A Formative Assessment Example : Word Association Test

This research was carried out to determine the effectiveness and functionality of the word association test (WAT), which is a formative assessment tool that is frequently emphasized on today’s modern education systems. The study group consisted of 60 students in a public school in Kocaeli in the school year 2018-2019. Participants were identified using convenience sampling technique. The data of the study were obtained by using pre-test and post-test quasi-experimental design with no control group. The data were categorized by subjecting to content analysis. The findings were tabulated using the cut-off technique and analyzed using the Wilcoxon signed-rank test. When the results of the study were examined, it was concluded that conceptual change and development occurred in participants’ minds and there was a significant difference in the results of the Wilcoxon test performed before and after the implementation. It was observed that the students wrote 1669 words before the implementation, and the number increased to 2193 after it. This shows that the students associate the key concept of “migration” with more words after the implementation and thus there is a wider connotation related to migration in their minds. In addition, the results of this research reveal that the WAT is suitable for formative assessment and can be used in educational studies.


Introduction
As a result of the review of the relationship between teaching and assessment concepts for the last 20 years, it has been shown that "assessment" guides students' learning skills (Biggs & Watkins, 1996).
Today's education systems aim to provide students with a higher quality education and to educate individuals who continuously improve themselves academically, forcing societies to canalize to more contemporary education approaches. This has played an important role in the adoption of a constructive approach by many developed countries such as the United States, Australia, Finland and New Zealand (Aşkar et al., 2005). The constructivist approach in Turkey, which was adopted in 2005, is a modern approach that supports the match of each student with a learning material appropriate to his/her level of learning and encourages students to learn by experiencing knowledge (Fosnot & Perry, 1996;Lainema & Makkonen, 2003). The constructivist approach has brought many innovations to our education system. One of these innovations is formative assessment and evaluation although it is not fully adopted yet.
The students are confronted with many concepts during the education process and learn them through mental processes. How these concepts form a structure in the mind can only be explained based on students' thoughts on key concepts (Gilbert, Boulter, & Rutherford, 1998). The abstract ones of these concepts prevent the formation of a certain structure in students' minds (Knippels, Waarlo, & Boersma, 2005;Krawczyk, 2007;Quinn, Pegg, & Panizzon, 2009). Accurate understanding and interpretation of each concept encountered in the education process will contribute to the academic life of the students. For a concept to be well understood and stick in the mind, other words related to that concept must be associated (Bahar, 2006). The more students associate a concept with words, the easier it will be to understand the concept (Kaya & Taşdere, 2016). Particularly, constructivist learning approach supports revealing the cognitive structure of students by giving importance to the use of different techniques in determining the conceptual understanding and the relationships between concepts (Vance, Miller, & Hand, 1995). It is difficult to correct the concepts that are misunderstood and cause misconceptions in the mind by using traditional teaching strategies and assessment and evaluation techniques (Posner, Strike, Hewson, & Gertzog, 1982). This requires focusing on formative assessment techniques. Word Association Tests, which are one of the formative assessment techniques, are very effective in revealing the cognitive structures of the students about the concepts they have.
Word association tests (WATs) are the most common alternative evaluation and assessment techniques used to analyze the cognitive structure of the students, to determine whether the relationships between the concepts stored in the mind during education and training are sufficient and to determine students' perceptions of the concepts (Cachapuz & Maskill, 1987;Johnstone & Moynihan, 1985;Özatlı & Bahar, 2010). The technique involves writing a concept about the key concept for any subject, without limiting students' minds, and in a limited time independently (Bahar, Johnstone, & Sutcliffe, 1999;Sato & James, 1999). Bahar and Özatlı (2003) stated that students should write the answers related to the key concept as a list and this may prevent the risk of chain response. The risk of chain response is the risk of writing the concept suggested by the first concept if the student does not return to the key concept after writing a concept for each key concept (Polat, 2013). To eliminate this risk, students should be encouraged to write one by one as a list. The order in which the students write their own sentences in the word association test is a measure of the semantic affinity of the stimulating (key) words (Deese, 1965). After writing key concepts, students may be asked to form a sentence about the given concept. These sentences are used to analyze students' cognitive structure (Gunston, 1980). During the process, students were given various time periods to write the words. These periods are 30 seconds (Bahar, Johnstone, & Sutcliffe, 1999;Kempa & Nicholls, 1983;Steinberg, Bieliauskas, Smith, & Ivnik, 2005), 45 seconds (Dinç, Sezer, Üztemur, & İnel, 2018), 1 minute (Ercan, Tasdere, & Ercan, 2010), 1.5 minutes (Kaya & Tasdere, 2016) and 3 minutes (Aladag & Yilmaz, 2014). There are also studies in which a 20-second period is allowed for the students (Keskin & Örgün, 2015).

Research Model
Quantitative research designs were used in this study which was conducted in order to determine how the word association test allows formative assessment. In order to determine to what extent the effect of the process on students can be measured by using WAT, pre-test and post-test were designed as a quasi-experimental design with no control group. With this design, the significance of the change between the pre-test and post-test scores of a single group can be determined (Büyüköztürk, Çakmak, Akgün, Karadeniz, & Demirel, 2010).

Participants
The sample of research consisted of 60 fourth-grade students of a public school in the migration region of Kocaeli in the academic year 2018-2019. Since the suitability of WAT to formative assessment will be discussed in the evaluation of migration issue, it is important to conduct the study in a migration region. The sample designed using convenience sampling technique, was preferred since the immigrant children in the migration region had to be selected in order to serve the purpose of the study and since the school in which the study was carried out was the school where one of the researchers worked.

Measurement Tools
WAT was used as data collection tool in the research. WAT is a measurement tool in which participants are asked to write the first word that comes to their minds about key concepts in a short period of time (Atasoy, 2004). Word association tests can be used to evaluate process and formatting and to identify pre-learning (Güneş & Gözüm, 2013). It is also a very functional technique in terms of its preparation, application and use in all classes and large groups (Bahar, Nartgün, Durmuş, & Bıçak, 2010). Within the research, 10 key concepts related to migration were determined in line with the opinions of 3 field experts. Each of these key concepts was prepared on a page and presented to the participants. In order to eliminate the risk of chain response, key concepts were written in a list (Bahar & Özatlı, 2003). The key concepts are: Border, Farewell, Loneliness, Passenger, Immigrant, Foreign Land, Moving, Longing, Migration and Journey. These key concepts are common concepts that emerge in line with the opinions of the field experts on the children's picture books to be used in the research. Necessary explanations about WAT and process were given to the participants. The time periods for writing the key concept in the literature may vary according to the group in which the WAT is applied. Since the group of participants in this study was 4th grade primary school students, it was decided to give the participants a 3-minute period (Aladağ & Yılmaz, 2014). A page of the measuring tool is shown below as an example. Related Sentence …………………………………………………………………………….

Procedure of the Research
This research was conducted using quantitative research designs to determine to what extent the Word Association Test allows formative assessment. The research was carried out with the participation of 60 4th grade students, who were identified with convenience sampling method. In line with the sample selection, a school in the migration region was preferred. In this context, it was aimed to determine to what extent the word association test serves formative assessment in revealing the cognitive structures and vocabulary of the students.
In the first step of the research, concepts that will explain the subject of migration were determined in line with expert opinions. In the light of these words, 15 illustrated children's books which can be used in practice and are accessible are determined. The names of these books and authors are as follows; "Stepping Stones, Nizar Ali Badr", "Why Am I Here?, Constance Orbeck-Nilssen", "Op de Vlucht, Prim Van Hest", "The Journey, Francesca Sanna", "The boy in the Boat, Sevde Tuba", "Welcome, Barroux", "Me and My Fear, Francesca Sanna", "Amin Loses His Way, Aleix Cabrera", "I walk with Vanessa, Kerascoet", "The Journey Home, Fran-Preston Gannon", "I Have the Right to be a Child, Alain Serres", "The box, Merve Çirişoğlu Çotur", "The Three Questions, Jon J. Muth", "Mother Bruce, T. Higgins", "Marwan's Journey, Patricia De Arias", "The other lions, Gökçe Gökçeer". Four of these books were selected by taking into consideration the characteristics of the student group, the peritextual characteristics of the books, the ways in which they deal with migration and the expert opinions.
These books are found appropriate to use: "Journey" written by Francesca Sanna and translated into Turkish by Zeynep Sevde, "Me and My Fear" the continuation of the same book and also written by Francesca Sanna and translated by Zeynep Sevde into Turkish, "The Boy in the Boat" written by Maya Mizuno and Vonne Hemels, about a Syrian child who had to leave his country, "Mother Bruce" written by T. Higgins and translated into Turkish by Nurten Hatirnaz about migration of the goose.
Before the implementation of the WAT, information was given about the research and the word association test.
After being sure that all participants learned how to complete the WAT forms, the students were asked to complete WAT forms in 3 minutes for each concept. In this context, prior knowledge of children, cognitive structures related to the concept of migration and vocabulary related to the identified keywords have been determined and the implementation phase has been started. During the implementation phase, children's picture books were read to the groups of 20 students in an interactive way. Reading of each book lasted for 1 lesson and reading process was carried out in 12 lesson hours. After reading the books, WAT forms were reapplied in order to quickly identify the changes in students and give feedback in accordance with formative assessment.
The data obtained after the implementation was analyzed and compared with the situation observed before the implementation and the research was finalized.

Data Analysis
The data of the research were analyzed using content analysis and categorized by forming certain titles. Content analysis is a scientific approach that involves categorizing various data types by adopting an objective perspective and presenting in an understandable way for the reader (Yıldırım & Şimşek, 2006). The use of content analysis may vary depending on the researcher's problem and how he/she approaches the problem (Weber, 1990). The words written by the participants were analyzed by 3 researchers and then converted into quantitative data. The reliability formula proposed by Miles and Huberman (1994) was used to determine the consensus among the researchers. In the resulting calculation, the reliability value of the study was determined as 88%. In line with this formula, the consensus among coders should be at least 80% (Miles & Huberman, 1994;Patton, 2002). In order to reveal the cognitive structures of the students about migration, repetitive words were identified, and mind maps were created. In this stage, the cut-off point technique, which was put forward by Bahar et al. (1999) was used. Some of the data were analyzed with the help of SPSS program (IBMCorp., 2016) in order to determine how much each participant's data distribution changed after the implementation. In this context, Wilcoxon signed-rank test was used for the non-normally distributed dependent groups.

Results
The results of this study aimed to measure how much the word association test allows formative assessment was obtained through content analysis applied to WAT forms. In this direction, the vocabulary and cognitive structures of the participants were tried to be determined. In order to compare the participants' post-implementation and pre-implementation vocabulary, they were analyzed in SPSS program using Wilcoxon signed-rank test. The findings obtained after these procedures were tabulated.  Table 1 shows the number of words obtained from the WAT forms of the participants. When Table 1 is examined, it is seen that various numbers of words are written for each key concept. Minimum words are written for the key concept of "Migration" key concept (149) and the maximum words are written for the key concept of "Passenger" (208) and all the key concepts have 1669 words. Approximately 63% of the words associated with the key concepts by the participants were identified as related to the key concept and 36% of them were identified as unrelated to the key concepts.   (251) is the key concept in which participants write the most words as in the pre-test results. When all the key concepts are examined, it is seen that the number of words written for each key concept increases. The post-test data consisted of 2193 words (1473 related, 720 unrelated). When the 2 another. T longing is the key co It is seen t post-test. T The key co concept of "stork", w org 21+ post-test c he key concep associated wit ncept of borde that word asso The key conce oncept of passe f loneliness is a while the key co hat none of the ated with the w moving is asso or 16-20 word e cut-off point n associated w f journey, in ad concept of migr he word "name ds e key concepts word "bird". T ciated with the d t formed betwe with the key con ddition to the w ration is assoc e". Vol. 13,No. 8; are associated The key conce e word "house" een 16-20 wor ncept of migra word "car".  amined, it is s gration at the c tween 5-9 wor oint between 5key concept of ed with the key Also, the key pt of longing. . ords een that many enter. In additi rds -9 words. Whi f border is asso y concept of lo y concept of . In addition t Vol. 13,No. 8; y key concept ion, 9 key con ile the key con ociated with the oneliness. Ther foreign land h to them, many 2020 s are cepts cepts e key e is a has a y key ies.ccsenet.org International Education Studies Vol. 13, No. 8; 2020

Discussion and Conclusion
In this study, the effectiveness of the word association test (WAT), which is one of the formative assessment techniques emphasized by modern education systems, is examined. It was concluded that the conceptual change and development occurred in the participants' minds related to the key concept of migration and there was a significant difference in the analysis of the words they wrote about the given key concept before and after the implementation.
Before reading the children's picture books on the theme of migration, WAT was applied to the participants and the data were collected and analyzed. After reading the children's picture books, the WAT was applied again, and the answers of the children were analyzed, and their cognitive connections were revealed. The results indicate that the participants' thoughts about the migration have changed, their knowledge about the concept of migration has increased and they have become aware of the migration.
In addition, the total number of words that each participant wrote on the key concept of migration before and after reading the children's picture book was calculated. It was seen that the participants wrote 1669 words before the implementation, and 2193 words after the implementation. In the pre-test, 1053 of the words written by the participants about the concept of migration were related and 616 of them were unrelated. In the post-test study, it is seen that 1473 words are related, and 720 words are unrelated. In addition, the cut-off sampling was used to determine changes in the cognitive structures of the participants before and after the implementation using WAT.
When the pre-test and post-test cut-off points were examined, it was seen that the words associated with the key concepts increased after the implementation and the key concepts were related to each other. These results show the effectiveness of the WAT in showing how the words written by the participants about a key concept have changed at the end of the implementation and how relevant the words written by the participants to the key concept. These results also reveal the effectiveness of WAT in showing the change in the cognitive structure of the students.
In addition, these data were analyzed in SPSS package program to see if there was a significant difference and then Wilcoxon test was applied. As a result of the results, it was concluded that there was a significant difference between the number of words of written by students before and after the implementation (p<.05). According to the results of Wilcoxon test, which calculated the relationship between the total number of words written by the participants before and after the implementation, it was found that there was an increase in the number of words written by 42 participants and decrease in 13 participants and there was no change in 2 participants after the implementation.
Social Studies course does not attract students' interest and does not remain in students' minds because some teachers do not give the necessary importance to the course or teach the course with traditional teaching methods (Freeman & Lestik, 1988). Since the concept of "migration" is known to be a part of the social studies curriculum (MEB, 2019), underestimation of this course may cause students not to perceive this concept sufficiently. It is very difficult for students to learn the concept of migration permanently unless they are subjected to appropriate education (Rüzgar, 2014). Within the research, the lesson has been made more active thanks to the children's picture books and the meaning of the increase in the number of words has become more significant when it is considered that the abstract concept of "migration" is assimilated to the students.
There are several studies that have similar results to those obtained during the implementation of this research. In these studies, WAT has been effective in revealing the cognitive structures of the participants' minds (Deveci et al., 2014;Karaca, 2018;Doğan et al., 2018;Ekici & Kurt, 2014). There are also similar studies stating that WAT helps to reveal participants' misconceptions on a key concept and to conduct studies to eliminate them (Aydemir, 2014;Işıklı et al., 2011;Keskin & Örgün, 2015;Ercan et al., 2010;Kaya & Flow, 2015). It was found that the WAT is very useful in determining prior knowledge and students' misconceptions about the subject during the research implementation and it is suitable for formative assessment. However, WAT aims to reveal the cognitive structures of students about a subject, but it ignores the underlying causes of the words that students associate with key concepts. A word in a student's mind described by researchers as related to key concepts can be included in the category of unrelated word-misconception because the teacher may not be aware of the student's experiences. Karaca (2018) stated in his thesis, which aims to reveal the perceptions of the students towards the professions, although there are no misconceptions about the professions, the words they associate with the profession do not reflect their occupational characteristics. In this research, the words that the students associate with the professions do not reflect the occupational characteristics. At this point, it is thought that the information obtained through WAT forms might not reveal necessary prior knowledge about the students' lives. Bahar and Özaltı (2003) stated in their study, aimed to reveal the cognitive structures of the students for the basic ies.ccsenet.org International Vol. 13, No. 8; components of people, that when they use WAT; the students' prior knowledge was spread over a wide area and they included words that were not related to the subject, but they mentioned that the students provided more scientific answers in the data obtained after the implementation. In this study, when the cut-off points were examined, it was seen that the number of common words associated increased in the post-test data and the students started to associate key concepts with each other. In formative assessment, it is important to evaluate the progress of the students in a planned manner at regular periods in the process. Before starting to teach a subject process, it is thought that WAT forms can be helpful while designing a process suitable for formative assessment by determining the instructional needs of students by applying WAT. In his study, Polat (2013) applied the WAT forms to the same students 2 weeks after the teaching and 1 year after the teaching and determined that there was a high decrease in the total number of answer words given by the students in the second implementation. He also stated that WAT is a very powerful alternative measurement tool in diagnosing and identifying. In this respect, Polat's (2013) views are similar to the ideas reached as a result of the study.
In the studies carried out by Polat (2013), Bahar and Özatlı (2003), Ercan et al. (2010) and Kaya and Taşdere (2016), a group of participants were subjected to WAT before and after being subjected to an education process and the number of words they wrote for the key word was calculated. In these studies, the number of words increased after the implementation. In this study, the increase in the number of words of participants before and after the implementation was successfully demonstrated. Considering this situation, it is thought that the WAT can be used to implement the feedback element which can be considered as the most important element of formative assessment and to examine the development process of the students.
In conclusion, this study has once again showed that WAT is a formative assessment tool. The WAT can be used as an alternative assessment tool suitable for formative assessment in the teaching process. Moreover, studies can be organized to determine the common ideas of the students on two different topics, experimental studies can be designed by using WAT over a long period and the effect of these evaluations on student achievement can be examined. In-service trainings may be offered to teacher candidates in order to popularize WAT forms and to emphasize their suitability for formative assessment.