An Examination of 1st, 2nd And 3rd Grade Elementary School Students’ Story-Telling Skills Based on Narrative Analysis

Narrative skills are important in terms of the development of children's academic skills (reading, writing, language). This research was planned to assess the story-telling skills of elementary school 1st, 2nd and 3rd grade pupils exhibiting typical growth with the objective of making a comparison between grades and genders. The study recruited 120 students enrolled in the 1st, 2nd and 3rd grades of an elementary school in the city center of Ankara. The wordless picture book "Frog, where are you?" was used in the research. A "Story-telling Assessment Questionnaire" was used to assess the stories that the children told. The SPSS 16.01 statistics program was used for the analysis of the data in the research. According to the results of the study, it was found that story-telling skills developed with age and that the rate of story comprehension of children is generally similar across ages and genders. It has been reported that the story-telling method is a useful way of assessing a child's language and academic qualities.


Introduction
Narratives, chats, explanations, discussions, persuasive talk and descriptions are all forms of discourse.A more developed form of describing an event is story-telling.In telling a story, the narrator forms grammatical structures; telling stories is a skill that begins to be learned starting from an early age (Topbaş, 2005).
Assessing a child's verbal narration skills is important in childhood and for this reason this has been a topic that has attracted the interest of researchers.Today verbal narrative skills have become a key component of school curriculums in the United States and each state has defined these skills and made them a part of their general education programs (Heilmann, Miller & Nockerts, 2010).
In a socio-cultural context (environment, peers, teachers and parents), narration provides a wide perspective on children's learning and development.It is particularly useful in determining the educational needs of children (Bourke & Mentis, 2010).At the same time, narration is frequently used to analyze a child's cognitive and linguistic skills in clinical evaluations.There are many reasons for this.According to Westby (1989), verbal narration serves as a bridge between written narratives.Stories are customarily used in narration assessments.Story-telling improves a child's verbal language and listening skills.These two basic skills are of the greatest importance in mastering the skills of reading and writing.Story-telling has been found to be important in helping children use their language correctly, widen their vocabularies and become skillful in finding the right words to use in speaking (Caulfield, 2000).Stadler and Ward (2005) have stated the importance of narrative skills in terms of three perspectives.These perspectives are: 1) Narrative skills are useful tools in assessing verbal language.
2) Verbal narratives are bridges between reading and writing.
3) Narratives are directly related to the development of concepts.
It is believed that narrative skills are an indicator of both mental skills and language skills.Various studies about narrative skills have been conducted with children who exhibit typical development as well as with children with special needs.In their study, Bishop and Adams (1992) found that children with specific speech difficulties were delayed in developing their narrative skills compared to children exhibiting normal development and an association was detected between narrative skills and reading skills.Thus, narrative skills are important in the assessment of children with speech problems and in interventions in this context.
According to Adams (2002), narratives also help to assess children's language skills from a pragmatic perspective, greatly facilitating the identification of children with language difficulties.The methods of retelling or story generation are frequently used in assessing competence in narration.Other methods used are describing skills or routine events, creating stories by looking at a wordless picture book, and narrating personal experiences.
Narrative analysis is important in determining the academic status of school-age children.This is because story-telling skills make it easier to understand a child's cognitive processing skills, helping to determine where academic difficulties lie.Additionally, story comprehension is a significant predictor of cognitive development since narratives entail a developmental process.For this reason, a child's narrative can provide a key to his/her developmental level (Renz, Lorch, Milich, Lemberger, Bodner, & Welsh, 2003).

Purpose of the research
Story-telling requires skills involving the organization of knowledge, planning and selfmonitoring (executive functioning).There are few studies in Turkey on narrative analysis.Therefore this research uses narrative analysis to assess the story-telling skills of children enrolled in the 1st, 2nd and 3rd grades of elementary school, and makes a comparison in terms of age and gender.Consequently, the research questions are the following: 1. Is there a difference in story-telling skills between children in the 1st, 2nd and 3rd grades of elementary school (in terms of age)?
2. Is there a difference in story-telling skills between children in the 1st, 2nd and 3rd grades of elementary school in terms of gender?

Research Universe
The universe of the research comprised children enrolled in the 1st, 2nd and 3rd grades in the elementary schools connected to the Ministry of National Education in the city center of Ankara.

Research Sample
Ten schools connected to the Ministry of National Education in the city center of Ankara were selected for the research by the random sampling method.The approval of the Ministry of National Education was first obtained for the study.The study recruited 4 children each from the 1st, 2nd and 3rd grades of each school (2 girls and 2 boys from each class), a total of 120 participants.The children were selected again by the random sampling method from among the children who wished to participate.The classroom teachers confirmed that the selected pupils were not in any integration program, did not have academic failures and did not have any obstacle standing in the way of their participation.The data were collected over the period April-May of the academic year and all of the children knew how to read and write.

Research Model
The research was of descriptive and comparative design.The students' story-telling skills were reviewed from a descriptive perspective and their language skills were evaluated according to gender and grade at school.

Data Collection Tools
Wordless Picture Book The 24-page wordless picture book "Frog, Where are You?" written by Mercer Mayer in 1969 was used for the study.This book has frequently been used in studies on narrative analysis.According to Wigglesworth (1991), the story was first used by Bamberg to assess the narratives of German-speaking children.The content of the book is appropriate for an assessment of children's narrative and story organization skills.The same storybook has also been used by Slobin (1987Slobin ( , 1988) ) in language-learning and comparative language studies.The story is about a child, his dog and a frog that has run away from home.The child and the dog set off to find the frog in the story, which goes on to relate the events that happen to the child and dog.

Story-Telling Assessment Form:
The researcher developed a "Story-telling Assessment Form" for the purpose of evaluating the pupils' story-telling skills.The control list that was drawn up was tested for content validity.For this, the "Story-telling Assessment Form" was subjected to the review of five specialists and the necessary revisions were made.The specialists were university faculty members in the psychology, foreign languages and special education departments; the specialists had previously worked in narrative analysis.

Internal Consistency of the Instrument:
The analysis for the internal consistency of the instrument that was developed was performed using the SPSS 16.01 package program as well as the Office 2007/Excel program.The Kuder-Richardson 20 formula was used to establish reliability; the Excel program was used in the analysis.The Kuder Richardson 20 formula applies 1 point to correct answers and 0 points to incorrect or missed items (Tekin, 2004).The formula is the following: KR-20: rx=n/n-1[1-Σp.q/s 2 ]q

Results of the reliability calculations of the instrument:
In the first analysis of the preliminary control list, containing 116 items, the item strength index and the item discrimination power index tested low and therefore 19 items were removed from the control list.The removed items were the following: in the setting component, 1 item from the section on "the initiating event"; in the attempt component, 2 items from the section on "actions in the face of obstacles; 1 item from the consequence component; 2 items from the errors section; 2 items from the "name the pictures differently"; 3 questions from the evaluation tasks, and the 8 items that comprised the responses to these questions.The result obtained in the second analysis performed on the remaining items can be seen in Figure 1.

Sections of the Instrument:
The "Story-Telling Assessment Form" used to evaluate the stories the children told in the study comprised a total of 97 items.The control list is made up of the following sections: the story's setting, the attempt, the consequence, story comprehension, errors and evaluation tasks.Besides these sections, there are 15 items related to story length and story-telling rate.

Stylistic Analysis of the Story
The children's stories were reviewed in terms of language according to the characteristics below.a) Story length: In this section, the stories the children told were examined for the following numerical data: the number of C units, number of simple utterances, total word count, number of distinct words, distinct vocabulary rate, number of total utterances, number of completely understandable utterances, rate of utterances with distinct meanings, number of words repeated, word repeat rate.
b) Story-telling rate: This section comprises time (seconds), utterance production speed, rate of production of utterances with distinct meaning, vocabulary production speed, distinct vocabulary production speed

Data Collection Process
The data collection process was undertaken by meeting with each child in a classroom using the face-to-face interview technique.The researcher gave all of the children participating in the study a copy of the book "Frog, Where are You?" and told them that the book was about a child, his dog and a frog that was kept as a pet at home.The children were asked to look at the whole book first, study the pictures and then retell the story looking at the book.During the process, each child's story-telling was recorded on an audio recording device.

Analysis of Data
The researcher then listened to the stories recorded on the device and transcribed them.After another specialist listened to 30% of the recordings, inter-rater reliability testing found an agreement of 0.9711 (range: 86.9%-100.0%).The transcribed stories were evaluated by the researcher, who accordingly filled out the "Story-telling Assessment Form."Another specialist evaluated 30% of the transcribed stories according to the control list and inter-rater reliability testing this time found an agreement of 0.9913 (range: 90.1%-100.0%).The SPSS 16.01 package program was used for the evaluation of the data.
The story-telling skills in the study findings were analyzed in terms of the grade and gender of each member of the sample group (n=120) .The chi-square test was used for the analysis by grades of the items on the story's setting, attempt, consequence, story comprehension, errors and evaluation tasks.Sümbüloğlu and Sümbüloğlu (2009) state that data defined as qualitative should be used in the chisquare test.These parts of the Story-telling assessment form are scored as "Yes-No" and therefore have qualitative value as data.Since in the evaluation made in this section, 3 classes and 2 types of scoring (yes-no) were used, manifold tables (2x3 matrix) were used.For the items where there was a significant difference, a comparison of 1st and 2nd grades, of 1st and 3rd grades and of 2nd and 3rd grades, that is, a manifold matrix (2x2) was set up to determine where the difference was coming from.
The data on story length and the story-telling rate comprised qualitative and continuous data.For the analysis of these variables by class, the Levene Variance Homogeneity Test was employed to understand whether or not the variables displayed normal distribution.Because all of the variables were of homogeneous distribution, the evaluation of the data on story length and the story-telling rate was performed with the Multivariate Analysis of Variance (MANOVA).According to Tabachnick and Fidel (2001), the multivariate analysis of variance, MANOVA, is used for the comparison of the group mean vectors for these variables in studies considering more than one dependent variable.In the case of the variables that revealed significance in the story length and story-telling rate variables, the Scheffe test was employed to find the origin of the difference.

Results
This research was planned to use narrative analysis to assess the story-telling skills of children enrolled in the 1st, 2nd and 3rd grades of elementary school, and to make a comparison in terms of age and gender.The results of the research can be seen below.As seen in Table 1, the differences in the story components were examined and the items that revealed significance were subjected to a paired comparison.Accordingly, there was a difference between the 1st and 3rd grades in favor of the 3rd grade in the determination of time and in the initiating event components, and in the search action and the attempt components.The obstacles encountered was also a component that displayed a significant difference between the 1st and 2nd grades.While no difference was found between the 2nd and 3rd grades, the 2nd grade was more successful in recounting details compared to the first grades.In the consequential event component, the 3rd grade was found to be more successful compared to the 1st and 2nd grades; this difference was significant (p<0.05).A difference by gender was only found in the two story components.It was seen that girls gave more of a description in introducing the characters and the location and in recounting the location of the frog (p<0.05).According to Table 2, no difference was seen between the grades in terms of the story comprehension and errors sections.All of the children (100%) were found to be successful in relating the sequence of events.In relating the obstacles encountered in the story during the narration of the story, a difference in gender was found in favor of the boys (p<0.05).
According to Table 3, the difference between the genders in the evaluation tasks section only emerged in describing the obstacles encountered.The girls were more successful than the boys in the questions asked in this section.If we consider the grades, the 1st grade was less successful than the other grades in naming the story while all of the children in the 2nd and 3rd grades were able to find a name for the story.In naming the characters, there were differences between all the groups and it was found that the higher the grade, the more the children could remember the characters in the story.The first grades were less successful than the other grades in describing the actions and relating the places where they looked for the frog.In describing the obstacles encountered, there was a difference between the 1st and 3rd grades in favor of the 3rd grade (p<0.05).As a result of the Levene Variance Homogeneity Test, because all of the variables were homogeneous (p>0.05), the MANOVA was performed on the scores for story length and the storytelling rate.Accordingly, the results showed that the students displayed significant differences in terms of class level [Wilks Lambda (λ) =0.627.F (15. 103)=1.80.p=0.019<.01].In the case of the variables that revealed significance in the story length and story-telling rate variables, the Scheffe test was used on the variables that were significant by grade in order to find the origin of the differences.It was found that the group that caused the differences in the number of C units, vocabulary production speed and in the distinct vocabulary production speed was the 1st grade.The mean of the first grade's number of C units was found to be lower than that of the 2nd and 3rd grades.There were no differences between the second and third grades for these items.The difference for the number of simple utterances derived from the difference between the 1st and 2nd grade.It was found that the number of simple utterances was higher in the 1st grade compared to the second grade.Of the other variables, the total word count, the number of distinct vocabulary, the distinct vocabulary rate, the number of total utterances, the number of completely understandable utterances, the rate of utterances with distinct meaning, the number of words repeated, the word repeat rate, time (seconds), utterance production speed and rate of production of utterances with distinct meaning showed no differences between the classes.It is seen in Table 3 that in the evaluation tasks section, the only difference between the genders was exhibited in describing the obstacles encountered.The girls were more successful than the boys in the questions asked about the story in this section.If we consider the grades, the 1st grade was less successful than the other grades in naming the story while all of the children in the 2nd and 3rd grades were able to find a name for the story.In naming the characters, there were differences between all the groups and it was found that the higher the grade, the more the children could remember the characters in the story.
In describing the actions and relating the places where they looked for the frog, the first grade was less successful in this than the other classes.In describing the obstacles encountered, there was a difference between the 1st and 3rd grades in favor of the 3rd grade (p<0.05).Because all of the variables were of homogeneous distribution, the evaluation of the data on story length and the story-telling rate was performed with the Multivariate Analysis of Variance (MANOVA).The results of the MANOVA reveal that there was a significant difference between girls and boys [Wilks Lambda (λ) =0.875.F (15. 104)=0.98.p=0.474>.01].Accordingly, it was seen that in terms of the variables in the story, that is, the number of simple utterances (F (1.118)= 5.799.p=0.018<0.05),number of total utterances (F (1.118)= 5.429.p=0.022<0.05)and the number of completely understandable utterances (F (1.118)= 6.583.p=0.012<0.05),there was a difference of 0.05 in favor of the girls.After a look into the other variables, it can be said that there were no significant differences between girls and boys in the items on number of C units, total word count, distinct word count, distinct vocabulary rate, rate of utterances with distinct meaning, number of words repeated, repeated word rate, time (seconds), utterance production speed, rate of production of utterances with distinct meaning, vocabulary production speed and distinct vocabulary production speed (p<0.05).

Discussion, Conclusion and Suggestions
In this study that assessed the story-telling skills of children attending the 1st, 2nd and 3rd grades of elementary school, the story components were examined in terms of the grades and gender and a paired comparison was made of the items that proved to be significant.Accordingly, it was seen that the 3rd grade students were more successful with their descriptions than the children in the 1st and 2nd grades in the several story components.
One of the studies that support the present research is a work by Wigglesworth (1991).This study used the story "Frog, Where are You" with children in the 4, 6, 8, 10 age group and it was found that in the story's setting component, the character that was most described by children of all ages was the boy in the story; it was seen that a description of the dog was generally not made and in this, the children of younger ages made less of an attempt to describe the dog.Cameron and Wang (1999) saw in their study with children ages 4-6 and 8 that the statements offered in the setting section of the story increased with age.Again, Roth and Spekman (1986) and John (2001) found that there were differences by age in the number of events described in the story and that the children in the older age group were more successful in describing the events in the story compared to the younger ones.
Stories have a setting, characters, the element of time, a description of the setting, an initiating event of the story, an internal response and a plan/purpose section.The setting section in general discerns by and large the main framework of the story.This section is where the planning of the story takes place; it is the place where the main components of the story are described.For this reason, it can be said that age plays a role in the performance of these kinds of skills and that organization skills also increase with age.
A difference by gender was only found in the setting part of the story.In this, it was seen that girls gave more of a description in introducing the characters and the place and recounting the situation of the frog.Greenhalgh and Strong (2001) state in their study that a group of children, ages 7-10, displayed a difference in their story-telling skills according to gender.These results are similar to those gleaned from the present research.Similarly, Gow (1986) in a longitudinal study, found that boys were better at certain language skills such as the number of words used and using the language meaningfully.It was however found that girls were more successful in basic language skills and for this reason exhibited better reading skills.John, Lui and Tannock (2003), in their study, showed that there was no difference in terms of gender in the story components or in story comprehension but that girls were better achievers in the components on questions about the interpretation of the story.Another study by Horgan (1976) reported that the language utterance production of girls was greater than boys and that they made fewer mistakes when speaking.
Research results show that girls are generally higher achievers in the initial story components because of their skills in creating and organizing a basic framework for a story.The study, besides bringing out certain details in the story, did not detect any significant difference between the genders.Girls were more skillful in expressing details compared to the boys.It can be said that in general, girls' oral vocabularies were better.
According to the story comprehension and error sections, no difference was seen between the grades.All of the children were found to be successful in relating the sequence of events.Reaping similar results to the present research, Greenhalgh and Strong (2001) and Kaderavek, Gillam, Ukrainetz, Justice & Eisenberg (2004) asserted that there was no difference in story-telling skills according to gender.In the study by John (2001), it was set forth that making mistakes in relating the sequence of events diminished with time.According to the results of the research, it was found that children in this particular age group did not make sequence mistakes when relating the story drawn from a picture book.It was also noted that only a few children made the mistakes of straying from the topic, changing or providing a different interpretation.In this section, a difference in favor of the boys was revealed for providing detailed information during the task of relating the story.Kaderavek, Gillam, Ukrainetz, Justice & Eisenberg (2004) reported that among children in the 5-11 age group, girls were more successful at producing stories than boys.John, Lui and Tannock (2003) stated in their study that there was no difference between the genders in story comprehension.In the study, it can be said that girls provided more detailed information in the setting section but boys were more focused on the details in the attempt section.
In the section that comprised the evaluation tasks related to the story, it was again seen that success in relating the details of the story increased with age.It was found that the 7 year-olds were less successful than the children in the other age groups.John, Lui and Tannock (2003) asserted in their study that the level of response to questions about a story increased with age.These results support the results of the present study.
In the evaluation tasks section, the difference between the genders was only seen in the obstacles encountered.Again, girls were more successful than the boys in providing information with details.In a similar study, John (2001) asserted that girls were more successful in answering the interpretation questions.These results are consistent with those of the present study.
The study considered the children's language skills from a stylistic point of view.Story length and story-telling rates were compared by age.In this, it was found that there were differences in terms of the number of simple utterances, the number of complex sentences (the C unit), vocabulary production speed and distinct vocabulary production speed when compared by age.Greenhalgh and Strong (2001) stated in their study that there was an increase in the number of complex sentences as the child got older but that this increase did not represent a significant difference for children in the 7-10 age group.Contrary to this finding, İpek and Bilgin (2007) found a significant difference in their study between the 1st and 3rd grades in terms of vocabulary.The higher the grade, the more were the number of words in the children's vocabulary.Acarlar (2005) has reported an increase in the use of words as children grow.Also, Roth and Spekman (1986) stated that the use of sentences with conjunctions increased with age.Sturm and Seery (2007) found in their research that when they looked at the talking speed of children in the 7-11 age group, they saw that the 7 year-olds spoke more slowly than the children 9 and 11 and that the total number of words used, the vocabulary and utterance production speeds were also slower.These findings are consistent with those of the present study.In terms of language structure, it is said that children's use of sentences with conjunctions and their talking speeds develop as they grow older.
Girls were found to be more successful in the variables of story length and story-telling speed, number of simple utterances, total number of utterances and in the number of completely understandable utterances.Looking into the other variables did not reveal any difference between the genders.
There are not many studies that have examined the difference between the genders in terms of story-telling rate and story length.John, Lui and Tannock (2003) found in their study that story-telling rate and story length or the number of words used did not exhibit a difference by gender.Sturm and Seery (2007) however found in their research that when they looked at the talking speed of children in the 7-11 age group, they saw that girls spoke faster and the number of words in their vocabulary was greater.
According to the results of the research, it can be said that there is no difference between boys and girls in terms of talking speed.It can also be said that the fact that the girls were better in story length only in terms of the number of utterances and the number of completely understandable utterances may be explained by the fact that girls' verbal language production is greater.
When we look at the results of the study, it may be said that story-telling skills develop with age.It was found that in terms of comprehending the story, children were generally similarly successful by age and gender.It was also found that detailed descriptions increased with age and that girls were more successful in the setting part of the story and boys in the development part.In general, it can be said that language skills do not exhibit large differences based on age and gender in the 1st, 2nd and 3rd grades.Bamberg (1977a) has examined many approaches to narrative analysis and has asserted that this analysis should be used in research for educational and clinical purposes.This is because stories do not only reflect language structure in individuals but also encompass cognitive structure.They are of great importance in determining academic skills in educational environments.It has been stated that pre-school is the predictor of the reading abilities of children in 1st and 2nd grade and also the predictor of reading comprehension skills in 3rd grade students (Akt.Silliman and Champion, 2002).For this reason, narrative analysis tools should be frequently used for educational purposes.It is important however that the tools used are suitable for the cultural setting they are to be used in.This method should be used particularly with pre-school and elementary school children to determine the characteristics of their language and cognitive development and their developmental needs.Scientific studies should be carried out in this area and educational revisions should be made in the light of the results of these studies.

Figure 1 .
Analysis of the Finalized Instrument

Table 1 .
Chi-square Results of the Narrative Skills related to the Story's Setting, Attempt and Consequence Components

Table 2 .
Results of Story Comprehension and Errors Sections

Table 3 .
Chi-square results of the Section on the Evaluation Tasks

Table 4 .
Scheff Test Results in the Comparison of Story Length and Story-telling Rate by Grades

Table 5 .
MANOVA Results of the Significant Items in the Comparison of Story Length and Storytelling Rate by Gender