Список использованной литературы
1. Программа курсов повышения квалификации педагогических работников РК. Третий
(базовый уровень). – 3-е изд. – ЦПМ АОО НИШ Астана: 2013, с.151
2. Модель оценивания тренера уровневых курсов. – ЦПМ АОО НИШ Астана: 2013, с.
65
3. Фрумин И. Оценка качества образования: между контролем и поддержкой. Ж-л.
Первое сентября – 2000, с.34
4. Балакирев В., Кетхудов Р., Бикег Сэмюэл, Боровых А. Новые тендеции в оценке
развития. – 2000, с.34
5. Бахмутский А. Оценка качества школьного образования. Автореферат – 2004
6. Загводкин В. Традиционные и современные формы оценки учителя и их эффективность.
Ростов-на-Дону, 1999, с.3
7. Практическое руководство для учителей. Формирующее оценивание. Оценивание
для обучения.2011, с.4
8. Материалы конференции «Инновационная Оценка в 21 веке» Оценивание в системе
повышения квалификации педагогических работников Казахстана, Мусарова В. 2014
9. Материалы конференции «Инновационная Оценка в 21 веке» Теоретические основы
изучения оценочной деятельности педагога В. Брожик, Р.Г. Кетхудов, А.А. Ивин, 2014
IMPROVING WRITING OUTCOMES AT GRADE 11: AN
COLLABORATIVE
ACTION RESEA RCH
Victor Avasi, Nagima Sarsenbayeva,
Nazarbayev intellectual school of physics and mathematics,
(NIS PhM), Taraz, KAZAKHSTAN
Keywords: process writing, feedback, teacher-student interaction, writing practice.
Ключевые слова: письмо как процесс, обратная связь, взаимодействие между учителем и
учеником, писменная практика
Түйінді сөздер: жазу-үрдіс ретінде, кері байланыс, мұғалім мен оқушы арасындағы
қарым- қатынас, жазу дағдыларын қалыптастыру тәжірибесі
Abstract
This study set out to contribute to the improvement of students writing outcomes at Grade 11
in NIS PhM Taraz. The study population consisted of grade 11 students at NIS. 23 students were
assigned naturally to the experimental group and another 23 to the control group. In January 2015, a
pretest measure was administered to both groups under similar conditions. The experimental group
was exposed to process writing for two months. Teacher mentoring and coaching lasted for four
months. The intervention took place in a partially controlled environment. In May 2015 a post-test
was administered on both groups. The hypotheses generated were tested with reliability-corrected
ANCOVA. There was a significant effect of the intervention on student’s writing outcomes after
controlling for the effect of pre-existing differences between the experimental and control groups, F(1,
39) =45.47, p=0.0005. Gender, Language_group and their combinations had no effect on students’
writing outcomes.
Это исследование было проведено с целью внесение вклада в улучшение результатов
14
5
навыков письма у учащихся 11 класса НИШ ФМН г. Тараза. В состав исследуемых групп
вошли учащиеся 11 классов. Эксперементальная группа состояла из 23 учащихся, количество
учеников контрольной группы тоже составило 23 ученика. На начальном этапе исследования
в действии, в январе 2015 г. в обеих группах были проведены оценивание навыков письма
в идентичных условиях, которое позволило определить стартовые знания учеников. В
экспериментальной группе в течение двух месяцев велись работы по улучшению навыков
письма. В течение четырех месяцев проводился коучинг и менторинг учителя. Исследование
проводилось в частично контролируемой среде. В мае 2015 года было проведено финальное
оценивание в обеих группах. После того как, все предложенные гипотезы были проверены на
надежность, был применен ковариационный анализ (ANCOVA). Даже после проверки влияния
уже существующих различий между экспериментальной и контрольной групп можем сказать,
что исследование дало значительные результаты, F (1, 39) = 45,47, р = 0,0005. Гендерные
различия, родной язык и их комбинации не влияют на результаты письменных работ учеников.
Бұл зерттеу Тараз қаласындағы ФМБ НЗМ-де 11 сынып оқушыларының жазу дағдыларын
жетілдіру мақсатында өткізілді. Зерттеуге 11 сынып оқушылары қатысты. Эксперименттік
тобы да, бақылау тобы да 23оқушыдан құралды . 2015 жылдың қаңтарында оқушылармен іс-
әрекеттегі зерттеуді өткізуден бұрын, жазу дағдыларын анықтайтын бағалау жүргізілді. Бағалау
екі топта да бірдей жағдайларда өткізілді. Эксперименттік топта екі ай бойы жазу дағдыларын
жақсарту шаралары жүргізілді. Сабақ жүргізетін мұғалімге коучинг және менторинг өткізілді.
Зерттеу жартылай бақыланатын ортада өткізілді. 2015 жылдың мамыр айында екі топта да
қорытынды бағалау өткізілді. Барлық болжамдар сенімділік талдауынан өткеннен кейін,
коварияциялы (ANCOVA) талдауы пайдаланылды. Эксперименттік және бақылау топтары
арасында айырмашылықтар бақылаудан өткеннен кейін де, зерттеу айтарлықтай нәтижелерге
жетті деп айтуға болады F (1, 39) 45.47, р = 0,0005. Оқушылардың гендерлік айырмашылықтары,
сондай- ақ қазақ немесе орыс сыныбында оқуы, олардың ағылшын тілінен жазатын жазба
жұмыстарының нәтижелеріне әсер етпейтіндігі анықталды.
2. Introduction
‘The State program of education development in the Republic of Kazakhstan’ June 29, 2011
№ 110, Kazakhstan aims to increase the share of country’s population that speaks English to 10% by
2014, to 15% by 2017, to 20% by 2020, and to increase of the share of population that speaks three
languages (Kazakh, Russian and English) to 10% by 2014, to 12% by 2017, to 15% by 2020 [32,
p.1]. NIS schools are created under the national strategic framework, Kazakhstan 2030 [32, p.29-31].
Writing is part of the curriculum offerings of formal school [14, p.8-9; 15, p.3]. Prior to the study,
writing outcomes were not as high as expected. In the IELTS exam of December 2014, students
performed dismally in writing. In summative assessments, writing outcomes are not always as high as
they ought to be and writing can be dismal. If this problem was not addressed students would continue
to rely on external help since they relegate self-regulated learning and manifest impaired executive
functioning [27, p.95; 4, p.4-6]. The intervention was approached from a change management
perspective concerning the student, English language teacher and their interaction. Using appreciative
inquiry student-centered educational practices that lead to optimal outcomes in writing at grade 11
were considered and a desirable scenario developed [5, p.20, 10, p.5]. A healthy student-teacher
interaction and writing practice as the conditions that improve writing outcomes was conceptualized.
Extrinsic and intrinsic factors related to the student might explain why writing outcomes are as they
are. One key extrinsic factor is teachers as they affect students more than anything or anyone else
[8, p.1]. Students perform better in schools where they are engaged inclusively [29, p.32, 166-164;
22, p.339]. Teacher support refers to students’ beliefs that their teachers care about them, and value
15
and establish personal relationships with them [28, p.5; 30, p.456]. Teacher support leads to higher
attendance and test scores [18, p.9]. A study to improve writing outcomes has to involve the teacher
at the center [22, p.339]. Intrinsic factors within the student drive students’ achievement motivation
[19, p.12, 14], and explain how they approach set tasks including essay writing [26, p.146; 22, p.349].
Concerning intrinsic factors, satisficing was detected in students’ essay writings. Satisficing
explains the tendency to select the first option that meets a given acceptability threshold when one
has to make a choice [34, p.1178]. Satisficing as a strategy conserves energy and resources; settles
for what is good enough, but it can lead to sub-optimality. It happens to the detriment of mastering
good writing, internalizing the skill and intrinsic motivation [19, p.12, 14; 22, p.349]. Satisficing
gradually proceeds to maximizing in repeat situations of choice making [34, p.1178]. This can be
achieved through the unfreeze-change-refreeze model developed by Kurt Lewin. The status quo of
the individual has to be challenged to create an opportunity for the desired change to be introduced
and sold to the person or people in the situation. The process of unfreezing disconfirms a person’s
incumbent belief system [31, p.5]. The beliefs, attitudes, decisions and actions [22, p.293] of students
in grade 11 can be challenged in this way. From this point the teacher can scaffold for the learners
to create the necessary change [22, p.317-318; 24, p.308]. Indirect psychosocial support though
modelling to students [3, p.197], exercise in writing essays [1, p.3, 8, 9] and new programming
[31, p.5] can be applied to cause refreezing into the new culture of writing essays thoughtfully and
skillfully.
Listening and speaking arises naturally in a normally developing child but reading and writing
have to be taught [1, p.2]. If teachers make assumptions about their students, learning outcomes can
be suboptimal. Students learn to write well when the teacher’s expectations are clearly communicated
to them [20, p.159; 39, p.212-216]. Successful teachers communicate high standards and adhere to
them even in the face of opposition, accepting students’ best efforts only [20, p.173]. One expectation
is that learners know how to address their audience appropriately when they write. Writing should be
shaped according to what the audience needs and expects [17, p.15-17]. How to skillfully address an
audience can be developed through the process approach [1, p.6, 8, 13]. Ideas in an essay have to be
presented in an orderly way [29, p.97-98; 38, p.117-120].
Students learn to write by writing [12, p.4]. Since writing is an outcome of an internal thought
process (ibid.), long essays should provide good diagnosis of the intrinsic factors such as self-
efficacy; maturity, and creativity [22, p.316, 349, 351-352] of each student in relation to writing.
A writing intervention should have a core that applies to all learners while allowing for customized
intervention for each individual [22, p.339, 25, p.5]. The process approach to teaching writing is a
multi-stage process including prewriting or inventive actions; drafting; getting feedback from peers
or the instructor; revising globally then by paragraph, proofreading, and creating a final text within
a set timeframe [35, p.151-155]. Using the process approach, students see how they are progressing
against known success criteria and work deliberately towards attaining them. The success criteria for
English at grade 11 under the NIS system are to: plan, write, edit and proofread work independently;
write with grammatical accuracy; use style and register to achieve an appropriate degree of formality;
develop coherent arguments supported with examples and reasons, and independently use appropriate
layout at text level on a wide range of general and curricular topics [14, p.8, 9; 25, p. 10-24].
Grade 12 students in NIS framework take the IELTS language proficiency in December as
part of their university entrance requirement. For example Nazarbayev University requires an IELTS
minimum of 6.5 for admission to The Schools of Medicine [41, p.1]. The IELTS performance bands
run up from 0 for no attempt, to 9 for an expert user with full command of the language [13, p.6; 16,
p.4; 41, p.1]. The Academic module has two writing tasks that take up 60 minutes (ibid.). CIE A- and
AS-level examinations such as 8693_ms_2006 allocate one hour per question and students write 600-
900 words. In process writing prewriting precedes actual writing [37, p.1]. One tool for prewriting is
16
5
mind-mapping. Students can benefit from using mind maps in planning their essays. Students learn
to plan and organize their writing by developing and using a mind map [6, p.12]. Each node on the
mind map can be used to constitute the paragraphs of the essay. Each node can be expanded into sub-
nodes from which paragraphs can be explained meaningfully (ibid.). Paragraphs should be developed
systematically with a narrow topic sentence and at least three supporting sentences. They posit that, a
well-developed paragraph includes at least three supporting details or examples [7, p.70]. Other than
ideas, paragraphs consist of applied language structure and syntax. Some practitioners advocate for
focusing more on teaching students to communicate than on teaching grammar [2, p.3;11, p.261, 265;
40, p.424]. However it is not clear what the contextual implications of such studies are where English
is a second- or third language. Nonetheless students in Kazakhstan could benefit from more teaching
of grammar rules and applications across the key stages [29, p.3, 7].
3. Subject of the study
The intervention in this study would help students with their specific grammatical needs.
Following the pretest, an intervention, based on the conceptual framework, was administered to the
experimental group. The following research questions were raised based: What effect would the
intervention have on the writing outcomes of grade 11 students? What effect would gender have
on the writing outcomes of grade 11 students? What effect would Language_grouping have on the
writing outcomes of grade 11 students? The following null- and alternative hypotheses were tested
at a 0.05 level of significance: H01: The intervention would have no significant effect on outcomes
in essay writing. H11: The intervention would have a significant effect on outcomes in essay writing.
H02: Gender would have a significant effect on outcomes in essay writing. H12: Gender would not
have a significant effect on outcomes in essay writing.
H03: Language_grouping would have a significant effect on outcomes in essay writing. H13:
Language_grouping would not have a significant effect on outcomes in essay writing.
4. Research Methods
4.1 Sample
The study population consisted of grade 11 students at NIS PhM Taraz. A sample of 46 [31%] of
the students were naturally assigned to the study. 23 students were assigned to the experimental group
and control group respectively. This sampling strategy was chosen in order to study the effect of an
intervention on a smaller group in one school prior to scaling up to a larger population and because of
logistical limitations [35, p.290].
4.2 Measures
The unit of analysis was a student. Pre-validated questions from CIE A and AS level examinations
were administered. These questions are used globally for qualification and certification and are
standard with proven face-, concurrent- and predictive validity. The baseline question was: ‘If you
were granted three wishes, what would they be – and why? [42, p.2]’. The post-test question was:
‘Cosmetic surgery does more harm than good.’ What do you think – and why? [43, p.2].
The essay-grading standards at NIS PhM Taraz include content, organization, expression
(communicative approach) and mechanics (language) and they are used internationally to grade
students’ essays at various levels [1, p.2]. From the standards, 16 criteria were adopted and expanded
for greater precision in the study to minimize subjectivity in checking, acknowledge achievement
appropriately and maximize intervention on specific weaknesses of each student. They were sorted by
essay-grading standards as follows. Content: Number of words; Self-efficacy; Maturity, and Creativity.
Organization: Use of a margin; Three-phase structure, paragraphs and Logical Flow. Communicative
17
Approach: Understandable; Minimal errors; Minimal hanging phrases, and Minimal slang Language:
Tenses; Articles; Conjunctions and Punctuation. Students’ essays were checked using the 16 success
criteria as Likert items. The success criteria were assigned scores based on a 3-point Likert scale.
Each student could score 3 or 2 or 1 depending on the quality of their work on a particular criterion.
The best score on each criterion was 3 giving a total possible score of 48 marks. The least total score
possible was 18 from achieving one point per criterion. All essay scripts were marked by one grader
to eliminate inter-rater differences [36, p.268-269]. Posttest essay scripts were crosschecked using
the 16 Likert items by an independent grader, for objective marking and confirmation of achievement
trend by students in the sample.
4.3 Design
According to [9, p.23], a design structures a research, showing how the samples or groups,
measures, treatments or programs, and methods of assignment coordinate to address the central
research questions. A Non-Equivalent Groups Design (NEGD) was used in this study. [36, p.105]
identifies NEGD as the most frequently used design in social research. It is structured like a pretest-
posttest randomized experiment without random assignment. The NEGD is susceptible to the internal
validity threats of selection. Differences between the groups at pretest tend to affect the outcome of
the study such that causality is not correctly attributed. Language_grouping of students and gender
mixes of the study groups were potential confounders. Consequently the data was disaggregated by
Language_grouping and gender. Leakage of the intervention to teachers and to students in the control
group was prevented as far as was possible. All students at NIS PhM Taraz study on government
scholarship which creates some equivalence between all students at school.
4.4 Procedure
Validated questions were administered to both the experimental and control group at the pretest
in January 2015 and at the posttest in May 2015. The conditions of the test were controlled for
uniformity. The intervention was two-pronged focusing on both the line teacher of English and the
experimental group. Attention was devoted to assess the status quo of classroom practice. Face-to-
face discussions were held to get the teacher’s self-report of teaching and learning. Mentoring and
coaching sessions for the teacher were held. Discussions of intervention areas for writing to improve
were held. PSPP statistical software (Build 0.8.4-g267362) was used to conduct factor analysis on
the 16 success criteria to reveal the underlying factors, which were then used as inputs to guide the
design of the intervention. A grammar intervention was developed from the needs assessment. The
environmental setting and how it facilitated writing in English was examined.
The performance of the experimental group was tracked. Each participating form group wrote
one essay per week and received feedback in the same week. The intervention was intensive on the
writing process [1, p.5-6] and the mental software which is crucial for writing. The CIE writing model
of essays of 600-900 words was adopted. This was in line with the theories of zone of proximal
development [ZPD] and assessment of learning (AfL) [22, p.308, 317-318, 346, 379; 23, p.308-311;
24, p.231-233]. The rational was that with a line density of eight (80 words per line), the word count
would be at 640 words in the allocated time. While 640 words was close to the lower limit, it was
considerably high for students who had grown accustomed to writing only 250 words per essay. It
was necessary to actively avoid creating a sense of frustration among the grade 11 students in case
they worried about too high a word count to be attained [1, p.3, 5]. 115 essays written by students in
the experimental group were checked and annotated in detail to highlight the aspects that a student
needed to do extra work on to improve. The teacher focused on providing constructive feedback and
motivating the students to write. Keeping the written essays in a file, locked away, minimized leakage.
18
5
4.5 Analysis
Data were captured using MSEXCEL (2010). The same program was used to determine the
reliability of the pretest and posttest data given the non-random assignment of the study group [36,
p.83]. Cronbach’s alpha and test-retest reliability coefficients were generated and data adjusted.
Scatter plots for data from both adjustments were generated. The same software was used to generate
graphs of reliability-adjusted means at pretest and posttest. The program SPSS was used to conduct
Levene’s test for homogeneity of variances when the posttest scores were regressed against adjusted
pretest data, gender and Language_grouping as fixed factors. SPSS was used to conduct reliability-
corrected ANCOVA [36, p.78-85].
5. Results
The Cronbach’s alpha for the experimental group was 0.93 while the test-retest coefficient was
0.89. The Cronbach’s alpha for the control group was 0.75 while the test-retest reliability coefficient
was 0.65. The adjustment of the pretest scores for the experimental group was not different from their
raw scores. The adjustment for the pretest scores of the control group using both coefficients visually
demonstrated homoscedasticity for both adjustment factors. A scatter plot of the reliability adjusted
data was created using the test-retest factors. The graph exhibited a crossover pattern. Levene’s test
[21, p.99, 100] verified the equality of variances in the samples (p>.05). The equality of variances
between the experimental and control groups made it possible to further test other conditions for using
ANCOVA to analyze the data gathered during the study. The between-subject factors were Group,
Gender and Language_grouping. Group had two coded variables, Control (N=23) and Experiment
(N=23). Gender had two coded variables, Female (N=31) and Male (N=15). Language had two coded
variables, Kazakh (N=34) and Russian (N=12). Tests of Between-Subjects showed that the adjusted
pretest scores and the group were significant at the 0.0005 level. Gender, Language_grouping and
their combinations were not significant. An ANCOVA (between-subjects factor: Pretest (Experiment,
Control); covariates: Group, Gender and Language_group) revealed a main effects of Pretest, F(1,
39) = 45.47, p = .000, η
p
2
= .538, and Group, F(1, 39) = 36.92, p = .000, η
p
2
= .486. The model used
explained almost 70% (R
2
=0.699, adjusted R
2
=0.653) of all the variability of the response data around
its mean of 38.351 for the experimental group. The mean difference between the Experiment and
Control group at a 95% Confidence Interval was 7.029 with a standard error of 1.096. The mean
difference was significant at the .05 level.
The hypotheses generated were tested with reliability-corrected ANCOVA and the decision was
taken at a 0.05 level of significance. There was a significant effect of the intervention on student’s
writing outcomes after controlling for the effect of pre-existing differences between the experimental
and control groups, F(1, 39) =45.47, p=0.0005. On the basis of the result the null hypothesis H01 was
rejected. The alternative hypothesis could not be rejected. H11: The intervention had a significant
effect on outcomes in essay writing of the students in the experimental group in essay writing.
There was no significant effect of Gender on student’s writing outcomes, Gender, F (1, 39) = .102,
p=.75. On the basis of this result, the null hypothesis, H02: Gender would have a significant effect
on outcomes in essay writing, was rejected. The alternative hypothesis could not be rejected. H12:
Gender would not have a significant effect on outcomes in essay writing. There was no significant
effect of Language_grouping on student’s writing outcomes, Language_group, F (1, 39) = 1.426, p =
.24. On the basis of this result, the null hypothesis, H03: Language_grouping would have a significant
effect on outcomes in essay writing, was rejected. The alternative hypothesis could not be rejected.
H13: Language_grouping would not have a significant effect on outcomes in essay writing.
Достарыңызбен бөлісу: |