Kendi Kendine (Öz) Liderlik Ölçeğinin Eğitim Örgütleri İçin Türkçeye Uyarlanması, Geçerlik Ve Güvenirlik Çalışması

The paper aims to carry out the validity and reliability analyses of the Turkish version, adapted for educational organizations, of Self Leadership Scale developed by Houghton and Neck (2002). Self-leadership means a process of self-influence including self-motivation and one’s directing himself/herself to accomplish or realize something. This type of leadership is also described as a process encompassing controlling, influencing and directing own behaviors by means of special behavior sets and cognitive strategies. The self-leadership process includes three important strategies, that is (a) behavior-focused strategies, (b) natural reward strategies and (c) constructive thought pattern strategies. Method: The reliability and validity analyses regarding the scale adapted for educational organizations in Turkish culture and language were carried out by gathering the opinions of two independent working groups, both composed of participants working in Malatya during 2013-2014 academic year and all voluntarily participated into the study. As data collection tool in this study was used the Turkish from, adapted for educational organizations, of Self Leadership Scale developed by Houghton and Neck (2002). The self-leadership scale is composed of behavior-focused strategies, natural reward strategies and constructive thought pattern strategies and nine sub dimensions. There are a total of 35 items in the scale. First an EFA was carried out, then a CFA was carried out to test model – data fit regarding the factor structure revealed out of EFA. As for the reliability, corrected item-total correlation coefficients and Cronbach's alpha coefficients of internal consistency were calculated. Findings: As a result of analyses, it was found out that there exist 29 items in the scale and these items are grouped under nine sub dimensions as they are in the original scale. Items in the analyses explain 63,18 % of the variance and factor loadings vary between .47 and .84. Communalities of factors explained vary between .49 and .77. In addition, corrected item-total correlation coefficients were calculated to test item validity ant homogeneity. The corrected item-total correlation coefficients of the scale for the data gathered out of the first working group vary between 37 and .68. A CFA was carried out to test model – data fit regarding the factor structure revealed out of EFA. The goodness of fit values for CFA are as follows: X2/df=1.85, GFI=0.94, AGFI=0.90, NNFI/TLI=0.92, CFI= 0.94, RMSEA=0.062, RMR=0.027, SRMR=0.063. Confirmatory factor analysis results revealed that the model obtained is similar to that of the original scale model for Self-leadership Scale developed by Houghton and Neck (2002). As a result of the reliability analyses, as a result of analysis for the data of the first working group, the Cronbach's alpha coefficients of internal consistency for nine subscales were calculated as .70, .75, .72, .67, .75, .75, .69, .71 and .72 respectively. Discussion and Conclusion: As a result of the validity and reliability analyses, in the final form of the scale are 29 items and nine sub dimensions. Self-leadership stands as one of the leadership approaches playing a prominent role both on individual performance and organizational performance. In Self leadership, everyone takes energy from his/her own leadership styles. According to this approach, new criteria of leaders are considered to be “to determine their own future right” and “to recognize own skills and develop them to the highest level”. Each item in the scale is evaluated and analyzed on a Likert-type scale, which consists of options "1 = Never", "2 = Rarely," "3 = Sometimes", "4 = Often", "5 = Always". Based on the findings of the validity and reliability analyses, it was obtained proofs regarding the fact that the Turkish version of the Self-leadership Scale adapted for educational organizations is a valid and reliable tool to measure the self-leadership characteristics and qualities of teachers.

Özet Bu araştırmanın amacı Houghton ve Neck (2002) tarafından geliştirilen Kendi Kendine (Öz) Liderlik ölçeğinin eğitim örgütleri için uyarlanan Türkçe formunun geçerlik ve güvenirlik çalışmalarını yapmaktır. Ölçek formunun geçerlik ve güvenirlik çalışmaları, 2013-2014 eğitim-öğretim yılında ve araştırmaya gönüllü katılan birbirinden bağımsız iki çalışma grubunun görüşleri alınarak gerçekleştirilmiştir. Birinci çalışma grubu 259 öğretmenden; ikinci çalışma grubu ise 225 öğretmenden oluşmaktadır. Birinci çalışma grubundan alınan verilerle yapılan Açımlayıcı Faktör Analizi (AFA) sonucunda, analize alınan 29 maddenin toplam varyansın %63.18’ini açıkladığı, maddelerin faktör yüklerinin .47 ile .84 arasında değiştiği, belirlenmiştir. AFA sonucunda elde edilen dokuz faktörlü yapının test edilmesi için ikinci çalışma grubundan elde edilen verilerle Doğrulayıcı Faktör Analizi (DFA) yapılmıştır. Doğrulayıcı faktör analizi sonucunda elde edilen modelin özgün ölçek olan Houghton ve Neck (2002) tarafından geliştirilen Kendi Kendine (Öz) Liderlik ölçeğinin modeli ile benzer olduğu sonucuna ulaşılmıştır. Güvenirlik çalışmaları kapsamında yapılan analizler sonucunda, Cronbach Alpha iç tutarlılık katsayıları, dokuz boyut için sırasıyla .70, .75, .72, .67, .75, .75, .69, .71, .72 olarak hesaplanmıştır. Yapılan geçerlik ve güvenilirlik çalışmaları sonucunda kendi kendine (öz) liderlik ölçeğinin eğitim örgütleri için uyarlanan Türkçe formunun öğretmenlerin kendi kendine (öz) liderlik özelliklerini ölçmede kullanılabilecek geçerli ve güvenilir bir ölçme aracı olduğu yönünde kanıtlara ulaşılmıştır.