

F

Sig.

1

Regression

9.009

1

9.009

8.03

.008b

Residual

9.714

49

.111

Total

18.723

50

a. Dependent Variable: Expository text (intermediate)

b. Predictors: (Constant), SILL – intermediate

The Table 4.23 showed that (F(1,49) = 8.03, p .05). As can be seen, p value is lower than the assumed level of significance; therefore, intermediate EFL learners’ use of reading strategies is a significant predictor of their comprehension of expository text i.e., the regression model is a suitable for the data. Therefore, the eleventh null hypothesis of the study was also rejected.

In order to test the 12th research hypothesis of the study in finding whether advanced EFL learners’ use of reading strategies is a significant predictor of their comprehension of expository text, a multiple regression analysis was performed.

Table 4.24 provides the details.

Table 4.24 Model Summary

Model Summary

Model

R

R Square

Adjusted R Square

Std. Error of the Estimate

1

.577a

.332

.353

.540

a. Predictors: (Constant), SILL

As shown in Table 4.24, the coefficient of multiple correlations R is the measure of the prediction of the dependent variable; in this case, comprehension of expository texts by advance learners. A value of 0.577 indicates a good level of prediction. The R2 value indicates that 33% of the variability of the expository text comprehension of advanced EFL learners’ is explained by the use of reading strategies.

A one-way ANOVA was performed in order to determine whether the provided model is a good fit for the data. The results are shown in Table 4.25.

Table 4.25 ANOVA of regression model

ANOVAa

Model

Sum of Squares

df

Mean Square

F

Sig.

1

Regression

4.05

1

4.05

6.70

.000b

Residual

6.703

23

.111

Total

10.708

24

a. Dependent Variable: Expository text (advanced)

b. Predictors: (Constant), SILL – advanced

The F value in the Table 4.25 showed that (F(1,23) = 6.70, p .05) showing that the advanced EFL learners’ use of reading strategies is indeed a significant predictor of their comprehension of expository text and that the regression model is a suitable for the data. Therefore, the twelfth null hypothesis of the study was rejected.

In order to test the 13th research hypothesis of the study in finding whether EFL learners’ use of reading strategies is a significant predictor of their comprehension of argumentative text across different proficiency levels, a multiple regression analysis was performed. Table 4.26 provides the extent to which variability in the dependent variable variables (argumentative text comprehension across three proficiency levels) is accounted for by the independent variable (reading strategies use).

Table 4.26 Model Summary

Model Summary

Model

R

R Square

Adjusted R Square

Std. Error of the Estimate

1

.275a

.756

.736

.7354

a. Predictors: (Constant), SILL

As shown in Table 4.26, the coefficient of multiple correlations presented in the “R” column is the measure of the prediction of the dependent variable; a value of 0.27 indicates a rather good level of prediction. The R2 value, the proportion of variance in the predicted variable that can be explained by the predictor indicates that EFL learners’ use of reading strategies can explain 75% of the variability of the argumentative text comprehension across different levels of proficiency.

In order to determine whether the provided is a good fit for the data, a one-way ANOVA was performed. The results are shown in Table 4.27.

Table 4.27 ANOVA of regression model

ANOVAa

Model

1

Sum of Squares

df

Mean Square

F

Sig.

Regression

2.669

1

2.669

4.934

.028b

Residual

63.816

118

.541

a. Dependent Variable: Argumentative text

b. Predictors: (Constant), SILL

The F value in the Table 4.27 shows the fitness of overall regression model for the data. The results showed that (F(1,118) = 4.93, p .05). As show in the Table above, p value is lower than the level of significance therefore, the EFL learners’ argumentative text comprehension across three proficiency levels can be significantly predicted by their use of reading strategies. Therefore, the thirteenth null hypothesis of

the study was not accepted.

In order to test the 14th research hypothesis of the study in finding whether beginner EFL learners’ use of reading strategies is a significant predictor of their comprehension of argumentative text, another multiple regression analysis was performed.

Table 4.28 provides the extent to which the independent variable (beginner EFL learners’ use of reading strategies) can be a predictor of the dependant variable (comprehension of argumentative text).

Table 4.28 Model Summary

Model Summary

Model

R

R Square

Adjusted R Square

Std. Error of the Estimate

1

.758a

.575

.562

.3976

a. Predictors: (Constant), SILL

As shown in Table 4.28, the coefficient of multiple correlations presented in the “R” column as the measure of the prediction of the dependent variable has a value of 0.75 indicating a rather low level of prediction. The R2 value indicates that 57% argumentative text comprehension is explained by EFL beginner learners’ use of reading strategies.

In order to determine whether the provided model (argumentative text comprehension of beginners as independent and reading strategy use as a dependent variable) is a good fit for the data, a one-way ANOVA was performed. The results are shown in Table 4.29.

Table 4.29 ANOVA of regression model

ANOVAa

Model

Sum of Squares

df

Mean Square

F

Sig.

1

Regression

7.842

1

3.921

6.845

.000a

Residual

4.858

42

.921

Total

11.700

43

a. Dependent Variable: argumentative text (Beginner)

b. Predictors: (Constant), SILL – Beginner

The F value in the Table 4.29 showed that (F(1,42) = 6.84, p .05) confirming that beginner EFL learners’ use of reading strategies can be a significant predictor of their comprehension of argumentative text (i.e., the regression model is a suitable for the data). Therefore, the fourteenth null hypothesis of the study was rejected.

In order to test the 15th null hypothesis of the study in finding whether intermediate EFL learners’ use of reading strategies is a significant predictor of their comprehension of argumentative text, another multiple regression analysis was conducted.

Table 4.30 provides the results.

Table 4.30 Model Summary

Model Summary

Model

R

R Square

Adjusted R Square

Std. Error of the Estimate

1

.412a

.169

.171

.2864

a. Predictors: (Constant), SILL

As shown in Table 4.24, the coefficient of multiple correlations as the measure of the prediction of the dependent variable equals 0.41and indicates a good level of prediction. The R2 value indicates that intermediate EFL learners’ use of reading strategies explains