A Content Analysis on Articles Related to Reading Skills

This study aims to examine the articles related to the reading skills written between 1990 and 2013. In line with this general purpose, articles related to the reading skills were examined published in twenty-nine national journals in the field of educational sciences and indexed by SSCI (f=5), ULAKBIM (f=13) and other (f=11) databases in Turkey. 232 articles were reached in total and examined using the “Reading Skills Papers Classification Form” (RSPCF) in accordance with the subjects, years, number of authors, universities, number of hypotheses, research methods and patterns, data gathering tools, samples, and data analysis methods. The data gained from this research was coded by using the SPSS 17.0 software package and analyzed with the content analysis method. Frequencies and percentages were used as descriptive statistical methods for analyzing the data. Results of this research revealed that most of the articles (48; 20.6%) were written about reading comprehension; articles usually had a single author (59.9%) and a hypothesis was tested (39.2%); survey method a non-experimental quantitative research method, was used for the majority (43.5%); attitude and perception tests etc. were the most frequently used data gathering tools (28.1%); middle-school students (28%) were preferred as samples; sample size was usually between 101-300 (27.2%) and the random sampling method (44.4%) was used; and a quantitative descriptive analysis method, frequency/percent (18.3%) method as data the analysis method, was preferred most frequently. Reading skills


Introduction
There are four basic skills areas in native language teaching: reading, writing, listening, and speaking.Two of these skills (reading, listening) are accepted as "receptive skills", and the other two (speaking, writing) are accepted as "productive skills" (Razı, 2008).For a very long time, it has been emphasized that basic language skills areas comprise an entirety, and should not be considered separately.Nevertheless, reading and writing skills have always been prioritized in native language teaching.Listening area integrated into secondary school education in Turkey in 1981.On the other hand speaking area was encompassed in 1949 under the title of "Verbal Expression".However the reading area had been included in the primary school education dated back to 1924 (Temizyürek and Balcı, 2006).In addition, changing world circumstances let people to make the most of their communications through reading.Developed countries which have been aware of such facts have realized that reading area is very important in human model generated by the contemporary world and emphasized a special importance on this issue within their education systems (Coskun, 2002).
"The purpose of reading education and teaching, which has an important place in native language teaching, is to bring up students who can comprehend, interpret and criticize what they read.Having people that speak well, think while listening and write with a creative intelligence in the future is dependent on developing these skills" (Gökçe, 2012, p. 826).
There have been many different definitions on the concept of reading, which has a wide range of meanings from vocalizing written symbols to building-up and interpreting meanings.Akyol (2012, p. 1) defines reading as "a process of establishing a meaning by using preliminary information, based on effective communication between writer and reader, in line with a proper method and purpose, performed in an organized environment".According to Harris and Sipay (1990, p. 10) reading is "the meaningful interpretation of written language".According to Yalçın (2006, p. 47) reading is "the process of reception by sensory organs of special symbols, which were predetermined by people, and their interpretation and evaluation by the brain".Constructive approach considers reading as "an active process in which the individual creates new meanings by integrating his/her preliminary information with the new information."(Güneş, 2009, p. 3).Karatay (2010, p. 459) defined reading as "comprehending the emotion, thought and messages in a written text, beyond vocally or silently decoding known writing characters".
As the basis of learning, reading which is a complex process consisting of the eyes, sounds and various functions of the brain such as sight, perception, vocalizing, understanding, and reconstruction in the mind, is affected by many elements (Güneş, 2009).These elements are divided into two; reader-dependent factors and environmental factors.Reader-dependent factors are sight, hearing, intelligence, language development, neurologic maturity, general movement ability and gender (Özbay, 2011).Environmental factors are family, school, social circle and libraries.The socioeconomic status of the family also affects reading.Families should have their children feel that reading is a necessity, provide books suitable for their age and attention levels and most importantly, have a bookshelf at home.It is a fact that the social structure of the individual, particularly his/her cultural and social environment affects his/her development of intelligence, attitude and behavior.If the friend circle of an individual has an interest in books and reading, this will reflect positively on the individual; if it is the contrary, this will affect reading skills negatively (Temizkan, 2009).Having a library at home, in the classroom, at school or in an easily accessible place may positively contribute to the improvement of reading skills.Libraries must be made attractive for students, and books must be updated to draw interest.One of the most important elements of reading education is the teacher.The teacher has great responsibility from the first reading studies in making reading a habit.Also, because teachers are role models, their attitudes are very important (Özbay, 2011).
In Turkey, reading education is usually carried out by vocalizing the text of a book in sections, in line with its relevant directives.However, reading education is an extensive process, covering issues such as what students must pay attention to while reading a text, what kind of deductions they can make from the text, and how they should act with different types of texts (Topuzkanamış & Maltepe, 2010).
In Turkish Language Lesson Curriculum (6, 7, 8 th Grades), reading skills "help students reach various sources and face new information, incidents, situations and experiences" and it states that these skills include a process that provides learning, researching, interpretation, discussion, and critical thinking (MEB, 2006).
Reading is a skill that can improve and be improved (Akçamete & Güneş, 1992).One of the most important goals of Turkish Language teaching is to teach students the habit and joy of reading.Thus, reading must become a habit beginning in childhood.Özbay (2007) expresses that reading skills improvement and maintenance is based on the foundation created in mandatory education.Therefore, at this stage, families and educators have great duties.
There is a significant correlation between reading skills and other skills (Çeçen, 2012).This relation can occur as skills completing or supporting each other, or as executing the same goal (Arıcı, 2012).It is hard for a student to reach success with a lack of reading skills.Consequently, it must not be forgotten that in a rapidly improving world, the societies that will lead the future will be societies that care about reading.Coşkun, Özçakmak and Balcı (2011) have assessed postgraduate theses written between 1981 and 2010, by classifying them in accordance with their types, years, universities, target groups and subjects.Şahin, Kana and Varışoğlu (2013) performed a study to identify the trends of subjects, methods, data gathering tools, data analysis methods and sample characteristics of postgraduate theses prepared in Turkish Language education between 2000 and 2011.Doğan and Özçakmak (2014) performed a study for evaluating postgraduate theses on listening skills, one of the basic language skills of Turkish Language.Varışoğlu, Şahin and Göktaş (2013) identified the trends of Turkish Language education researches, subjects, methods, data gathering tools, data analysis methods and sample characteristics of theses prepared in between 2000 and 2011 in journals indexed by SSCI and ULAKBIM Social Sciences databases.
Lack of assessment studies for articles on reading skills is an important factor in performing this study.It is thought that specifying the required field of studies with regards to reading area and conducting a content analysis in order to orient the studies towards such field shall be useful.Determining which particulars such as methods, data gathering tools and sample groups, etc., are often used in studies so as to lead the way for future studies related to reading skills are among the aspects making this study important.
In this study, articles related to reading skills were examined in terms of descriptive information on the identity of the articles, their subjects, methods, data gathering tools, sample and data analysis methods; and answers to the following questions are sought: 1. How are the articles related to reading skills, written between 1990 and 2013, distributed according to their subjects?
2. How are the articles related to reading skills, written between 1990 and 2013, distributed according to years?
3. How are the articles related to reading skills, written between 1990 and 2013, distributed according to number of authors?
4. How are the articles related to reading skills, written between 1990 and 2013, distributed according to distribution of authors to institutions/universities? 5. How are the articles related to reading skills, written between 1990 and 2013, distributed according to number of hypotheses?
6. How are the articles related to reading skills, written between 1990 and 2013, distributed according to research methods and patterns?
7. How are the articles related to reading skills, written between 1990 and 2013, distributed according to data gathering tools?
8. How are the articles related to reading skills, written between 1990 and 2013, distributed according to samples? 9. How are the articles related to reading skills, written between 1990 and 2013, distributed according to data analysis methods?

Method
In this research, the content analysis method was used.Used as a qualitative research technique in the field of social sciences; the content analysis is a systematic, repeatable technique summarizing certain words of a text with smaller content categories using codes based on some certain rules (Büyüköztürk et al., 2013)."The main process of the content analysis method is collecting similar data within the framework of certain concepts and themes, and organizing and interpreting them in a conceivable order for the reader" (Yıldırım and Şimşek, 2011, s. 227).Content analysis is a method for examining the literature of a scientific field (Falkingham and Reeves, 1998).Çalık and Sözbilir (2014) considered the content analyses under three titles: meta-analysis, meta-synthesis (thematic content analysis), and descriptive content analysis.This research can be considered under the title of descriptive content analysis because it provided information on the trends related to the studies on reading skills in Turkish Language Education."Descriptive content analysis is a systematic review that aims to identify and describe the general trends and research results in a particular research discipline" (Çalık and Sözbilir, 2014, s. 34).

Population/Sample
The population data of this research consists of twenty-nine national journals published between 1990 and 2013 in the field of educational sciences, and indexed by SSCI (f=5), ULAKBIM (f=13) and other (f=11) databases in Turkey.Studies published in these periodicals were specialized in the field of reading.Since the department of Turkish Language Education commenced in Turkey with the master and doctorate programs had been opened between the years of 1989-1990(Güzel, 2003)), the research on Turkish language education has intensified.Hence, the interval of 1990-2013 was selected for the research.Limited focus was set on "Turkish Education" and the articles on "reading skills" (Aytaş, 2005;Güneş, 2009;Okur, 2013;Razı, 2008;Sallabaş, 2008), which are considered to be important in the field, were examined in the research.In this research, issues of these journals, which are available online, have been examined.This research is limited regarding such feature thereof.Randomly selected journals were examined and then, the articles on reading skills were gathered from these journals to create an article pool.The sample of study consist of 232 articles on reading skills, selected by the purposeful sampling method.Names of the journals reviewed, the publication index, the interval of years in which the articles have been published and the numbers of articles on reading skills are given in Appendix 1.
Accordingly, among the journals that published Turkish language education research; Turkish Studies (f=35), Journal of National Education (f=21), Educational Sciences: Theory & Practice (f=18), Eurasian Journal of Educational Research (f=15) and e-Journal of New World Sciences Academy (f=15) are the journals with the highest numbers of publications on reading skills.

Data Gathering Tool
In this study, "Reading Skills Papers Classification Form" (RSPCF) was used.The researchers developed the form CRSBF which was developed by Sözbilir and Kutu (2008) by reorganizing it in compliance with the objectives and content of the study.The RSPCF consists of six sections including the descriptive information about identity of the article, issues, methods and design, data collection tools, methods of sampling, and data analysis.For the validity of the form, the opinions of experts from the fields of Turkish Education, Educational Sciences and Linguistics were gathered.In line with the received comments, a draft form was created.A pilot study of the draft form was applied on a group of randomly selected ten articles.Possible reasons of the disputes among the raters were determined and the specimen rating forms were compared; deficiencies and weaknesses in the classifications were identified.Sample assessment forms were compared to determine the incomplete and weak points in the classifications.The final shape was given to the form by resolving the problems occured during the creation of the sections and classification of data in line with the consensus of the experts.

Data Analysis
In this study, articles related to the reading skills, which were examined in twenty-nine national journals published in the field of educational sciences and indexed between 1990 and 2013 by the SSCI (f=5), ULAKBİM (f=13) and other (f=11) databases in Turkey, have been classified and analyzed under nine titles.During the analysis, each article was examined and classified according to their subjects, years, number of authors, universities, number of hypotheses, research methods and patterns, data gathering tools, samples, and data analysis methods.The data were coded by using SPSS 17.0.First, data related to the article identity information (name, author(s), year, journal, volume published, issue number, institution(s) of the author(s), number of hypotheses etc.) were entered and then; their subjects, methods, data gathering tools, sampling and data analysis methods were recorded.While performing these stages, it is understood that, there was not sufficient descriptive information on the subjects, methods, data gathering tools and data analysis methods particularly in some of the articles.This situation has been discussed between the researchers and coded upon a consensus.Frequencies and percentages were used as the descriptive statistical methods for analyzing the data coded with the SPSS.

Results
The quantitative results reached by the analysis of articles on reading skills published in twenty-nine national journals between 1990 and 2013 in the field of educational sciences, indexed by SSCI (f=5), ULAKBIM (f=13) and other (f=11) databases in Turkey, were examined under 9 subjects as represented in the following sections.

Distribution by Years
While examining the distribution according to years, range of time was kept wide due to the lack of articles between 1990 and 2000.11 articles were published between 1990 and 2000, and 18 articles were published between 2001 and 2005.Starting from 2006, a great increase is seen in the number of articles, therefore, a separate classification was made for each year, beginning from 2006.Of the total 232 articles on reading skills, 52 were published in SSCI covered journals.Distribution of the articles on reading skills per year is given in Table 2.

Number of Authors
Among analyzed articles, it was observed that studies with a single author (59.9%) or two authors (29.3%) were frequently preferred in publications related to the reading skills, while studies with four (0.9%) or more authors were not preferred.Distribution of the articles per their number of authors is given in Table 3.

Distribution of the Authors by Institutions/Universities
While classifying articles according to the institutions/universities of the authors, the fact that some articles with multiple authors was taken into account.Names of the institutions/universities that published five or more articles are included in the table, while the institutions/universities that published less than five articles are evaluated under the category "other".Based on the distribution of articles according to the institutions/universities, it is seen that the most studies on reading skills were performed by Gazi University (14.7%), the Ministry of National Education (9.7%), Anadolu University (6.4%) and Ankara University (5.8%).The distribution of articles on reading skills per the institutions/universities is given in Table 4.

Number of Hypotheses
As a result of the analysis, it is understood that the researchers usually used one hypothesis (f=91) in the articles.It is seen that there is a significant number of articles with no hypothesis (f=21).Distribution of articles according to the number of hypotheses is given in Table 5.

Research Methods and Patterns
It is revealed that the researchers most frequently preferred the survey method (43.5%) among non-experimental quantitative research methods.This was followed by the experimental method (17.2%), descriptive method (4.7%) and case method (4.3%).It is seen that the quantitative research (71.5%) was preferred over the qualitative research (12%) by the researchers.Survey method is the most preferred (43.5%) method among quantitative researches while the case studies are the most preferred (4.3%) method among the qualitative researches.
Another salient point is that there are plenty of literature reviews.Out of the articles, 13.4% (f=31) are literature reviews.Distribution of the articles according to the research methods and patterns is given in Table 6.

Data Gathering Tools
Among the articles related to the reading skills, it is seen that the attitude tests, perception tests etc. are the most used (28.1%)data gathering tools.Observation is the least preferred data collection mean by the researchers.Usage ratios of other data gathering tools are close to each other.Distribution of the articles on reading skills according to the data gathering tools is given in Table 7.

Sample
In this section, the articles are analyzed in terms of the sample level, sample numbers and sample selection method.Regarding sample level, middle-school students formed the greatest sample group with a ratio of 28% (f=65).After this, the sample is formed of university students (22%, f=51) and primary school students (14.2%, f=33).Distribution of the articles on reading skills per sample level is given in Table 8.When the sample numbers of articles on reading skills are examined, it is seen that the sample numbers between 101 and 300 are used most frequently (27.2%).This is followed by the sample numbers of 31-100 (20.7%) and 301-1000 (15.9%).It is seen that the sample numbers higher than 1000 is the least preferred sample number (2.2%).Another salient result is the high number of articles with undefined sample numbers [f=55(23.7%)].Distribution of the articles on reading skills per sample numbers is given in Table 9.It is seen that the random sampling is the most (44.4%) common method among the analyzed articles.This is followed by the convenience (18.1%), and purposeful sampling methods (15.9%).It is seen that the studies covering the whole population are very rare (1.3%) in sampling method selections.It was found that other sampling methods have a high ratio (20.3%).Distribution of the articles on reading skills per the sampling method is given in Table 10.

Data Analysis Methods
When the analysis results were examined, it was seen that the frequency/percentage method is the most used method (18.3%) among the quantitative descriptive analysis methods.This is followed by the t-test (17.6%), descriptive analysis (13.8%), mean/standard deviation (11.2%), and ANOVA/ANCOVA (10.2%) methods.It is seen that the MANOVA/MANCOVA (0.2%) and graphic display methods are the least preferred data analysis methods.Distribution of the articles on reading skills according to data analysis methods is given in Table 11.Upon the examination of Table 11, it is seen that the quantitative research (f=347) was preferred more than the qualitative research (f=73) by researchers.Among quantitative data analysis methods, the most used method is the frequency/percent method (18.3%), while the descriptive analysis (13.8%) is the most used data analysis method among the qualitative data analysis methods.

Discussions, Conclusions and Suggestions
There have been many studies conducted on reading skills.It is seen that these studies gradually increase, and it is to be expected that such an increase may reflect in the variety of article subjects.However, it is seen that studies focus more on certain issues.When articles on reading skills are classified according to their subjects, it is seen that "reading comprehension" (20.6%) articles have the highest ratio.This ratio is important because it reflects our approach to reading skills.Çiftçi and Çeçen (2009) state that comprehension must absolutely happen at the end of the reading.A reading without comprehension is unsuccessful.For the reading to achieve its purpose, "reading comprehension" studies are very important.The second-most preferred subject is (15.5%)"reading habit", because it is "a final goal aimed by reading education" (Özbay, 2011, p. 75), it is an expected result that reading habit is among the most preferred subjects.
The third-most preferred subject related to reading skills is "attitudes towards reading" (10.7%).Attitudes, "just as other receptive variables, are known to be important factors to gain and improve reading skills, and in reading success" (Balcı, 2013, p. 34).Attitude studies on reading skills must be continued and increased.Articles with the subject "reading strategies" are among the subjects of great interest to researchers (9.4%).As "solution ways intentionally used by reader to give a meaning to a text during a reading process" (Ülper, 2010, p. 80), reading strategies are necessary for better and more efficient reading.If readers apply these strategies before, during and after reading, they will reach this goal faster.Another often used subject in articles related to reading skills was "reading evaluation" (7.4%) because "evaluation plays an important role for the educational system to reach determined goals" (MEB, 2006, p. 214).In the process and at the end of reading, evaluation studies are important for the reading to achieve its purpose.The fact that the studies on these subjects were preferred can be explained by gathering the respective data mostly through a success test or scale.
It is seen that sufficient studies have not been performed on some subjects related to reading skills.Subjects such as reading acquisitions, relation of reading with other skills, reading preparation, position of reading in Turkish Language curriculum were rarely studied subjects.These subjects are also important for Turkish Language education and if researchers focus on these issues, it will contribute to complete the missing aspects on this field.The requirement of longer time and more effort to gather and analyze the data can be counted among the reasons why these subjects have been studied less.
Studying subjects such as the relationship of reading skills with other language skills or other fields is important for the development of the field.It is understood that the researches related to reading skills has mostly concentrated on certain subjects.Those who are willing to do studies on reading skills may also be recommended to aim at fields such as critical reading, visual reading, media literacy, reading motivation, reading anxiety, and other similar areas.
The distribution of articles on reading skills as per the years in which they were written, show an increase especially after 2006.Between the years 1990 and 2000, there was an article almost every year.A total of 18 articles were published during the 5 year period between 2001 and 2005.Displaying a great increase in 2009 the number of articles (f=32) continued around the same mean for the next two years, with another increase in 2012, number of articles (f=32) continued around the same mean.The increase in the number of articles has been greatly dependent on the increase in the number of Turkish Language Teaching departments and of the researchers studying in these departments, a countrywide spread of postgraduate licenses and doctorate programs in Turkish Language Education Departments and their ability to produce graduates after 2005, especially with the restructuring of Education Faculties in 1998, can be listed as one of the most important reasons of this situation.
When published articles are classified, it is seen that approximately a quarter (f=52) of these articles (f=232) are published in journals under SSCI.Although it appears to be a significant number when we take into account the lack of journals under SSCI, it cannot be argued that this ratio reflects the truth.There are many journals active in the field of education.In this study, almost all journals active under SSCI in the field of education were examined, while only some of the other journals were examined.This result is important for providing the number of articles on reading skills published in Turkey between 1990 and 2013 in journals under SSCI.
Among analyzed articles, it is seen that studies with a single author (59.9%) or two authors (29.3%) are frequently preferred in publications related to reading skills and that studies with four (%0.9) or more authors are not preferred.Saban et al. (2010) found in their content analysis on qualitative researches that studies usually had a single author (56.6%) or two authors (29.2%), and there were very few studies with a higher number of authors.Varışoğlu, Şahin and Göktaş (2013) found in their research that the majority of studies had a single author (66.7%) or two authors (25.6%), and there were fewer studies with a higher number of authors.The requirement to have a singleauthor article published in indexes such as SCI and SSCI is among the criteria of application for becoming an associate professor in Education Sciences and Field Training and can be considered an important factor on the number of article authors.Since the articles published by an associate professor together with their postgraduate students are deemed as written by a single author, the same situation could have happened for the two-author articles in this concept.With regards to applications for associate professorship there may be an increase in both the number and the quality of the articles publicized by joint writers if a requirement of a number of articles is requested regardless of the number of writers in place of an article written by a single writer publicized in an indexed periodical.
Based on the distribution of articles according to institutions/universities, it is seen that most studies on reading skills were performed by Gazi University (14.7%),Anadolu University (6.4%), Ankara University (5.8%) and the Ministry of National Education (9.7%) among 69 institutions or universities.While most of the studies were performed by researchers employed by the universities, employees of the Ministry of National Education and other institutions provide an important contribution.Plenitude of teachers in central universities, and assignment to such universities of research officers with their staff working at other universities because until recently doctorate programs were mainly at such universities, can be shown as a reason for the increase in the number of articles.In recent years, doctorate programs have been opened in various universities.Similar researches in future can present more valid information on the reasons of these findings.
As a result of the analysis, it is understood that researchers usually used one hypothesis (39.2%) in their articles.Göktaş et al. (2012) observed in their study on articles that the number of articles containing one hypothesis was higher than others (53.8%).It could be suggested that researchers prefer using one hypothesis since it allows the focus to remain on only one point and does not disrupt the subject integrity.It was seen that there were a considerable amount of articles with no hypothesis (9.1%).Which is usually found in literature reviews.Although they were literature reviews, using a hypothesis was an expected trend in every study.This may be explained by the lack of information on scientific research methods.Giving a wider coverage to the lectures and seminars on the issue of the reporting of scientific researches may provide a contribution to the solution of this problem.
It is understood by the methods of the articles that, for studies on reading skills, researchers most frequently prefer survey method (43.5%) among non-experimental quantitative research methods.This is followed by experimental method (17.2%), descriptive method (4.7%) and case method (4.3%).Survey method is the most preferred (43.5%) method among quantitative researches while case study is the most preferred (4.3%) method among qualitative researches.Varışoğlu, Şahin and Göktaş (2013) found in their research, in which they examined the method distribution of articles, that survey method was preferred more according to other methods among quantitative researches.Karadağ (2010) found in his research on theses related to educational sciences that survey method is one of the most frequently used methods.Here, it is seen that other quantitative research methods are usually not preferred.The fact that studies prepared by survey method cost less effort and time can be considered as a reason for this situation.
It is seen that quantitative research (71.5%) was preferred according to qualitative research (12%) by researchers.Selçuk, Palancı, Kandemir and Dündar (2014) found in their study that quantitative methods are used more (76.6%)than qualitative methods (14.8%).Arık and Türkmen (2009) also found in their study that quantitative methods are used more (68%) than qualitative methods (19%).Qualitative research has been widely used in recent years in education sciences and linguistics, just as in many other fields (Balcı, 2011).Despite this, articles related to reading skills, written between 1990 and 2013 reveal that this is quite the opposite to Turkish Language teaching in Turkey.If we consider the international trends, we may accept this as an indication for failure in keeping up with the advances in the area."Qualitative research provides deeper information than quantitative research.Qualitative research is necessary to find answers for questions that cannot be expressed with traditional research methods (Büyüköztürk et al., 2013, p. 234).The fact that qualitative research requires more time and deeper research, and that data requires more effort to gather and analyze; can be counted among the reasons for why researchers opt for quantitative methods.However, it would be useful to undertake qualitative research to reveal the quality of the reading and the problems related to reading in a more adequate manner.In addition, combined method studies using qualitative and quantitative research methods can be concurrently used.
As a result of the analysis, it is understood that out of the articles published, 13.4% were literature reviews.Varışoğlu, Şahin and Göktaş (2013) detected that out of 558 articles they analyzed, 63 were literature reviews.Since it is classified a verbal skills, such a ratio is expected in studies related to reading skills.
In articles related to reading skills, it is seen that attitude and perception etc. tests are the most frequently used (28.1%)data gathering tools.It is understood that ratios of data gathering tools such as interviews, success tests, surveys, documents and alternative tools are near each other.In articles related to reading skills, it is seen that attitude, perception etc. tests are the most used (28.1%)data gathering tools (Selçuk, Palancı, Kandemir and Dündar, 2014;Varışoğlu, Şahin and Göktaş, 2013).The researchers put more emphasis on rarely used data collection means such as observation form, performance tests, control lists and portfolio rather than frequently used data collection means such as attitude or perception tests may contribute to the emergence of more qualified researches.
Sample levels in articles show that middle-school students form the greatest sample group with a ratio of 28%.After this, the sample is formed of university students (22%) and primary school students (14.2%).These results show that primary school and middle-school students, who are the targets of Turkish Language education researches, are reached.It is noteworthy that sample selection has plenty of university students.On this basis, it can be said that researchers opted for university students, who are easier to reach.Varışoğlu, Şahin and Göktaş (2013), in their research where they examined the distribution of sample levels used in educational science researches, found that primary school (23.2%), middle-school (20.7%) and undergraduate (22.3%) level students were used the most, and academics, parents and postgraduate students were used less for gathering data.Gökçek et al (2013) determined in their research that teacher candidates (27.5%), middle-school students (18.1%) and primary school students (15.5%) are the most studied sample groups.
Groups those are least likely to take place in samples are; postgraduate students (0.4%), preschool students (0.4%), academics (1.3%), parents (2.2%), teachers (4.7%) and secondary school (highschool) students (5.6%).This is quite a low ratio.Reading is not only a skill continued in school, but Akyol (2012) states that reading must continue for life.Reading is important for parents, academics, teachers and secondary school students as well.Moreover, knowledge and experiences of academics must be exploited more.For this purpose, studies towards these sample groups must be increased.
Another result is the high number of articles with undefined sample levels (18.5%).The reason that this ratio is high is that most of these articles are literature reviews.
When the sample numbers of articles on reading skills are examined, it is seen that sample numbers between 101 and 300 are used the most (27.2%).This is followed by sample numbers of 31-100 (20.7%) and 301-1000 (15.9%).It is seen that sample numbers higher than 1000 is the least preferred sample number (2.2%).Selçuk, Palancı, Kandemir and Dündar (2014) found in their study that most preferred number of samples was between 301 and 1000 (24.1%).It was found that after this range, 31-100 (20.8%) and 101-300 ranges (20.6%) were used the most.Research is usually performed in a certain province or school and consequently samples are insufficient to represent the universe can be considered as reasons for this situation.Performing more extensive research that represent the country in general or more qualitative researches may increase the number of publications on sample groups with various sizes.
It was found according to the research results that researches with sample numbers higher than 1000 are found less (10.6%) according to others.Another salient result is the high number of articles with undefined sample numbers (23.7%).It can be said that since these studies are usually literature reviews, sample numbers are not included.
It is seen that random sampling method (44.4%) is the most used method among the analyzed articles.This shows that, in the studies conducted on reading skills, the sample's power to represent the universe is high (Büyüköztürk et al., 2013).This is followed by convenience (18.1%), and purposeful (15.9%) selection methods.It is seen that studies covering the whole population are very rare (1.3%) in sample selections.It could be said that researchers abstain from studying on the whole of the universe due to reasons such as the difficulty of reaching the whole of the universe, material shortages and requirement of a very long span of time.In addition to that the opinion of studying on the samples would be sufficient from the aspect of the prediction of the conclusion to be obtained could be considered as a reason for that.
When analysis results were examined, it was seen that frequency/percentage method was the most used method (18.3%).This is followed by t-test (18.3%), descriptive analysis (13.8%), mean/standard deviation (11.2%) and ANOVA/ANCOVA (10.2%) methods.It was seen that MANOVA/MANCOVA (0.2%) and graphic display methods (1.4%) was are the least preferred data analysis methods.It is seen that quantitative research (82.6%) were preferred more according to qualitative research (17.3%) by researchers.It is seen that quantitative data analysis methods (82.6%) were preferred more than qualitative data analysis methods (17.3%) by researchers.In a study, where it is seen that quantitative research (71.5%) was preferred more than qualitative research (12%) by researchers; it is natural to have such a result.Küçükoğlu and Ozan (2013) found in their research that qualitative data analysis methods (81.6%) were used more than qualitative data analysis methods (18.3%).In order to increase qualitative data analysis methods, area researchers must opt for qualitative research.For this, it can help that institutions support and facilitate qualitative researches and related practices.
It would be useful to address the less studied subjects, to use qualitative data gathering techniques, and to undertake research associated with other fields for placing the literature in the field on more solid foundations.
Performing such studies on writing, listening/monitoring and speaking skills which are the basic linguistic skills of Turkish Language education, would contribute to the creation of more necessary researches.Besides teaching Turkish as a mother tongue, there is need for studies on reading skills in the fields of its teaching to foreigners and bilinguals.

Table 1 .
Distribution of the Articles on Reading Skills according to their Subjects

Table 2 .
Distribution of the Articles on Reading Skills per Year

Table 3 .
Distribution of the Articles per the Number of Authors

Table 4 .
Distribution of the Articles on Reading Skills per the Institutions/Universities

Table 5 .
Distribution of the Articles according to the number of hypotheses

Table 6 .
Distribution of the Articles according to the Research Methods and Patterns

Table 7 .
Distribution of the Articles on Reading Skills according to the Data Gathering Tools

Table 8 .
Distribution of the Articles on Reading Skills according to the Sample Levels

Table 9 .
Distribution of the Articles on Reading Skills per Sample Numbers

Table 10 .
Distribution of the Articles on Reading Skills according to the Sampling Method

Table 11 .
Distribution of the Articles on Reading Skills according to the Data Analysis Methods Appendix 1. Examined Journals and Numbers of Articles Related to Reading Skill