Article Title Using Dictation to Improve Language Proficiency Author Mohammad Rahimi Shiraz University Biography: Dr. Mohammad Rahimi (PhD in TEFL) is currently a full-time assistant professor at the Department of Foreign Languages and Linguistics of Shiraz University, Iran. He teaches both graduate and undergraduate courses in EFL teaching, language testing, research methods, and writing. His areas of interest are research in testing and teaching writing, listening, and speaking skills. . |
Abstract
Dictation, though widely known as a testing device, can be considered a good learning technique to improve learners’ proficiency. This study investigates the effect of dictation with 65 EFL learners to see if it helps them improve their language proficiency. To this end, an experimental group (EG) and a control group (CG) were selected. They both took a proficiency test as a pre-test. All the conditions, especially teaching materials, were the same for both groups except that the EG had regular practice with dictation for one semester. The post-test was administered in a semester after the experiment had terminated. The results showed that both groups showed improvement in their performance on the whole proficiency test. However, with respect to the components of the test, the EG showed improvement in grammar, vocabulary, reading, and listening comprehension, while the control group showed improvement just in the vocabulary section of the test.
Introduction
Dictation has been widely known as a testing device for measuring language proficiency. However, using dictation as a language testing technique has suffered huge oppositions. One of the strongest claims made against dictation has been offered by Lado (1961). He considers dictation as having very little value in testing the language. He believes that since everything is dictated to the learner, dictation cannot measure any aspect of the language. Harris (1965), too, rejects dictation as an “uneconomical” and “imprecise” technique of language testing. Somaratne (1957) goes even further and considers it as just a test of spelling. Finally, Stanfield (1985) rejects using dictation as a test of foreign language proficiency due to its strong association with the Grammar Translation Method.
On the other hand, the proponents of dictation advocate it as a good device for measuring language proficiency, due to the high correlation it has shown with other integrative language proficiency tests such as cloze and TOEFL (Oller 1971 and Irvine et al., 1974). Besides, Oller and Streiff (1975) believe that as dictation triggers the internalized expectancy grammar of the learner, it can give us information about his overall language proficiency.
With respect to its teaching value, a few people believe that using dictation in class and practicing with it has nothing to do with the improvement of the learners' proficiency. For instance, Cartledge (1968) believes that dictation is not a teaching device. He claims that it can merely help learners have some practice in oral comprehension. Although Valette (1964) considers dictation as a method of both testing and learning, she believes the fact that learners are proficient in dictation does not necessarily mean that they are proficient in other aspect of language learning.
Nonetheless, research shows that dictation can help learners improve different aspects of their proficiency. For instance, Sawyer and Silver (1961) believe that dictation is primarily a learning exercise and only secondarily a testing device. They assert;
On the other hand, the proponents of dictation advocate it as a good device for measuring language proficiency, due to the high correlation it has shown with other integrative language proficiency tests such as cloze and TOEFL (Oller 1971 and Irvine et al., 1974). Besides, Oller and Streiff (1975) believe that as dictation triggers the internalized expectancy grammar of the learner, it can give us information about his overall language proficiency.
With respect to its teaching value, a few people believe that using dictation in class and practicing with it has nothing to do with the improvement of the learners' proficiency. For instance, Cartledge (1968) believes that dictation is not a teaching device. He claims that it can merely help learners have some practice in oral comprehension. Although Valette (1964) considers dictation as a method of both testing and learning, she believes the fact that learners are proficient in dictation does not necessarily mean that they are proficient in other aspect of language learning.
Nonetheless, research shows that dictation can help learners improve different aspects of their proficiency. For instance, Sawyer and Silver (1961) believe that dictation is primarily a learning exercise and only secondarily a testing device. They assert;
It is true that such a dramatic change does not mean that the student’s control of his new language is proportionately that great. He has, however, learned to listen, to concentrate, to write from dictation; he has become familiar with the teacher’s particular voice quality; but these abilities are also part of learning a language. In acquiring them he has learned some part of his second language (p. 37).
Morris (1983), too, drawing on the mistakes made by EFL learners on three dictations, concludes that dictation is a technique which can be used both as a testing technique and, more importantly, as a learning activity which helps students develop their accuracy in listening and writing and reinforces structure and vocabulary.
Moreover, Whitaker (1976) considers dictation as a good teaching device. He says, “... wherever aural comprehension is prized, together with literacy and ability to read the FL, dictation may be found to be both profitable for teaching, and valid for testing” (p. 92). Similarly, Pappas (1977) considers dictation as a good means of developing learners’ listening comprehension.
Valette (1964) asserts that practice with dictation can help students learn the language. She believes the teacher’s concentration on different component of the language, including sound, sentence structure, etc, encourages the students to correct their papers. In fact, she believes dictation can stimulate the students’ awareness of the written language.
Some other researchers believe that dictation can be utilized as a teaching technique due to the amount of input it provides for the learner. Stanfield (1985) mentions that since in dictation the learner employs more than one faculty, he is more successful in internalizing the language. Furthermore, Neisser (1967) considers dictation as an effective technique since the language learner goes through two cognitive processes, i.e., synthesis and analysis. Likewise, Jafarpur and Yamini (1993) claim that, “Dictation allows the language learner to both comprehend and produce the language in the context of meaningful discourse” (p. 360). Accordingly, they call dictation “dual-access processing” because “in taking dictation the subject alters and harmonizes his perception, conception, and expression.” Nevertheless, the results of their study with Iranian EFL learners show that practice with dictation cannot improve learners’ proficiency. However, as they mention, since language learning is a process which improves over time and needs an incubation period before any learning can be seen in the performance of the learners, the results of their study may not hold for the impact of practice with dictation when longer period of time is concerned.
The incubation period was first pointed out by the proponents of the natural approach to language teaching in such methods and approaches as Total Physical Response, and Natural Approach. In fact, they believe in the priority of listening to speaking in language learning. That is, we should wait for the language to which the learners are exposed to, to be internalized and then show its effects in the learners' production. This incubation or silent period is generalized to other areas of language learning. According to Alwright (1984) and Lightbown (1983), any teaching or learning technique does not immediately show its effect, so we should wait for some time (the incubation period) to see the effect of teaching and learning in the learners' performance.
Therefore, there seems to be an urgent need for another study to avoid the problem of the short time length between practicing with dictation and the post test given to the students to investigate the effect of dictation on the improvement of EFL learners' proficiency. This study is an attempt to deal with this issue.
Moreover, Whitaker (1976) considers dictation as a good teaching device. He says, “... wherever aural comprehension is prized, together with literacy and ability to read the FL, dictation may be found to be both profitable for teaching, and valid for testing” (p. 92). Similarly, Pappas (1977) considers dictation as a good means of developing learners’ listening comprehension.
Valette (1964) asserts that practice with dictation can help students learn the language. She believes the teacher’s concentration on different component of the language, including sound, sentence structure, etc, encourages the students to correct their papers. In fact, she believes dictation can stimulate the students’ awareness of the written language.
Some other researchers believe that dictation can be utilized as a teaching technique due to the amount of input it provides for the learner. Stanfield (1985) mentions that since in dictation the learner employs more than one faculty, he is more successful in internalizing the language. Furthermore, Neisser (1967) considers dictation as an effective technique since the language learner goes through two cognitive processes, i.e., synthesis and analysis. Likewise, Jafarpur and Yamini (1993) claim that, “Dictation allows the language learner to both comprehend and produce the language in the context of meaningful discourse” (p. 360). Accordingly, they call dictation “dual-access processing” because “in taking dictation the subject alters and harmonizes his perception, conception, and expression.” Nevertheless, the results of their study with Iranian EFL learners show that practice with dictation cannot improve learners’ proficiency. However, as they mention, since language learning is a process which improves over time and needs an incubation period before any learning can be seen in the performance of the learners, the results of their study may not hold for the impact of practice with dictation when longer period of time is concerned.
The incubation period was first pointed out by the proponents of the natural approach to language teaching in such methods and approaches as Total Physical Response, and Natural Approach. In fact, they believe in the priority of listening to speaking in language learning. That is, we should wait for the language to which the learners are exposed to, to be internalized and then show its effects in the learners' production. This incubation or silent period is generalized to other areas of language learning. According to Alwright (1984) and Lightbown (1983), any teaching or learning technique does not immediately show its effect, so we should wait for some time (the incubation period) to see the effect of teaching and learning in the learners' performance.
Therefore, there seems to be an urgent need for another study to avoid the problem of the short time length between practicing with dictation and the post test given to the students to investigate the effect of dictation on the improvement of EFL learners' proficiency. This study is an attempt to deal with this issue.
Method
Subjects
The subjects of the study consisted of 65 English majors enrolled in Conversation II classes. They took part in two separate classes: one taken as the experimental group (EG) consisted of 34 students and the other, the control group (CG), consisted of 31 students. They varied in age from 18 to 25 years with an average of 22.
Subjects
The subjects of the study consisted of 65 English majors enrolled in Conversation II classes. They took part in two separate classes: one taken as the experimental group (EG) consisted of 34 students and the other, the control group (CG), consisted of 31 students. They varied in age from 18 to 25 years with an average of 22.
Instrumentation
The main instrumentation of the study was Shiraz University Placement Test. It consisted of three sections: grammar, vocabulary, and reading comprehension with 60, 40, and 40 items, respectively. In addition, since the test did not contain any listening comprehension items, 25 listening items selected from Sharpe (1995) were added to it. Altogether, the test included 165 multiple-choice items (the test format was like the TOEFL). It was used as both pre-and post-tests to measure the subjects’ overall language proficiency.
In addition, the subjects took a listening comprehension multiple-choice test and a dictation consisting of 150 words as their final exam. In the former they listened to an unheard excerpt from Soars and Soars (2001) followed by some multiple-choice questions constructed by the researcher. The latter was excerpted from Hartley and Viney (1995) played for the examinees in the form of a systematic dictation. The scores of these tests, however, were not included in the analysis since they did not contribute any information to the study.
The main instrumentation of the study was Shiraz University Placement Test. It consisted of three sections: grammar, vocabulary, and reading comprehension with 60, 40, and 40 items, respectively. In addition, since the test did not contain any listening comprehension items, 25 listening items selected from Sharpe (1995) were added to it. Altogether, the test included 165 multiple-choice items (the test format was like the TOEFL). It was used as both pre-and post-tests to measure the subjects’ overall language proficiency.
In addition, the subjects took a listening comprehension multiple-choice test and a dictation consisting of 150 words as their final exam. In the former they listened to an unheard excerpt from Soars and Soars (2001) followed by some multiple-choice questions constructed by the researcher. The latter was excerpted from Hartley and Viney (1995) played for the examinees in the form of a systematic dictation. The scores of these tests, however, were not included in the analysis since they did not contribute any information to the study.
Procedure
In the first session of the conversation class, all the subjects took the proficiency test as the pre-test. Except for the listening comprehension part, there was no time limit for the test so that all the students would finish the test and a complete evaluation of their proficiency would be made.
The conversation class met three one-and-a-half-hour sessions a week for a total of 17 weeks - 51 sessions on the whole. The objective of the course was to enable the students to produce and understand spoken English at an intermediate level. That is, according to the syllabus, they must be able to use English at a comprehensible level while interacting with the teacher and each other rather fluently in spite of the mistakes in grammar, pronunciation, and vocabulary choice. With respect to the listening ability, too, they must be able to have a thorough, though not detailed, understanding of the spoken English in ordinary situations and contexts. For instance, when they listen to a dialog or a short lecture, they must be able to understand the whole message and be able to answer some global and general questions on them.
Class activities were divided into two parts: each week, in the first and the second session, they practiced New Interchange and New Headway. In the third session, the students gave lectures on already-determined and agreed-upon topics. Part of their course grades were based on these lectures. Both sections followed the same syllabus.
The EG, in addition, took a total of 50 dictations during the semester--each session, one dictation. The passages were selected from different sources, but mainly from books 2, 3, and 4 of Lado English Series (Lado, 1990). The reason was that the texts increase in length and difficulty levels as we move from one unit to another and from one book to another. The length of the passages ranged from 50 to 150 words. In each session the difficulty level and the length of the passages were increased. Thirty passages were dictated by the researcher; 20 others were read by other teachers, tape recorded, and were then played for the subjects. Each dictation exercise took about 10 minutes at the beginning of each session. It was read three times. First, it was read at a normal speed while the students just listened. Then, it was read in chunks with pauses between each chunk and the students wrote down what they heard. Finally, it was read at normal speed again. This time the students reviewed their writing to restore any missing parts.
The students’ writings were collected and corrected by the researcher and were given back to the students in the next session. The missing words and misspelled ones which changed the meaning of the words were considered mistakes. For instance, writing peace for piece was considered as a mistake but not peece for peace. For each mistake one point was subtracted from the total number of the words the passage contained. In the next session, before starting a new dictation, the corrected ones were returned to and reviewed with the subjects. In the review, the words that had been missed by the majority of the students were discussed. This procedure was followed since dictation was used as a teaching device and the purpose was for the students to learn from the dictations they had written and the errors they had made.
The students were asked to guess what part of speech the missing word had been. Then they were asked to say what the word would be with respect to the meaning of the whole text and the sentence it appeared in. In some occasions, they were asked to use reading skills, such as recognizing cohesion and coherence to guess the missing words/phrases. For instance, in the sentence “Then, he would blow the…… off,” they were asked to guess what part of speech the missing word was. They used such clues as the existence of the article the before the missing word and its appearance after a verb to conclude that they needed a noun. Then, the researcher made them think what things are blown off. Finally, since the topic was on “a birthday party,” they managed to find that the missing word was “candles.”
If they had made a spelling mistake due to the wrong recognition of the word while listening, some pronunciation practice was also done. For example, if they had written “sing” for “hing,” due to lack of the existence of the sound /Ө/ in Persian, this sound and the words in which it appeared were presented and practiced in class.
In order to compensate for the time spent on the dictations, the EG class was held 10 minutes longer than the CG class. The subjects were not informed of the objectives of the study.
After a lapse of one semester, at the end of the next semester (after about10 months), the same proficiency test--the post-test this time--was given to the same subjects. Since the test was relatively long and the time interval between its two administrations was long enough it was very unlikely that the students remembered the contents of the test from the first administration.
The conversation class met three one-and-a-half-hour sessions a week for a total of 17 weeks - 51 sessions on the whole. The objective of the course was to enable the students to produce and understand spoken English at an intermediate level. That is, according to the syllabus, they must be able to use English at a comprehensible level while interacting with the teacher and each other rather fluently in spite of the mistakes in grammar, pronunciation, and vocabulary choice. With respect to the listening ability, too, they must be able to have a thorough, though not detailed, understanding of the spoken English in ordinary situations and contexts. For instance, when they listen to a dialog or a short lecture, they must be able to understand the whole message and be able to answer some global and general questions on them.
Class activities were divided into two parts: each week, in the first and the second session, they practiced New Interchange and New Headway. In the third session, the students gave lectures on already-determined and agreed-upon topics. Part of their course grades were based on these lectures. Both sections followed the same syllabus.
The EG, in addition, took a total of 50 dictations during the semester--each session, one dictation. The passages were selected from different sources, but mainly from books 2, 3, and 4 of Lado English Series (Lado, 1990). The reason was that the texts increase in length and difficulty levels as we move from one unit to another and from one book to another. The length of the passages ranged from 50 to 150 words. In each session the difficulty level and the length of the passages were increased. Thirty passages were dictated by the researcher; 20 others were read by other teachers, tape recorded, and were then played for the subjects. Each dictation exercise took about 10 minutes at the beginning of each session. It was read three times. First, it was read at a normal speed while the students just listened. Then, it was read in chunks with pauses between each chunk and the students wrote down what they heard. Finally, it was read at normal speed again. This time the students reviewed their writing to restore any missing parts.
The students’ writings were collected and corrected by the researcher and were given back to the students in the next session. The missing words and misspelled ones which changed the meaning of the words were considered mistakes. For instance, writing peace for piece was considered as a mistake but not peece for peace. For each mistake one point was subtracted from the total number of the words the passage contained. In the next session, before starting a new dictation, the corrected ones were returned to and reviewed with the subjects. In the review, the words that had been missed by the majority of the students were discussed. This procedure was followed since dictation was used as a teaching device and the purpose was for the students to learn from the dictations they had written and the errors they had made.
The students were asked to guess what part of speech the missing word had been. Then they were asked to say what the word would be with respect to the meaning of the whole text and the sentence it appeared in. In some occasions, they were asked to use reading skills, such as recognizing cohesion and coherence to guess the missing words/phrases. For instance, in the sentence “Then, he would blow the…… off,” they were asked to guess what part of speech the missing word was. They used such clues as the existence of the article the before the missing word and its appearance after a verb to conclude that they needed a noun. Then, the researcher made them think what things are blown off. Finally, since the topic was on “a birthday party,” they managed to find that the missing word was “candles.”
If they had made a spelling mistake due to the wrong recognition of the word while listening, some pronunciation practice was also done. For example, if they had written “sing” for “hing,” due to lack of the existence of the sound /Ө/ in Persian, this sound and the words in which it appeared were presented and practiced in class.
In order to compensate for the time spent on the dictations, the EG class was held 10 minutes longer than the CG class. The subjects were not informed of the objectives of the study.
After a lapse of one semester, at the end of the next semester (after about10 months), the same proficiency test--the post-test this time--was given to the same subjects. Since the test was relatively long and the time interval between its two administrations was long enough it was very unlikely that the students remembered the contents of the test from the first administration.
Data analysis
The results of the tests were subjected to the following statistical procedures. Two independent t-tests were run for the difference between the mean scores of the EG and the CG on the pre-test as well as the post-test. A paired t-test was run to find the difference between the means of the scores on the following tests: the pre- and post- proficiency tests for the CG as well as for the EG to see if there is any difference between the performance of the subjects on the pre- and post-tests; different components of the pre- and post-tests for the CG as well as for the EG to see if there is any significant difference between the performance of the subjects on different components of the pre-test and the post-test. In addition, an independent t-test was run to find the difference between the means of the scores of the subjects in the EG and the means of the scores of the subjects in the CG in the pre-test; the same test was run in the case of the post-test, too. Furthermore, scores on the whole pre-and post-tests and their components for both groups were subjected to Kuder-Richardson 21 to estimate their reliability coefficients. In addition, true mean gains were calculated for the pre- and post-proficiency tests and their components for both groups to see if there is any difference between the gains of the subjects in the CG and the EG. Finally, an ANOVA test was run for the mean gains of the subjects in the components of the post-test.
The results of the tests were subjected to the following statistical procedures. Two independent t-tests were run for the difference between the mean scores of the EG and the CG on the pre-test as well as the post-test. A paired t-test was run to find the difference between the means of the scores on the following tests: the pre- and post- proficiency tests for the CG as well as for the EG to see if there is any difference between the performance of the subjects on the pre- and post-tests; different components of the pre- and post-tests for the CG as well as for the EG to see if there is any significant difference between the performance of the subjects on different components of the pre-test and the post-test. In addition, an independent t-test was run to find the difference between the means of the scores of the subjects in the EG and the means of the scores of the subjects in the CG in the pre-test; the same test was run in the case of the post-test, too. Furthermore, scores on the whole pre-and post-tests and their components for both groups were subjected to Kuder-Richardson 21 to estimate their reliability coefficients. In addition, true mean gains were calculated for the pre- and post-proficiency tests and their components for both groups to see if there is any difference between the gains of the subjects in the CG and the EG. Finally, an ANOVA test was run for the mean gains of the subjects in the components of the post-test.
Results and discussions
The mean scores of CG and EG on the pre-test were compared to see if they were the same or different before the experiment started. Table 1 shows the result of t-test.
The mean scores of CG and EG on the pre-test were compared to see if they were the same or different before the experiment started. Table 1 shows the result of t-test.
Table 1
Independent t-test for EG and CG on pre-test
Independent t-test for EG and CG on pre-test
Mean Difference between EG and CG in Pre-test | CG | EG | df | t | Sig |
83.8 | 88.38 | 63 | -1.01 | 0.31 |
As evident in the table, there was not significant difference between the two groups before the experiment started and the two group started with the same proficiency level.
Table 2 illustrates the results of t-test for the performance of the two groups on the post test.
Table 2 illustrates the results of t-test for the performance of the two groups on the post test.
Table 2
Independent t-test for EG and CG on post-test
Independent t-test for EG and CG on post-test
As the results indicate, EG significantly did better than CG in the post-test. In other words, the experimental group who had regular practice with dictation made more improvement in their general proficiency than the control group. Since the two groups of students had almost the same classes during the time interval between the pre-and post-test, one can claim that the difference in their performance on the proficiency test was due to the fact that the experimental group had practice with dictation.
In order to see how much improvement each group had from the pre-test to the post-test, first descriptive statistics and then a matched t-test were conducted. Table 3 shows descriptive statistics for the results of the pre- and post-tests for the CG.
In order to see how much improvement each group had from the pre-test to the post-test, first descriptive statistics and then a matched t-test were conducted. Table 3 shows descriptive statistics for the results of the pre- and post-tests for the CG.
Table 3
Descriptive statistics for the results of the pre-and post-tests for the CG
Descriptive statistics for the results of the pre-and post-tests for the CG
Pre-test Post-test | ||||||||||
statistics | Whole test | Gr | Voc | R | L | Whole test | Gr | Voc | R | L |
K | 165 | 60 | 40 | 40 | 25 | 165 | 60 | 40 | 40 | 25 |
`X | 83.8 | 38.7 | 17.15 | 16.3 | 11.65 | 89.8 | 39.2 | 21.25 | 18.55 | 12.15 |
SD | 19.57 | 5.82 | 6.39 | 6.36 | 4.88 | 20.28 | 5.09 | 6.72 | 7.43 | 5 |
Range | 60 (58-118) | 19 (30-51) | 24 (4-28) | 22 (8-30) | 16 (4-20) | 67 (65-132) | 19 (30-51) | 24 (8-32) | 25 (8-33) | 18 (5-23) |
Skewness | 0.43 | 0.27 | -0.2 | 0.55 | 0.53 | 0.71 | 0.29 | -0.18 | 0.33 | 0.69 |
Kurt. | -0.99 | -0.45 | -0.52 | -0.16 | -0.77 | -0.63 | 0.2 | -0.52 | -0.7 | -0.4 |
KR-21 | 0.9 | 0.6 | 0.77 | 0.77 | 0.75 | 0.9 | 0.49 | 0.79 | 0.83 | 0.78 |
The mean of the pre-test for the CG is 83.8, and that of the post-test is 89.8. The standard deviation for the pre-test is 19.57, and that of the post-test 20.28. Besides, the range of the scores of the pre-test is 60, whereas that of the post-test is 67. Furthermore, the distribution of the scores of both pre- and post-tests is positively skewed, though the latter is more positively skewed (0.43 and 0.71). In addition, the distribution of the scores obtained from both tests is flat (-0.99 and -0.63). Finally, the reliability coefficients of both tests (measured through KR-21) are the same, i.e., 0.90.
With respect to the components of the pre-test and the post-test, in all the cases the means of the components of the post-test show some increase as compared with the pre-test. In the case of grammar, for instance, the mean for the pre-test is 38.7 and that of the post-test 39.2. With respect to vocabulary, too, the mean of the post-test is larger than that of the pre-test (21.25 and 17.15, respectively). The mean scores of reading comprehension and listening comprehension also show the same pattern.
Regarding the standard deviation, one cannot see a remarkable difference between the components of the two tests. The same is true with regard to the range of the scores of different components of the two tests. Concerning the distribution of the scores on the components of the tests, except for the vocabulary part whose distribution is negatively skewed in both pre- and post-tests (-0.2 and -0.18, respectively), in all other cases one can see positive skewness. This shows that in both pre- and post-tests the vocabulary part has been the easiest. Finally, as the kurtosis values for different components of the tests show, except for the vocabulary part, a large difference can be seen in the case of the other parts. That is, with respect to grammar, as it is evident in the table, the distribution of the scores in the pre-test is flat and that of the post-test is rather peaked (-0.45 and 0.2, respectively). With regard to listening comprehension and reading comprehension, though in both pre- and post-tests the distribution of the scores is flat, the degree of peakedness largely differs. In the case of listening comprehension it is flatter in the pre-test (-0.77) than in the post-test (-0.4), whereas it is quite the opposite in the case of reading comprehension (-0.16 in the pre-test and -0.7 in the post-test).
With respect to the components of the pre-test and the post-test, in all the cases the means of the components of the post-test show some increase as compared with the pre-test. In the case of grammar, for instance, the mean for the pre-test is 38.7 and that of the post-test 39.2. With respect to vocabulary, too, the mean of the post-test is larger than that of the pre-test (21.25 and 17.15, respectively). The mean scores of reading comprehension and listening comprehension also show the same pattern.
Regarding the standard deviation, one cannot see a remarkable difference between the components of the two tests. The same is true with regard to the range of the scores of different components of the two tests. Concerning the distribution of the scores on the components of the tests, except for the vocabulary part whose distribution is negatively skewed in both pre- and post-tests (-0.2 and -0.18, respectively), in all other cases one can see positive skewness. This shows that in both pre- and post-tests the vocabulary part has been the easiest. Finally, as the kurtosis values for different components of the tests show, except for the vocabulary part, a large difference can be seen in the case of the other parts. That is, with respect to grammar, as it is evident in the table, the distribution of the scores in the pre-test is flat and that of the post-test is rather peaked (-0.45 and 0.2, respectively). With regard to listening comprehension and reading comprehension, though in both pre- and post-tests the distribution of the scores is flat, the degree of peakedness largely differs. In the case of listening comprehension it is flatter in the pre-test (-0.77) than in the post-test (-0.4), whereas it is quite the opposite in the case of reading comprehension (-0.16 in the pre-test and -0.7 in the post-test).
Table 4
Difference between the means of the pre- and post-tests and their components in the CG
Difference between the means of the pre- and post-tests and their components in the CG
The whole tests | Gr | Voc | R | L | The whole pre-and post-test excluding voc. | |
Difference between the means | -2.12 (s) | -0.57 (ns) | -3.33 (s) | -1.80 (ns) | -0.73 (ns) | -0.62 (ns) |
According to Table 4, there is a significant difference between the results of the pre- and post-tests for the CG. However, looking at the components of the two tests, one will see that, except for the vocabulary part, there is no significant difference between the results of the two tests. Thus, one may doubt whether it is because of the scores on the vocabulary part that there exists a significant difference between the two tests. The hypothesis was verified when a t-test was run between the two tests excluding the vocabulary part. As it can be seen in the table, there is no significant difference between the two tests in this case.
Nevertheless, the results are quite different in the case of the EG. The descriptive statistics for the pre- and post-tests for this group are presented in Table 5.
Nevertheless, the results are quite different in the case of the EG. The descriptive statistics for the pre- and post-tests for this group are presented in Table 5.
Table 5
Descriptive statistics of the pre- and post-tests for the EG
Descriptive statistics of the pre- and post-tests for the EG
According to this table, the mean for the pre-test is 88.38, whereas that of the post-test is 114.32. There is a relatively high difference between the standard deviations of the two tests: 21.58 in the case of the pre-test and 13.71 for the post-test. This shows that the subjects have performed closer to homogeneously on the post-test. This claim can be verified referring to the large difference between the range of the scores on the two tests (121 for the pre-test and 61 for the post-test). Furthermore, the distribution of the scores on the two tests is negatively skewed (-1.39 and -0.19), though that of the post-test tends to be more normal. Besides, the distribution of the scores of the two tests is peaked (4.95 and 0.37). However, that of the post test seems to be much closer to a normal distribution.
Regarding the components of the two tests, one can see an increase in the means in all the cases. The highest increase, however, can be seen in the case of the reading part, i.e., from 17.87 to 27.19. Concerning the standard deviation, on the other hand, one can see a relative decrease in all the cases. With regard to the range of the scores, except for the vocabulary part which shows some increase from the pre-test to the post-test--23 to 25--, in all other cases the range has decreased, indicating the fact that the subjects have performed more homogeneously on the subtests. As for the skewness, rather large differences can be seen between the components of the pre- and post-tests. In the case of grammar, while the distribution of the scores of both tests is negatively skewed (-1.25 in the case of the pre-test and -0.19 in the case of the post-test), that of the post test is much closer to the normal distribution. With respect to vocabulary, the difference is more remarkable.
That is, the distribution of the scores in the pre-test is positively skewed (0.31), whereas that of the post-test is negatively skewed (-0.45), indicating the fact that the latter has been easier. Similarly, with regard to the reading comprehension part, one can see that the distribution of the scores tends to be more negatively skewed in the post-test. As it is evident in the table, the distribution of the scores in the pre-test is very much close to the normal distribution (-0.01), whereas it is rather highly negatively skewed in the post-test (-1.19). Finally, the distribution of the scores on listening comprehension of both tests is positively skewed (0.29 and 0.47), implying that this part has been difficult both in the pre- and post-tests. Concerning the peakedness of the distribution of the scores, the only component of the test which shows a difference in the pre- and post-tests is reading comprehension. That is, the distribution of the scores in the pre-test is flat (-0.67), while it has become peaked (0.86) in the post-test. This shows that the scores on the post-test have been much closer to the mean. In all other cases, the peakedness has remained unchanged in spite of some changes in the degree.
Table 6 represents the difference between the means of the two tests and their different components for the EG.
Regarding the components of the two tests, one can see an increase in the means in all the cases. The highest increase, however, can be seen in the case of the reading part, i.e., from 17.87 to 27.19. Concerning the standard deviation, on the other hand, one can see a relative decrease in all the cases. With regard to the range of the scores, except for the vocabulary part which shows some increase from the pre-test to the post-test--23 to 25--, in all other cases the range has decreased, indicating the fact that the subjects have performed more homogeneously on the subtests. As for the skewness, rather large differences can be seen between the components of the pre- and post-tests. In the case of grammar, while the distribution of the scores of both tests is negatively skewed (-1.25 in the case of the pre-test and -0.19 in the case of the post-test), that of the post test is much closer to the normal distribution. With respect to vocabulary, the difference is more remarkable.
That is, the distribution of the scores in the pre-test is positively skewed (0.31), whereas that of the post-test is negatively skewed (-0.45), indicating the fact that the latter has been easier. Similarly, with regard to the reading comprehension part, one can see that the distribution of the scores tends to be more negatively skewed in the post-test. As it is evident in the table, the distribution of the scores in the pre-test is very much close to the normal distribution (-0.01), whereas it is rather highly negatively skewed in the post-test (-1.19). Finally, the distribution of the scores on listening comprehension of both tests is positively skewed (0.29 and 0.47), implying that this part has been difficult both in the pre- and post-tests. Concerning the peakedness of the distribution of the scores, the only component of the test which shows a difference in the pre- and post-tests is reading comprehension. That is, the distribution of the scores in the pre-test is flat (-0.67), while it has become peaked (0.86) in the post-test. This shows that the scores on the post-test have been much closer to the mean. In all other cases, the peakedness has remained unchanged in spite of some changes in the degree.
Table 6 represents the difference between the means of the two tests and their different components for the EG.
Table 6
Difference between the means of the pre- and post-tests and their components for the EG
Difference between the means of the pre- and post-tests and their components for the EG
The whole tests | Gr | Voc | R | L | |
Difference between the means | -7.25 (s) | -4.13 (s) | -9.33 (s) | -7.45 (s) | -3.32 (s) |
The results show that there is a significant difference between the means on the pre- and post-tests and all their components. Since the mean scores of the subjects in all the cases have increased from the pre-test to the post-test, it can be claimed that practice with dictation has had a positive effect on the improvement of the proficiency of the subjects, as compared with the results of the tests taken by the CG where one can see such an increase only in the vocabulary part. The reason for the increase observed in both cases can be the fact that since the subjects majored in English Translation and in most of their courses much emphasis was laid on developing their vocabulary, after two semesters, both groups’ knowledge of vocabulary has improved rather equally.
The hypothesis that practice with dictation has probably been the major factor improving the proficiency level of the subjects in the EG can be verified by two pieces of evidence. Firstly, according to Table 5 (below), there is no significant difference between the scores of the subjects in the EG and the CG on the pre-test. In the post-test, however, a significant difference can be observed between the scores of the two groups. Since the mean of the scores for the EG is larger than that of the CG, one can truly claim that practice with dictation has helped students improve their proficiency in the English language.
The hypothesis that practice with dictation has probably been the major factor improving the proficiency level of the subjects in the EG can be verified by two pieces of evidence. Firstly, according to Table 5 (below), there is no significant difference between the scores of the subjects in the EG and the CG on the pre-test. In the post-test, however, a significant difference can be observed between the scores of the two groups. Since the mean of the scores for the EG is larger than that of the CG, one can truly claim that practice with dictation has helped students improve their proficiency in the English language.
Table 7
Difference between the means of the scores of the EG and the CG
Difference between the means of the scores of the EG and the CG
Difference between the means in the pre-test | Difference between the means in the post-test |
0.78 (ns) | 4.75 (s) |
Another piece of evidence supporting this claim is the difference between the gains of the two groups. Table 8 presents the gains of the subjects in the EG and the CG.
Table 8
The mean gain of the subjects in the EG and the CG
The mean gain of the subjects in the EG and the CG
As presented in the table, the true mean gain obtained by the subjects in the EG is much larger than that of the CG (25.93 and 4.06, respectively). In other words, though both groups have had some gain from the pre-test to the post-test, that of the EG who have made practice with dictation is much larger. The same is true regarding the vocabulary part. That is, the gain in the EG is approximately twice as much as that of the CG. With regard to other components of the test, since the difference between the pre- and post-tests in the CG was not significant, it was not possible to measure any gain.
Finally, in order to see if there is any difference between the mean gain of the subjects in the EG on different subtests, an ANOVA test was run. The results represented in Table 9 show that there is a significant difference among the mean gains obtained on different parts of the test
Table 9
ANOVA test for different components of the post-test for the EG
Finally, in order to see if there is any difference between the mean gain of the subjects in the EG on different subtests, an ANOVA test was run. The results represented in Table 9 show that there is a significant difference among the mean gains obtained on different parts of the test
Table 9
ANOVA test for different components of the post-test for the EG
According to Table 10 which presents the results of the scheffe, except for reading and vocabulary on the one hand and listening and grammar on the other, the means on all other components show significant differences. That is, there is a significant difference between the mean gain of the subjects on vocabulary and listening and grammar. The same holds true concerning reading comprehension and vocabulary and grammar. Thus, since the mean gain in vocabulary and reading comprehension has been larger than the grammar and listening comprehension, one can claim that practice with dictation has improved the subjects’ vocabulary and reading comprehension more than their grammar and listening comprehension.
Conclusions
This study investigated the effect of practice with dictation on the learners’ proficiency. There were an experimental and a control group. The experimental group took a dictation in every session of their class for a period of one semester. Each session, the researcher practiced the corrected dictation with them and elaborated on grammatical, phonological, and lexical points in the dictated passage. The results of the study showed that the experimental group made some improvement in their language proficiency. The control group, however, showed no improvement on the proficiency test and its components except for the vocabulary subtest.
The results of the study, to some extent, support those of Valette (1964) and Morris (1983), since they claimed that practice with dictation helps learners improve their language proficiency. However, they mostly emphasized the improvement in learners’ listening and writing skills, whereas in the present study listening comprehension and grammar turned out to be the language components least improved and reading and vocabulary the ones most improved.
The results of the study, on the other hand, contradict those of Jafarpur and Yamini (1993) in that they saw no improvement in the proficiency of the experimental group as compared with the control group. Nevertheless, since they administered the post-test at the end of the semester during which the experiment was conducted (after about 4 months), the ‘incubation hypothesis’ might have been partly disregarded (as they themselves stated). The results of the present study lend support to the above hypothesis which implies that a teaching technique may show its effects in the long run.
This study, nonetheless, suffers from a couple of shortcomings. Firstly, since the post-test was given a long time after the pre-test, i.e., after about 10 months, some of the subjects might have improved their proficiency via extracurricular activities, e.g., attending language institutes and reading short stories. The low number of the subjects participating in the study would amplify the effect of such activities on the results of the study.
Secondly, the pre- and post-tests measured just the learners’ recognition rather than production. Including tests of writing and speaking in the pre- and post-tests could have shown if practice with dictation has any effect on the improvement of such skills or not.
Finally, as mentioned above, the class for EG was held about 10 minutes longer than that of CG. Though it was inevitable, as the researcher had to ensure that everything else was the same for the two classes except for the dictation, this longer time for EG might have contaminated the results. If fact, the better performance of EG on the post-test might be partly due to having more language practice as compared to CG.
This study investigated the effect of practice with dictation on the learners’ proficiency. There were an experimental and a control group. The experimental group took a dictation in every session of their class for a period of one semester. Each session, the researcher practiced the corrected dictation with them and elaborated on grammatical, phonological, and lexical points in the dictated passage. The results of the study showed that the experimental group made some improvement in their language proficiency. The control group, however, showed no improvement on the proficiency test and its components except for the vocabulary subtest.
The results of the study, to some extent, support those of Valette (1964) and Morris (1983), since they claimed that practice with dictation helps learners improve their language proficiency. However, they mostly emphasized the improvement in learners’ listening and writing skills, whereas in the present study listening comprehension and grammar turned out to be the language components least improved and reading and vocabulary the ones most improved.
The results of the study, on the other hand, contradict those of Jafarpur and Yamini (1993) in that they saw no improvement in the proficiency of the experimental group as compared with the control group. Nevertheless, since they administered the post-test at the end of the semester during which the experiment was conducted (after about 4 months), the ‘incubation hypothesis’ might have been partly disregarded (as they themselves stated). The results of the present study lend support to the above hypothesis which implies that a teaching technique may show its effects in the long run.
This study, nonetheless, suffers from a couple of shortcomings. Firstly, since the post-test was given a long time after the pre-test, i.e., after about 10 months, some of the subjects might have improved their proficiency via extracurricular activities, e.g., attending language institutes and reading short stories. The low number of the subjects participating in the study would amplify the effect of such activities on the results of the study.
Secondly, the pre- and post-tests measured just the learners’ recognition rather than production. Including tests of writing and speaking in the pre- and post-tests could have shown if practice with dictation has any effect on the improvement of such skills or not.
Finally, as mentioned above, the class for EG was held about 10 minutes longer than that of CG. Though it was inevitable, as the researcher had to ensure that everything else was the same for the two classes except for the dictation, this longer time for EG might have contaminated the results. If fact, the better performance of EG on the post-test might be partly due to having more language practice as compared to CG.
References
Allwright, R.L. (1984). The importance of interaction in classroom language learning. Applied Linguistics, 5, 156-171.
Cartledge, H. A. (1968). A defence of dictation. ELT Journal, 22, 226-231
Harris, D.P. (1965). Testing English as a second language. New York: McGraw-Hill.
Hartley, B., & Viney, P. (1995). New American streamline: Connections. Oxford: Oxford University Press.
Irvine, P., Atai, P., & Oller, G.W. (1974). Cloze, dictation, and the test of English as a foreign language. Language Learning, 245-254.
Jafarpur, A. & Yamini, M. (1993). Does practice with dictation improve language skills? System, 21, 359-369.
Lado, R. (1961). Language testing. New York: McGraw-Hill.
Lado, R. (1990). Lado English series. New Jersey: Prentice-Hall.
Lightbown, P.M. (1983). Exploring relationships between developmental and instructional sequences in L2 acquisition. In H. W. Selinger & M. Long, (Eds.), Classroom-oriented research in second language acquisition, pp. 217-245. Rowley, MA: Newbery House.
Lord, E.M. (1963). Elementary models for measuring change. In C. W. Harris (Ed.), Problems in measuring change. Madison: University of Wisconsin.
Morris, S. (1983). Dictation-a technique in need of reappraisal. ELT Journal, 37, 121-126.
Neisser, U. (1967). Cognitive psychology. New York: Appleton-Century Crofts.
Oller, J. W. (1971). Dictation as a device for testing foreign-language proficiency. ELT Journal, 25, 254-259.
Oller, J. W. & Streiff, V. (1975). Dictation: A test of grammar-based expectancies. ELT Journal, 30, 25-36.
Pappas, G.S. (1977). You mean you still give dictations? Language Arts, 54, 936-939.
Richards, J.C., Hull, J., & Proctor, S. (1997). New interchange English for international communication. New York: Cambridge University Press.
Sawyer, J. & Silver, S. (1961). Dictation in language learning. Language Learning, 11, 33-42.
Sharpe, P.J. (1995). Barron's how to prepare for the TOEFL test of English as a foreign language. Woodbury, New York: Barron's Educational Series.
Soars, J. & Soars, L. (2001). New headway intermediate. Oxford: Oxford University Press.
Somaratne, W. (1957). Aids and Tests in the Teaching of English. Oxford: Oxford University Press.
Standfield, C. W. (1985). A history of dictation in foreign language teaching and testing. Modern Language Journal, 69, 121-128.
Valette, R.M. (1964). The use of the dictee in the French language classroom. Modern Language Journal, 48, 431-434.
Whitaker, S.F. (1976). What is the status of dictation? Audio-Visual Language Journal, 14, 87-93.
Allwright, R.L. (1984). The importance of interaction in classroom language learning. Applied Linguistics, 5, 156-171.
Cartledge, H. A. (1968). A defence of dictation. ELT Journal, 22, 226-231
Harris, D.P. (1965). Testing English as a second language. New York: McGraw-Hill.
Hartley, B., & Viney, P. (1995). New American streamline: Connections. Oxford: Oxford University Press.
Irvine, P., Atai, P., & Oller, G.W. (1974). Cloze, dictation, and the test of English as a foreign language. Language Learning, 245-254.
Jafarpur, A. & Yamini, M. (1993). Does practice with dictation improve language skills? System, 21, 359-369.
Lado, R. (1961). Language testing. New York: McGraw-Hill.
Lado, R. (1990). Lado English series. New Jersey: Prentice-Hall.
Lightbown, P.M. (1983). Exploring relationships between developmental and instructional sequences in L2 acquisition. In H. W. Selinger & M. Long, (Eds.), Classroom-oriented research in second language acquisition, pp. 217-245. Rowley, MA: Newbery House.
Lord, E.M. (1963). Elementary models for measuring change. In C. W. Harris (Ed.), Problems in measuring change. Madison: University of Wisconsin.
Morris, S. (1983). Dictation-a technique in need of reappraisal. ELT Journal, 37, 121-126.
Neisser, U. (1967). Cognitive psychology. New York: Appleton-Century Crofts.
Oller, J. W. (1971). Dictation as a device for testing foreign-language proficiency. ELT Journal, 25, 254-259.
Oller, J. W. & Streiff, V. (1975). Dictation: A test of grammar-based expectancies. ELT Journal, 30, 25-36.
Pappas, G.S. (1977). You mean you still give dictations? Language Arts, 54, 936-939.
Richards, J.C., Hull, J., & Proctor, S. (1997). New interchange English for international communication. New York: Cambridge University Press.
Sawyer, J. & Silver, S. (1961). Dictation in language learning. Language Learning, 11, 33-42.
Sharpe, P.J. (1995). Barron's how to prepare for the TOEFL test of English as a foreign language. Woodbury, New York: Barron's Educational Series.
Soars, J. & Soars, L. (2001). New headway intermediate. Oxford: Oxford University Press.
Somaratne, W. (1957). Aids and Tests in the Teaching of English. Oxford: Oxford University Press.
Standfield, C. W. (1985). A history of dictation in foreign language teaching and testing. Modern Language Journal, 69, 121-128.
Valette, R.M. (1964). The use of the dictee in the French language classroom. Modern Language Journal, 48, 431-434.
Whitaker, S.F. (1976). What is the status of dictation? Audio-Visual Language Journal, 14, 87-93.
Tidak ada komentar:
Posting Komentar