Obez Çocuklara Yönelik Tutum Ölçeği: Geçerlik ve Güvenirlik Çalışması

Obesity appears as the disease and problem of our age and continues to increase day by day. This continuing increase is the focus of attention especially for studies in the field of health (Kabalcıoğlu vd., 2012). Obesity is not regarded as only a health problem. It has also been of interest to psychology due to its adverse social and psychological effects on children (Buğrul vd., 2012; Puhl and Brownell, 2001; Chandler vd., 1994). These adverse effects are stated to be resulted from people’s negative attitudes toward obese children (Yılmaz vd., 2010). The present study aims to develop a valid and reliable Attitude Scale to be used to determine children’s attitudes toward their obese peers. Method: Within the development process of “Attitude Toward Obese Children Scale”, the aim of the study, data were obtained from 823 students in total, 479 (58.2%) of whom are females and 344 of whom are males (41.8%) studying in secondary schools in Samsun in 2012-2013 academic year. Students’ age ranged between 10 and 14. Further, 38.6% (n=318) were fifth-grade students; 16.4% (n=135) were sixth-grade students; 23.8% (n=196) were seventh-grade students and 21.1% (n=174) were eighth-grade students among those who participated in the survey. The first version of the scale developed to determine attitudes of obese children consisted of 95 items. Following the item analysis, 41 items remained. Exploratory and confirmatory factor analysis conducted for structure validity yielded two factors called “rejection of obese peer and acceptation of obese peer”. The last and most ideal version of the scale consisted of 16 items and two sub-dimensions. The reliability coefficient determined by the test-retest method and Cronbach's alpha. Findings: The scale consists of 16 items and two sub-dimensions. The lowest point and the highest point to be taken from the six-point Likert-type scale are 0 and 80 respectively. Higher the scores obtained from the scale indicates positive attitude toward obese peer. Sub-dimension of “Rejection of Obese Peer” consists of 9 items and accounts for 24.72% of the variance. Sub-dimension of “Acceptation of Obese Peer consists of 7 items and accounts for 20.23% of the variance. These two sub-factors together account for 44.95% of the total variance concerning the variable of attitude toward obese children. The factor loadings of the scale items range between .44 and .70. As a result of the confirmatory factor analysis, the proposed model with 16 observed variables and 2 latent variables was found significant. It was determined that the values whose standardized values are shown in the diagram were not above “1”. There was no inconsistency between observed and latent variables. Thus, the values obtained and factor loadings of the items show that the model is acceptable ( /sd=1.57, RMSEA= 0.04, GFI= 0.93, AGFI= 0.91, SRMR= 0.05, NNFI= 0.94 ve CFI= 0.95, p< .00). The internal consistency coefficient (Cronbach’s Alpha) of the scale conducted on the secondary school students was computed .76. The reliability coefficient determined by the test-retest method was calculated .78 (p<.001). The split-half reliability coefficient was .82 for the first half and .76 for the second. Conclusions and Recommendations: The values obtained from the present study indicate that the “Attitude toward Obese Children Scale” developed for Turkish Culture in order to determine students’ attitudes toward obese peers is a valid and reliable scale. This study was performed along with the data obtained from students studying in secondary schools. Scale versions appropriate for all age groups could be developed thereby carrying out validity and reliability studies of the scale with different groups and age ranges. Studies should not only determine levels, factors and results of obesity, but also primarily aim to inform parents and teachers about obesity as a public health problem and to prepare training programs concerning this issue. Considering psychological and social problems obese children experience, improving attitudes toward obese children is as essential as the prevention of obesity in terms of health. 

Bu çalışmanın amacı, obez çocuklara karşı akranlarının tutumlarını belirlemeye yönelik bir ölçme aracı geliştirmektir. Çalışma 2012-2013 eğitim öğretim yılında Samsun ilinde ortaokul düzeyindeki 823 öğrenci grubundan elde edilen veriler üzerinde gerçekleştirilmiştir. Geliştirilen “Obez Çocuklara Yönelik Tutum Ölçeği” 16 maddelik altılı Likert tipi bir ölçektir. Ölçeğin yapı geçerliliğini belirlemek amacıyla yapılan açımlayıcı faktör analizi sonucunda iki alt boyutlu bir yapıya sahip olduğu belirlenmiştir. Ölçeğin “Obez Akranı Reddetme” olarak adlandırılan birinci alt boyutu 9 maddeden oluşmakta ve varyansın %24.72’sini açıklamaktadır. “Obez Akranı Kabul Etme” olarak adlandırılan ikinci alt boyutu ise 7 maddeden oluşmakta ve varyansın % 20.23’ünü açıklamaktadır. Bu iki alt boyut birlikte obez çocuklara yönelik tutum değişkenine ilişkin toplam varyansın %44.95’ini açıklamaktadır. Ölçeği oluşturan maddelere ilişkin faktör yükleri ise .44 ile .70 arasında değişmektedir. Ayrıca yapılan doğrulayıcı faktör analizi ile modelin iyi uyum gösterdiği belirlenmiştir ( /sd=1.57, RMSEA= 0.04, GFI= 0.93, AGFI= 0.91, SRMR= 0.05, NNFI= 0.94 ve CFI= 0.95, p< .00). Ortaokul öğrencileri üzerinde geliştirilen ölçeğin iç tutarlılık (Cronbach’s alpha) katsayısı .76 ve test tekrar test güvenirlik korelasyon katsayısı .78 (p<.001) bulunmuştur. Test yarılama güvenirliği katsayısı ise, ilk yarı için .82, ikinci yarı için .76’dır. Elde edilen bu değerler Obez Çocuklara Yönelik Tutum Ölçeği’nin geçerli ve güvenilir bir ölçek olduğunu göstermektedir