The Effectiveness of Positive Psychology Intervention in Turkey: A Meta-Analysis

The aim of this study is to determine the effect of the results intervention designed in experimental and quasi-experimental model within certain fields of positive psychology in Turkey by using meta-analysis method. The National Thesis Center database and Google Scholar was used to collect the data. Experimental thesis and articles conducted in Turkey between 2000-2019 were included in the study. Relevant thesis and articles are gathered by searching with the keywords “positive psychology”, “psychological well-being”, “psychological resilience”, “mental well-being”, “hope”, “optimism”, “happiness”, “resilience”. The research focused on experimental and quasi-experimental studies, and studies with pre-test and post-test patterns were reviewed. Comprehensive Meta-Analysis program was used in the analysis of the research data. After the analysis of the data, effect sizes in various sub-dimensions were calculated. In this study, the dependent and independent variables in experimental studies carried out by the researchers in Turkey on certain topics of positive psychology and the effectiveness of the independent variables used together with the effect size calculation were examined. In order to make some calculations related to the metaanalysis of the thesis and articles included in the research, studies with the required arithmetic mean, standard deviation and sample number were examined. Studies outside the field of Psychological Counseling and Guidance, qualitative studies and studies in which related values couldn’t be calculated were not included in the research. 16 studies (2 for positive psychology, 1 for psychological well-being, 2 for psychological resilience, 4 for subjective well-being, 1 for hope, 3 for optimism, 1 for happiness, 2 for resilience) were included in meta-analysis. As a result; according to a random model this study general effect size found 0,8 (0,32-1,27), also fixed model effect size 0,6 (0,43-0,78). The data obtained has indicated that the experimental studies have effective results in acquiring certain social skills related to positive psychology and providing more psychologically happy, hopeful and positive perspectives. The small number of experimental studies reached is an important indicator of the need for different experimental studies in the field of positive psychology.


INTRODUCTION
It is seen that psychology used to focus on three areas before World War II: treating mental illnesses, making people's lives more productive and satisfying, identifying and educating superior abilities and intelligence. But after the World War II, pathology-oriented psychology was found to be insufficient to deal with factors such as world-wide chaos, difficulties encountered, economic depression and so on (Seligman & Csikszentmihayli, 2000). While the widespread belief for an individual to be considered healthy was the absence of visible ailments was valid until that time; after the war, World Health Organization's description of health as "not only the absence of disease or weakness, but complete physical, mental and social well-being" was effective in changing the focus of psychology (WHO, 1948).
During such a difficult period after the World War II, people didn't lose their hope despite losing their jobs, money, status and social support, which urged Csikszentmihayli to seek for an answer to the question "What are the resources that empower people? The fact that an answer cannot be found to this question with sciences such as religion, philosophy and history; and the schools of psychology (behaviorism and humanism) found in those years were inadequate has given rise to the emergence of positive psychology. Seligman denotes that "the field of psychology is not only composed of pathology, weakness or damage, medicine and treatment; but it includes terms such as job, education, love, development and joy. Thus, he states that positive psychology is not a wellintentioned thinking, faith, self-deception or a temporary fashion; but it intends to examine individuals' subjective problems in their entirety with the best scientific methods (Seligman & Csikszentmihayli, 2000). In this context, when we examine the historical development of psychology, two important models emerge: pathology-oriented and positive-oriented studies (Eryılmaz, 2015).

-1 0 +1
The aim of positive psychology is to bring individuals as close to +1 point as possible. It transmits this to individuals by eliminating the situations that disturb the individual and with empowerment strategies that can overcome these situations not through Pollyannism or a simple system of good thinking (Eryılmaz, 2015). This is reported by Hanson (2017) by using a mind-garden metaphor as follows: Suppose your mind is like a garden. You can content yourself with knowing where the things are or look at the weeds and flowers in the garden without judging or changing anything. Secondly, you can remove weeds by reducing the negative effects in your mind. Third, you can grow flowers by increasing the positive aspects in your mind. On the numerical axis, -1 point refers to the pathology point where weeds and flowers stay together, 0 point refers to the state of mind without any weeds, and +1 point refers to the situations in which individuals can plant flowers in their minds.
In this research, the basic concepts of positive psychology such as subjective well-being, mental wellbeing, psychological well-being, psychological resilience, hope, optimism, happiness, resilience which were examined. The first one is well-being, which is a concept that has been researched since Ancient Greek till present (Eryılmaz, 2015;Hefferon & Boniwell, 2018). It is described as a state beyond "feeling the good things". The definition of well-being by Ryff and Keyes (1995) includes six dimensions: self-acceptance, personal growth, life purpose, positive interpersonal relationships, autonomy and competence in community. According to the studies, there is a positive relationship among psychological, subjective and mental well-being, which are the sub-dimensions of well-being.
The definition of psychological resilience was first used in the literature by Block in 1950 (Karaırmak, 2006). In order to be able to define any situation with the phenomenon of psychological resilience, an individual must be faced with a life-threatening risk and the coping skills shown against this risk must be successful. When the studies in the literature are reviewed, there is no definite suggestion as to whether the source of psychological resilience is individual or environmental, but recent studies suggest that the interaction of these two factors is effective in improving psychological robustness (Gizir, 2007).
Hope is the factor that encourages an individual to reach his/her aims actively and help him/her define the ways to achieve his/her goals and proceed with intrinsic motivation in the use of these ways. (Snyder, Harris, Anderson, Holleran, Irving, Sigmon, Yoshinobu, Gibb, June & Charyle, 1991;Irving, Snyder & Crowson, 1998;Hefferon & Boniwell, 2018). It is seen that individuals with superior hopes set achievable goals and divide the aim necessary for this goal into sub-goals (Hefferon & Boniwell, 2018).
Optimism, in its simplest definition, is an individual's being able to look to the future with hope despite all the difficulties in life (Sapmaz & Doğan, 2012). Harris and Middleton (1994) defined it as an expectation of positive events rather than negative ones. Scheier and Carver (1985) put forth that it was a personality trait. Studies conducted by Scheier and Carver (1985), Chang (1998) stated that individuals with high levels of optimism were more active in solving their problems, used problem solving skills more and they differed from the pessimists in terms of the strategies they used. Öğülmüş (2001) translated the term "resilience" into Turkish as "yılmazlık" for the first time and it expresses flexibility and being able to revert to one's own type easily (Greene, 2002). It was defined by Benard (1991) and Rutter (1990) as being able to remain strong against the tough life events and look to the future in an optimistic way. Resilience is an inherent property of some individuals, but it is also a feature that can be acquired and developed later (Gürgan, 2006). Arastaman and Balcı (2013) has found that resilience is a feature that we should have the students gain and also the students who have this feature own higher levels of social competence, autonomy and self-esteem characteristics that are known to be related to resilience.

Research Design
In this study, a survey model was used to determine the experimental studies and their effect sizes. Meta-analysis method was used to calculate the effect sizes of the experimental studies. Metaanalysis is a quantitative research method based on examining the existing studies in the literature and calculating their effect sizes. Since meta-analysis studies are upper-level research of the existing studies, the name "meta-analysis" is used. Meta-analysis studies allow the evaluation of multiple studies at the same time and provide the basis for new studies (Crombie & Davis, 2009).

Literature Review
In this study, experimental thesis and articles conducted in the field of positive psychology in Turkey between 2000-2019 were reviewed. In order to find out the studies to be included in the research; the keywords "positive psychology", "psychological well-being", "psychological resilience", "mental well-being", "hope", "optimism", "happiness", "resilience" were searched on The National Thesis Center database and Google Scholar. After searching with the key words, experimental and quasiexperimental thesis and articles with the topics including "education and training", "psychology", "education-training and psychology" and written in "Guidance and Psychological Counseling" and "Psychology" departments were examined. As a result of these procedures, 24 thesis and 8 articles were found to meet the criteria.

Inclusion Criteria
As an inclusion criteria for the thesis and articles to be reviewed in the research, experimental and quasi-experimental design studies carried out in Turkey in the field of positive psychology between 2000-2019 were examined. In addition, arithmetic means of the pre-tests and post-tests, standard deviations, and sample size values were taken into consideration in order to calculate the effect size of these studies (Table 1).

Exclusion Criteria
As a result of the literature review, studies that were outside the field of guidance and psychological counseling, which were conducted according to quantitative research methods, that did not have pre-test post-test experimental-control groups and which did not have statistical values required for data analysis were not included in the analysis. At the end of the literature review carried out according to this criterion, it was found out that 12 thesis and 4 articles didn't have the aforementioned values, and 1 thesis was in limited category. Therefore, they could not be included in the analysis.
As a result, experimental studies which are carried out in the field of positive psychology in Turkey between 2000-2019 year and that are relevant to the aforementioned criteria were examined. Then meta-analysis was conducted with the data of 8 PhD thesis, 4 master's thesis and 4 articles ( Table 2).

Data Analysis
After the studies were selected in line with the inclusion criteria, coding process started. To tag the studies, the criteria below were used:  (Table 3).

Table 3. Distribution of the Sources by Type and Years of Publication
In the studies reviewed, the total sample number of experimental groups was found to be 288, that of control group was 253, and the total number of samples was 541. Comprehensive Meta-Analysis program was used for data analysis. Hedge's g was used to calculate the effect size. According to Borenstein, Hedges, Higgins and Rothstein (2009), the effect is considered to be small for g = .20-.50, medium for g = .50-.80 and large for g = .80 and above. Funnel plot (Funnel graphic), Egger's regression intercept, Duval and Tweedie's (2000) trim and fill, Begg and Mazumdar (1994) ran correlation, Rosenthal fail safe N were used in the study to determine publication bias. Operating characteristics of a rank correlation test for publication bias (Table 4).

FINDINGS
In meta-analysis studies, there are two options to determine the effect: a fixed and random effects model. In the fixed effects model, the effect size of each study is considered to be the same. However, the demographic variables (gender, age, education, health and financial status, etc.) of the sample often cause the effect size to vary. In addition, it is recommended that researchers use a random effects model in social sciences due to the difficulty in detecting the error involved in the research (Borenstein et al., 2009;Cumming, 2013). The Q test was performed to determine whether the data were homogeneous or heterogeneous, and the distribution of the studies was heterogeneous, so the random effects model was preferred (Table 5).  In the table, the average effect value of the studies included in the research according to the fixed effect model was 0,6; standard error of mean effect size was 0,8; the minimum confidence interval of the mean effect size was 0,43; the upper limit was found to be 0,78. For the homogeneity test of the included studies, Q analysis was used and it was found to be Q = 99,71; this shows that the studies are heterogeneous. Another way to test homogeneity is test. Contrary to the Q statistic, ranges that aren't affected by the current number of studies are as follows: 25% represents low heterogeneity, 50% represents moderate heterogeneity, and 75% represents high level heterogeneity (Cooper, Hedges & Valentine, 2009;Kasapoğlu & Kış, 2016). In this study, the overall effect size obtained according to the random effects model was found to be 85%, which represents high heterogeneity. In this case, the average effect value of the random effects model used in the study was 0,8; standard error of average effect size was 0,24; the minimum confidence interval of the mean effect size was 0,32; the upper limit was determined as 1,27.

Forest Plot
Forest plot, which is a frequently used method in meta-analysis studies, not only shows the confidence interval and effect size of each study analyzed, but also allows us to see the confidence interval and effect size for the whole study. The black boxes in the graph are found by the ratio of total sample by the sample of each of study. The horizontal lines represent the confidence interval. It is possible to say that when these lines are short, the confidence interval is narrow but the sensitivity is high, and when they are long, the confidence interval is wide but the sensitivity is low (Bakioğlu & Göktaş, 2018) (Table 6).

Funnel Plot
Funnel Plot which is used for the control of publication bias in meta-analysis studies is also used to observe the absence of bias in the graph. Each circle in the graph represents a study. In order to avoid publication bias in the research, the studies should exhibit a symmetrical distribution. The circles stacked in any direction in the graph express the bias of publication. In the Funnel plot graph, the horizontal axis shows the effect size and the vertical axis shows the sample size or variance. While the groups with high sample size lie in the upper parts of the graph and have the average effect size, studies with small samples are located below the graph and in a wide area (Bakioğlu & Göktaş, 2018) (Table 7).

Table 7: Publication Bias Graphic
In this study; according to the Funnel Plot graph given in Figure 1, it is seen that the effect sizes of the studies are close to symmetric distribution and the publication bias is not significant. In addition to Funnel Plot graph; Begg-Mazumdar Kendall's, Egger, Orwin Fail Safe Number (FSN) tests were used to investigate bias and Egger score was found p = 0,09. According to these data, it is seen that there is no publication bias. The Orwin Fail Safe Number test whose analysis results are given below, supports this data (Table 8).  (Sarı, 2018).

DISCUSSION
Since the beginning of the 2000s, the effect of increasing positive psychology movement is also seen in Turkey. When the literature is examined, it is seen that the studies are mostly in relational design.
The number of current studies in the experimental design is very small and their effectiveness has not been addressed before this study. The effectiveness of the studies prepared and applied in experimental design has been tested for the first time. Thus, it is aimed that it will contribute to the literature.
While statistical results provide information about how much the study data differ from each other and the expectations of the researcher, the effect size gives information about the relative and different effect sizes of the experimental study data. Within the scope of this study, the effect sizes of 16 experimental studies (12 thesis -4 Master's thesis and 8 PhD thesis-and 4 articles) were examined. According to the findings of this study, the overall effect size of the interventions was 0,80. It is seen that the major ratio among the studies within the scope of the research belongs to PhD thesis. The overall effect average of PhD studies was 0,71 (0,86-1,34); the overall effect average of master studies was 0,67 (0,15-1,20); and the overall effect average of the articles was 0,64 (-0,84-2,12). Based on these results gathered after the interventions, it is possible to say that the interventions have a positive effect in favor of the experimental group.
The effect size classification specified by Hedge is as follows: between 0.00 and ± 0.10 weak, between 0.10 and ± 0.30 small, between 0.30 and ± 0.50 medium, between 0.50 and ±, 80 strong, over 0.80 very strong. Thalheimer and Cook (2002) (Başar, Aşkın & Gelbal, 2016;Ören & Sarı, 2019). When both classifications are considered, it is possible to say that 0,80 effect size value obtained from this study is within the large classification. Therefore, it is seen that the interventions made affect the experimental groups positively.
When the studies carried out abroad are reviewed, the early start date of positive psychology practices shows the importance given to interventions. It is seen that the first intervention was made by Lichter, Haye and Kammann (1980) with the aim of "reducing depression and increasing psychological well-being". According to the data obtained from the pre-test, post-test and 6-week follow-up test of this study which includes experimental and control groups aged between 20-60, intervention program was found to be significant in favor of the experimental group. In the metaanalysis study by Bolier, Haverman, Westerhof, Riper, Smit and Bohlmeijer (2013), the effectiveness of 40 applications for subjective well-being, psychological well-being and depression were tested with the meta-analysis method. They were in experimental design and the overall effect size was found to be 0,34 for subjective well-being studies and 0,20 for psychological well-being studies. Practices made in the field of subjective well-being and psychological well-being in Turkey were examined in this meta-analysis and effect sizes were found 0,7. Based on these results, it is possible to say that the efficiency levels of subjective well-being and psychological intervention programs in Turkey is quite high.
The first study conducted abroad on hope which is another concept discussed in the scope of the study, was conducted by Berg, Snyder and Hamilton (2008) is pain tolerance development and hope vaccination programs with university students. This intervention program did not have an effect on the participants, contrary to expectations, but it is important because it is the first study in this field and contributes to other studies to be conducted. Weis and Speridakos (2011) found similar results at the level of effectiveness of programs in meta-analysis studies in which they measured the effectiveness of hope vaccination programs. It was found that hope vaccination programs had a low level of correlation with self-reported hope and life satisfaction, but no effect on depression, anxiety and psychological stress factors. Similar results were obtained in a study by Seligman, Rashid and Parks (2006) and as examined in this study by Yerlikaya (2006).
The fact that interventions in the field of positive psychology do not have the effect we have expected to see on the experimental group does not mean that the interventions are completely wrong, faulty or weak. This situation may result from the fact that the individuals participating in the therapy or interventions don't have same level of sensitivity (Casellas-Grau, Font & Vives, 2014), the selected intervention studies are not suitable for the experimental group (Sin, Della Porta & Lyubomirsky, 2011), the experimental and control group members have various cultural characteristics (Cohen & Fredrickson, 2010). Another research could prove this was carried out by Mongrain and Anselmo-Matthews in 2012. In their study, made a copy of the study conducted by Seligman, Steen, Park & Peterson (2005) with their own experimental and control groups. The implementation results of these two intervention programs, which are exactly the same, were found to be different than expected. While depressive group in Seligman's study continued to maintain their level of well-being after the intervention program, the effect observed in the experimental group was lower after the intervention program conducted by Mongrain and Anselmo-Matthew (2012). This was attributed to the fact that the experimental group included in Seligman's study was chosen from a group of people who visitors to the Web site created for Authentic Happiness book having read the authentic happiness book written by Seligman and that their participants were less depressive than the ones participating in the intervention by Mongrain and Anselmo-Matthew (2012). When the studies are examined through this perspective, the evaluation should be made considering that the same intervention programs may give different results on different groups.

LIMITATIONS AND SUGGESSTIONS
After the empirical studies in the field of positive psychology in Turkey were eliminated according to certain criteria, 16 studies were reviewed. In this study; moderator effects such as type of studies (master's thesis, PhD thesis, articles), sample size, gender distribution of sample, age of sample, city were not included in the calculation of effect size of the study. This issue can be taken into consideration in future studies.
The positive psychology practices in Turkey was first seen in PhD doctoral thesis publicated in 2006 (Gürgan, 2006;Yerlikaya, 2006). Scarcity of work due to the limited number of interventions in Turkey limits the opportunity to purely examine the concept of positive psychology. This point, which can be considered as a limitation of this study, can be solved by increasing the number of interventions to be made.
As understood from flow chart in Table 2, there are quite many studies in the field of positive psychology as a quantitative value in Turkey. However, most of the studies in the field of Guidance and Psychological Counseling appear to be relational. Positive psychology practices which are similar to the ones carried out increasingly abroad in recent years can also be done in Turkey. Intervention programs can be established for the areas needed by individuals just as the one in the study by Dunn, Uswatte & Elliott (2009). In the study conducted by Eryılmaz (2017), the necessary factors for effective psychotherapy were mentioned based on the existing psychotherapies. These points can be taken into consideration in new programs to be created.
It is seen that the age ranges of the individuals in the experimental and control groups in the studies reviewed are 11-61. Positive psychology intervention programs can be used for younger age groups after making the necessary arrangements (Payton, Weissberg, Durlak, Dymnicki, Taylor, Schellinger & Pachan, 2008;Korkut-Owen, Demirbaş-Çelik & Doğan, 2018). Another developmental period outside the study samples is elderliness. While the general expectation is that the level of well-being decreases with age, Ryan and Deci (2001) stated that this situation is the opposite. As denoted by Howell, Kern & Lyubomirsky (2007), it is seen that positive psychology approaches have an effect on elder groups compared to young groups due to features such as knowledge, experience and wisdom increasing with age. Therefore, intervention programs can be organized for this age group, as well.
It can be deduced from the fact that one of the competence areas of well-being is  (2019). Such studies can also be put into practice in Turkey so that the field of positive psychology is known by more people and it will be easier for the people in need to reach it.
In the literature, the integration of positive psychology into state and private schools and its importance is mentioned in the studies carried out by Terjesen, Jacofsky, Froh & Digıuseppe (2004) (2015) are available in recent years. However, it is not possible to say that they aren't available in sufficient number, so it can be focused on increasing these studies in schools.