administered as a means for homogenizing the participants was first given to 30 participants with almost the same characteristics of the target group for the piloting phase. Table 4.1 provides the descriptive statistics of piloting the KET.

Table 4.1: Descriptive Statistics for KET Proficiency Test piloting

Descriptive Statistics

N

Minimum

Maximum

Mean

Std. Deviation

Variance

KET-Total

Valid N (listwise)

30

17.00

28.00

21.9667

3.04544

9.275

30

The researcher felt safe in employing the above piloted test for the subject selection process. Following the piloting, the KET was administered to90 students with the aim of selecting 60 of them for the study. The descriptive statistics of this process are presented below in Table 4.2.

Table 4.2 Descriptive Statistics for KET Proficiency Test

Descriptive Statistics

N

Minimum

Maximum

Mean

Std. Deviation

Score

90

27.000000

53.000000

44.11111111

4.911450607

Valid N (listwise)

90

Item analysis was run on this test and the results showed that none of the items needed to be revised or changed. Hence, the reliability was calculated. Table 4.2 reports the reliability estimate of the piloting (an acceptable Cronbach’s Alpha Index of .775.).

Table 4.3: Reliability of the KET Proficiency Test Piloting

Reliability Statistics

Cronbach’s Alpha

Cronbach’s Alpha Based on Standardized Items

N of Items

.775

.772

81

4.3. Descriptive Statistics of the KET Main Administration for Homogenization

The sections of the KET test were used to ensure the homogeneity of the two groups in terms of their proficiency level at the outset of the study.

Table 4.4: Descriptive Statistics for KET Main Administration for

Homogenization

Descriptive Statistics

N

Range

Minimum

Maximum

Mean

Std. Deviation

Variance

Skewness

Statistic

Statistic

Statistic

Statistic

Statistic

Statistic

Statistic

Statistic

Std. Error

KET control group

30

9

40

49

44.40

2.931

8.593

-.028

.427

KET experimental group

30

8.00

40.00

48.00

43.8667

2.43159

5.913

.151

.427

Valid N (listwise)

30

Table 4.5: The Results of Normality Check of the Distribution of scores on KET

KET

Skewness Ratio

Experimental group

-.028/.427= -0.06

Control group

.151/.427= .0353

Since the values of skewness/standard error of skewness in both groups were between the range of -1.96 and +1.96, the normality of the distribution of scores was guaranteed.

Figure 4.1: The Histogram of Scores of KET Main Administration

In order to make sure that the two groups were homogeneous regarding their English Language proficiency prior to the treatment an independent sample t-test was conducted between the KET mean scores of the experimental and control groups.

Table 4.6: Independent Sample T-test for Control and Experimental Groups’ KET scores

Independent Samples Test

Levene’s Test for Equality of Variances

t-test for Equality of Means

F

Sig.

T

Df

Sig.(2-tailed)

Mean Difference

Std. Error Difference

95% Confidence Interval of the Difference

Lower

Upper

KET- main administration

Equal variances assumed

1.844

.180

.767

58

.446

.533

.695

-.859

1.925

Equal variances not assumed

.767

56.085

.446

.533

.695

-.860

1.926

As depicted in the table, the value of Sig. (2-tailed) (.446) is larger than .05, leading to the fact that the two groups were homogeneous in their language proficiency prior to the treatment.

4.4. Descriptive Statistics of the grammar Pre-test

Among the 90 students who took the test, 60 whose scores fell between one standard deviation above and below the mean were chosen as the participants of this study to be placed in the experimental and control groups. Accordingly, 30 were put in one group and 30 in the other. The descriptive statistics of the two groups appear in the Table 4.6. Below:

Table 4.7: Descriptive Statistics for the Results of the Pre-test

Descriptive Statistics for the pre-test

N

Range

Minimum

Maximum

Sum

Mean

Std. Deviation

Variance

Statistic

Statistic

Statistic

Statistic

Statistic

Statistic

Std. Error

Statistic

Statistic

Control Group

30

77.50

12.50

90.00

1882.50

62.7500

2.78158

15.23537

232.116

Experimental Group

30

72.50

12.50

85.00

1804.00

60.1333

3.26034

17.85764

318.895

Figure 4.2: Histogram of the Scores Obtained on the Grammar Pre-test of the Control Group

Figure 4.3: Histogram of the Scores Obtained on the Grammar Pre-test of the Experimental Group

Table 4.8: Results of Normality of Distribution of Scores for Grammar Pre-test

Grouping

N

mean

Std. deviation

Std. error mean

Skewness

Kurtosis

statistic

Std. error

statistic

Std. error

Control

30

62.750

15.235

2.781

– 1.028

.578

2.739

.833

experimental

30

60.133

17.857

3.260

– .970

.578

.551

.833

According to the above table and based on the results of dividing statistics to standard error of skewness (-.1.028 / .578= – 1.77) which is a value in the acceptable range of -1.96 and +1.96, the researcher came to the conclusion that the scores on the test of those participants who were grouped as the control group were normally distributed.

The same process was practiced for the scores of experimental learners and the result (-..970 / .578 = – 1.67) was located in the same range and guaranteed the normality of distribution. Now that the researcher was sure about the normality of her sample scores, she was persevered to run the independent samples t-test in order to find out about their homogeneity

Table 4.9: Independent Samples T- Test for Pre-test

Independent Samples Test for the pre-test

Levene’s Test for Equality of Variances

t-test for Equality of Means

F

Sig.

t

Df

Sig. (2-tailed)

Mean Difference

Std. Error Difference

95% Confidence Interval of the Difference

Lower

Upper

Scores

Equal variances assumed

1.061

.307

.611

58

.544

2.61667

4.28568

5.96206

11.19539

Equal variances not assumed

.611

56.596

.544

2.61667

4.28568

5.96659

11.19992

The significance level of Levene’s test in the first line of the table-which refers to equal variances assumed-is larger than .05. In fact, as depicted from the table, the amount of p-value is .544.05. Therefore, the researcher was assured that there was no significant difference regarding grammar knowledge between the two groups prior to the treatment. So it can be claimed that the groups were almost homogeneous in terms of grammar ability with 95% confidence.

Figure 4.4: Histogram of the Scores Obtained on the Grammar Post- test of the Control Group

Figure 4.5: Histogram of the Scores Obtained on the Grammar Post- test of the Experimental Group

Table 4.10: descriptive statistics for the results of the post-test

Descriptive Statistics for the post-test

N

Range

Minimum

Maximum

Sum

Mean

Std. Deviation

Variance

Statistic

Statistic

Statistic

Statistic

Statistic

Statistic

Std. Error

Statistic

Statistic

Control Group

30

80.00

20.00

100.00

1771.00

59.0333

3.29236

18.03298

325.189

Experimental Group

30

55.00

45.00

100.00

2170.00

72.3333

3.02986

16.59525

275.402

Table 4.11: Results of Normality of Distribution of Scores for Grammar Post-test

Grouping

N

mean

Std. deviation

Std. error mean

Skewness

Kurtosis

statistic

Std. error

statistic

Std. error

Control

30

59.033

18.032

3.292

– .213

.427

.390

.833

experimental

30

72.333

16.595

3.029

– .034

.427

– .857

.833

According to the above table and based on the results of dividing statistics to standard error of skewness (-.213 / .427 =-0.498) which is a value in the acceptable range of -1.96 and +1.96, the researcher came to the conclusion that the scores on the test of those participants who were grouped as the control group were normally distributed.

The same process was practiced for the scores of experimental learners and the result (-.034 / .427 = – .079) was located in the same range and guaranteed the normality of distribution. Now that the researcher was sure about the normality of her sample scores, she was persevered to run the independent samples t-test in order to find out about their homogeneity.

4.5. The Results of Testing the Null Hypothesis

In order to check the null hypothesis of the study which stated that alternative teaching model does not have a significant effect on EFL learners’ grammar achievement the researcher needed to compare the mean scores of the experimental and control groups of the reading post-test. For this purpose an independent sample t-test was legitimately run, as the distributions of scores for both groups proved to be normal. The subsequent table illustrates the corresponding statistics for the control and experimental groups’ performance on grammar post-test.

Table 4.12: Independent Samples Test for Post-test

Independent Samples Test for the post-test

Levene’s Test for Equality of Variances

t-test for Equality of Means

F

Sig.

T

Df

Sig. (2-tailed)

Mean Difference

Std. Error Difference

95% Confidence Interval of the Difference

Lower

Upper

scores

Equal variances assumed

.005

.942

2.973

58

.004

-13.30000

4.47434

22.25636

4.34364

Equal variances not assumed

2.973

57.604

.004

-13.30000

4.47434

22.25767

4.34233

Considering the