Game Therapy Studies : A Review and Meta-Analysis Sample

Accepted: 10.01.2022 Gaming is a natural process for the child, and with the play therapy technique used in this process, opportunities are offered to the child to express their problems and emotions by playing. Play therapy is used in many studies with children. The number of studies conducted on the effectiveness of play therapy in Turkey is increasing with each passing day. In this study, experimental graduate studies related to the play therapy education program from the past to the present (1987-2020) were examined and the results were brought together. A total number of 38 master's and 8 doctoral studies were reached in the YÖK (Council of Higher Education) thesis center related to the subject, but only 7 studies in accordance with the inclusion criteria were analyzed with metaanalysis method. While examining the postgraduate studies, care was taken toward the existence of an experimental and control group in which play therapy was applied. The studies in which play therapy was applied to the experimental group but not the control group were included in this study. In order to calculate the effect sizes, average (X), standard deviation (SS) and sample size (N) values from master's theses were processed in the Excel program. Later, CMA (Comprehensive Meta Analysis) statistical program was used for the analysis required for the study, "Hedges's g" was used to calculate the effect size and the Excel program was used for the collection-processing of the data. As a result of the research, it was found that the effect size of play therapy on children's behaviors has an excellent effect size. It was concluded that 5 of the studies included in the study were excellently effective, 1 was moderately effective, and 1 was broadly effective. These limited studies provide information that play therapy has a positive effect on children.


Introduction
Play is a natural process for children. When the literature was examined, it was seen that the play was defined in different ways. When defining the play, Turkish Language Association (2020) defines the play as an entertainment that improves talent and intelligence, Adlerist play therapy. Play therapy should be planned in accordance with the development and characteristics of each child and should be selected from the types of play therapy (British Association of Play Therapists, 2020; Deniz, 2019;Koç and Yumru, 2019;Temizel, 2019). VanFleet and Faa-Thompson (2010) stated that one of the advantages of play therapy is that it does not require language or verbal expression in children because they stated that this activates the natural tendency of children to play, that children express themselves, communicate and solve problems more readily than the speech-oriented and more cognitive methods designed for adults.
There are many studies in the literature that show that play therapy has a positive effect on children's behavior. In a study of Schottelkorb et al. (2020), It was concluded that childcentered play therapy has a therapeutic effect on the social sensitivity, externalization problems, attention problems and aggression of children with autism. In a study of Uysal (2020); it was concluded that Theraplay play therapy has a healing effect on the emotional and behavioral problems of Syrian refugee children. Ahbab (2019) concluded that play therapy has a healing effect on children's anger problems. In a study of Blanco et al. (2019), titled as the effect of child-centered play therapy on the academic success of at-risk kindergarten students, concluded that child-centered play therapy has a positive effect on academic achievement. In the study of Ray et al. (2009), it was concluded that child-centered play therapy has a curative effect on the behaviors of children with aggressive behavior. In their study, Kılınç and Saltık (2018) concluded that the problematic behaviors previously seen in children participating in play therapy decreased after the therapy. Kockaya and Siyez (2017) concluded in their research that play therapy has a significant effect on children's shyness behaviors. Montemayor (2014) concluded that child-centered play therapy positively affects children with behavioral problems in the preschool period. Chinekesh et al. (2013), concluded that play therapy has a positive effect on improving the social-emotional skills of children. Looking at the results of the relevant research, it is seen that play therapy is effective on children's emotional and behavioral problems, social skills, and developmental problems. In this study, the effect size of play therapy was tried to be calculated in thesis studies on play therapy and an answer was sought for the following question: • What is the effect size of play therapy on dependent variables?

Literature Review
Research data include graduate thesis about play therapy in Turkey. Sample selection was made with purposeful sampling method. Purposeful sampling is to conduct in-depth research by selecting information-rich situations depending on the purpose of the research. Purposeful sampling is a preferred sampling method when it is desired to work in one or more special situations that meet certain criteria or have certain characteristics (Koç Basaran, 2017). During the research, the words "oyun terapisi" in Turkish and "playtherapy" in English were used as key words on the YÖK (Higher Education Council) thesis page. A total of 46 studies from 1987 to 2020 were determined as a result of the screening. However, 7 studies that met the inclusion criteria were included in this meta-analysis study.

Inclusion criteria
The aim of this study is to identify postgraduate studies on play therapy that have been made in Turkey and reveal the general profile of these studies. It is aimed that all of the studies examined in this study are directed to play therapy. In a meta-analysis study, it is important to formulate the inclusion criteria for the respective studies. If inclusion criteria is kept very broad, the quality of the studies included may get weaken which would ultimately reduce the reliability of the output. If the inclusion criteria are kept strict, the results may be based on limited number of studies and generalisation would be more difficult (Lam and Kenedy, 2005). In this study, following criteria applied to avoid above two scenarios: • Access to postgraduate theses in full text in YÖK (Higher Education Council) thesis database, • Including numerical data that will enable the calculation of the effect sizes of the studies, • The scales used in the studies contain total score data other than sub-dimensions, • The language of the studies is Turkish or English, • Studies containing the following keywords "oyun terapi" and "play therapy", • The studies were conducted with children between the ages of 0-18, In the meta-analysis study, the effect size of 7 studies that met these criteria was calculated.

Including Sufficient quantitative Data
In order to calculate effect sizes in meta-analysis studies, quantitative data are needed for the experimental and control groups of the studies (Göçmen, 2004). For this purpose, studies with sample size (N), mean (X), standard deviation (SD) values of the experimental and control groups were included in the meta-analysis study.

Coding
The purpose of this study is to reveal a general review about postgraduate studies on play therapy and to obtain data on the effectiveness levels of play therapy education / program used in these postgraduate studies. First of all, demographic information, effect size calculations and qualitative summaries of the examined postgraduate theses were carried out in the study. The coding method used in the study was collected under 2 headings. The first part includes the demographic characteristics of the research (author names, the year of the study, the institute / department where the study was conducted, the language of the study, dependent and independent variables, the result of the research). In the second part, sample size (N), mean (X) and standard deviation (SS) obtained from experimental and control groups were included in order to calculate the effect size. Expert opinion was sought during the coding stage of this study and following that the necessary analyses has been performed.

Effect Size Calculation
Effect size forms the basis of the meta-analysis and expresses the incidence of the phenomenon (Kural, 2020). For effect size values based on arithmetic averages, effect sizes are grouped in six groups as insignificant, small, medium, large, very broad and excellent.
In this study, X, SS and N values from theses were processed in Excel program to calculate the effect sizes. Then, CMA (Comprehensive Meta Analysis) statistics program was used for the analysis required for the study. "Hedges's g" was used in the calculation of the effect size and Excel 2010 programs were used in the collection-processing of data. Results are given in Table 4.

Heterogeneity Test and Meta-Analysis Model Application
It is crucial to evaluate heterogeneity in meta-analysis study. Output of the heterogeneity test is very important as it is being used to determine the model for general impact calculation. If the p value is less than 0,05 or Q value is greater than the value corresponding to the df value in the chi-square table, it can be concluded that meta-analysis application has a heterogenous structure following the inclusion of the individual studies. That is, these studies are not of a similar nature. In this circumstance, the analysis should be performed under the random effects model (Dincer, 2004). In this study, heterogeneity analysis has been performed to determine which meta-analysis model to be used for the interpretation of study impact.  Table 1 shows p<0,05 (p=0,001) which implies significant difference between the groups.
Since study suggests heterogenous structure overall impact should be considered under the Random Impact Model.

Interpretation of Effect Size
The effect sizes obtained as a result of the meta-analysis can be interpreted according to the coefficient classification. "Hedge's g" was used in calculating the effect size in this study. The following classification was used to interpret the effect sizes in the study (Thalheimer and Cook, 2002):

Results
46 studies were compiled as a result of the scanning. The number of those who met the selection criteria among these studies was limited to 7. While screening, care was taken to for the presence of an experiment and control group in which play therapy was applied. This therapy was applied to the experimental group, but not to the control group. It is concluded that play therapy has an effect when the difference between these two groups is significant. The theses used are shown in Table 2. and Table 3. Table 2. General Summary of the Study A total of 46 studies were found as a result of the searching process conducted at the YÖK thesis center. The number of studies meeting the inclusion criteria in these studies was limited to 7. It is seen that 57.14% of the studies are master's thesis and 42.86% are doctoral thesis. In the studies, it is seen that 57.14% of the sample groups are children in the 0-18 age group, while 42.86% are disabled individuals. It is seen that 28.57% of the studies were conducted between 2011-2015 and 71.43% between 2015-2020. It is seen that 85.71% of the studies are experimental and 14.29% are semi-experimental. It is seen that 42.86% of the studies were worked in the Institute of Social Sciences and 57.24% in the Institute of Health Sciences. It is seen that most of the studies were studied in the Department of Psychology and the Department of Child Health and Disease Nursing (28.57%). It is seen that all of the studies are qualitative and 85.71% of them are published in Turkish. It is seen that 20% of the emotional / behavioral problem types examined together with play therapy in studies are social skills, social adaptation, and emotion regulation. Other types of problems are seen to be behavioral problem / disorder, behavior management, anxiety, anger problem, family attitudes and academic achievement, motor function, balance, functional independence, physical fitness. Among the types of play therapy in the studies; it is seen that only play therapy is used with a ratio of 71.43%, child-centered play therapy with a ratio of 14.29% and developmental play therapy with a ratio of 14.29%. The general summary of the master's theses examined within the scope of the research is given in Table 3 above.  Table 5.Meta Analysis It is aimed to obtain the average of effect size values obtained from each study used in metaanalysis. However, the mean value to be calculated in meta-analysis is not the arithmetic mean value. Since the sample size of each study used in meta-analysis is different, the contribution of each study to the general average is not the same. Weighted average is calculated in meta-analyzes considering the different sample sizes (Şen, 2019). When Table 4 is examined, it is seen that the effect size of play therapy is 1.558. When Table 6 is examined, it is seen that 5 studies have an excellent effect size, 1 study has a moderate effect size, and 1 study has a broad effect size.

Discussion
The number of studies conducted on the effectiveness of play therapy in Turkey is increasing every day. In this study, experimental graduate studies on play therapy education and program from the past to the present (1987-2020) were examined and the results were brought together. A total number of 38 master's thesis and 8 doctoral thesis studies were found in the YÖK (Higher Education Council) thesis center related to the subject, but 7 studies in accordance with the inclusion criteria were analyzed by meta-analysis method. As a result of the research, it was found that the effect size of play therapy on children's behaviors was 1.558 at excellent level. It was seen that 5 studies have an excellent effect size, 1 study has a moderate effect size, and 1 study has a broad effect size. These limited number of studies provide information that play therapy has a positive effect on children.
In the research of Blanco and Ray (2011), examining the effectiveness of child-centered play therapy in first year students with academic achievement risk, they stated that play therapy is effective in increasing the academic success of children and the results of the research support the use of play therapy as an intervention program in academic success. In a study of Altun et al. (2019), it was concluded that developmental play therapy is effective in reducing posttraumatic stress symptoms of orphans and children in need of protection. In a study of Soniya et al. (2019), in which the effect of play therapy on anxiety was examined in hospitalized preschool children, it was concluded that play therapy is an inexpensive treatment used to reduce anxiety. In a study of Ray et al. (2009), it has been concluded that child-centered play therapy reduces children's aggressive behaviors. Schumann (2010) concluded that childcentered play therapy is effective in reducing children's aggressive behaviors. Swan and Ray (2014), after studying the effect of child-centered play therapy on the hyperactivity and nervousness behaviors of two children with mental disabilities, concluded that child-centered play therapy reduced the hyperactivity and irritability behaviors seen in children and improved their behavior. Hayati et al. (2014), in their study examining the effect of childcentered play therapy on the aggression level of children aged between 5-10 years, concluded that child-centered play therapy is effective in reducing children's aggression levels. As a result of their research, Iswinarti and Hidayah (2020) stated that group play therapy improves children's empathy skills and reduces aggressive behaviors. As a result of the research, it was concluded that play therapy has an excellent effect on children's behavior. When the literature is examined, studies show us that play therapy is effective in reducing aggression behavior, stress symptoms and anxiety seen in children, and also supports the result of our study.
When the literature is examined, meta-analysis studies on play therapy that support the results of the findings obtained within the scope of the study were encountered. In their metaanalysis study, LeBlanc and Ritchie (2001) examined 42 studies conducted between 1947 and 1996 and concluded that play therapy is as effective as non-play therapies in the treatment of children with emotional difficulties and reported the effect size as 0.66. Ray et al. (2015) examined 23 studies conducted between 1975 and 2011 in their meta-analysis study and concluded that school-based play therapy has small and medium statistically significant effect sizes on children's problem behaviors, internalizing and externalizing problem behaviors, selfefficacy, and academic achievement. Lin and Bratton (2015) examined 52 studies conducted between 1995 and 2010 in their meta-analytical study of child-centered play therapy approaches and concluded that the effect size of child-centered play therapy was moderate (0.47). Bratton et al. (2005) examined 93 studies conducted between 1953-2000 in their study on the effectiveness of behavioral play therapy and found the overall effect size as 0.80. Slade and Warne (2016) examined 10 studies that met the inclusion criteria in their meta-analytical study, where they examined the effectiveness of trauma-focused cognitive-behavioral therapy and play therapy to help victims of child abuse. As a result of their research, they concluded that play therapy is beneficial for child abuse victims, but cognitive-behavioral therapy is more effective throughout the study. They also stated that play therapy may be more effective in children with externalized behavioral problems.

Conclusion and Recommendations
When the results obtained within the scope of the study were examined, it was found that the effectiveness of play therapy studies was high. The results obtained in this metaanalysis study including master's thesis studies about play therapy that were registered in YÖK (Higher Education Council) thesis center in Turkey were found to be compatible with studies which were conducted abroad. This study is limited to the studies included in the analysis. This meta-analysis study, which has been conducted with very few studies, is thought to be a guide for further meta-analysis studies, especially on play therapy. In new studies to be done, studies that will examine the effectiveness of play therapy studies in a more comprehensive way will be useful. Studies to reveal in which areas play therapy is more effective may be proactive for researchers. Studies to determine the factors that limit and increase the effect of play therapy in new studies will contribute to researchers and experts working on play therapy. Including statistical data in the play therapy studies will result more comprehensive meta-analysis work and more reliable results.