Sosyal Etki Ölçeği’nin Geliştirilmesi ve Psikometrik Yönden İncelenmesi
Füsun Ekşi, Nesrullah Okan, Mücahit Yılmaztürk, Halil EkşiBu araştırmanın amacı üniversite öğrencilerinin sosyal etki düzeylerini belirlemeye yönelik bir ölçme aracı geliştirmek ve psikometrik yönden incelemektir. Ölçek kendini ifade etmeye dayalı (self-report) yedili likert tipindedir. Araştırmanın örneklemi 2019-2020 eğitim-öğretim yılında çeşitli üniversitelerde (Marmara, Fırat ve Muş Alparslan) aktif öğrenim gören 235’i kadın 81’i erkek ve yaş ortalaması 23.08 olan toplam 316 öğrenciden oluşmaktadır. Yapılan faktör analizleri sonucunda 13 maddeden oluşan ölçeğinin tek faktörlü bir yapıya sahip olduğu ve açıklanan toplam varyansın %46,637 olduğu belirlenmiştir. Ölçeğin Cronbach’s alpha güvenirlik katsayısı .941 olarak bulunmuştur. Madde toplam-kalan analizleri için tüm değerler (p<,000) düzeyinde anlamlıdır. Ölçek toplam puanlarının ayırt ediciliğine ilişkin yapılan analizler neticesinde de gruplar arasında anlamlı bir fark olduğu belirlenmiştir (p<,000). Bununla birlikte yapılan doğrulayıcı faktör analizi sonucunda uyum iyiliği değerlerinin uygun referans aralıklarında olduğu görülmektedir (χ2/sd=2,786; RMSEA=0,076; SRMR= 0,09; CFI= 0,929; GFI= 0,915). Araştırma sonucunda elde edilen bulgular Sosyal Etki Ölçeği’nin (SEÖ), üniversite öğrencilerinin sosyal etki düzeylerini belirlemede geçerli ve güvenilir bir ölçme aracı olarak kullanılabileceğini ortaya koymaktadır.
Development of the Social Impact and Psychometrical Examination Scale
Füsun Ekşi, Nesrullah Okan, Mücahit Yılmaztürk, Halil EkşiThe aim of this study is to develop a measurement tool for determining and psychometrically examining the social impact levels of university students. The scale is a 7-point Likert-type self-reported scale. The sample of the study consists of 316 students (235 females, 81 males; average age = 23.08 years) actively enrolled at various universities during the 2019-2020 academic year. As a result of the factor analysis, the scale has been determined to consist of 13 items and have a single-factor structure that explains 46.637% of the total variance. Cronbach’s alpha of reliability for the scale is .941. For the item-total and item-remainder analyses, all values are significant at the level of p < .000. As a result of the analysis on the discriminatory power of the scale’s total scores, a significant difference was determined to exist between the groups (p < .000). However, as a result of the confirmatory factor analysis, the goodness-of-fit values were seen to be within the appropriate reference ranges (χ2 / df = 2.786; RMSEA = 0.076; SRMR = 0.09; CFI = 0.929; GFI = 0.915). The findings obtained as a result of the research reveal the Social Impact Scale (SIS) to be able to be used as a valid and reliable measurement tool in determining university students’ social impact levels.
Introduction
Human beings live within a social environment throughout their lives. One of the basic qualities of being human is social interaction. The relationship between humans and society is bidirectional, which means individuals both affect and are affected by the society in which they live (Yüksel, 2013). In other words, social environment has an important effect on the formation of human behavior (Kağıtçıbaşı & Cemalcılar, 2014). The interaction between human behavior and society has been examined by many disciplines in the field of social sciences from past to present and has been an important area of research in social psychology (Yüksel, 2013).
Social psychology aims to examine why and how people’s feelings, thoughts, and behaviors are affected in a socio-cultural context (Taylor, Peplau, & Sears, 2000). People have their own orientations, behaviors, and characteristics (Ertürk, 2012). Although individuals have various unique characteristics, they may have similar feelings and thoughts as a result of their interaction with the social environment and may react similarly to events (Gedikli, 2011). These similarities are of great importance in the regulation and continuity of social life (Arkonaç, 2005). Social norms are set forth to ensure this social order, and individuals act within the framework of these norms beginning in childhood (Eroğlu, 2015). Thus, some similarities in human behavior form as a result of the life-long joint learning that starts in childhood (Ertürk, 2012; Kağıtçıbaşı & Cemalcılar, 2014; Silah, 2005). Individuals can also sometimes react similarly to events that occur without joint learning (Kağıtçıbaşı & Cemalcılar, 2014). In such situations, individuals’ emotions, beliefs, and behaviors are consciously or unconsciously affected by others (Lee et al., 2006; Vanclay, 2003). The literature explains the occurrence of these changes in people as a result of the discourse and behavior of those around them through the concept of social impact (Sağlam, 2009).
Social impact is the direct or indirect change of people’s ideas, attitudes, feelings, and behaviors as a result of their interaction with others (Rashotte, 2007; Bilgin, 2006 & Choo, 2013). In this process, the decisions individuals make are affected by the environment in which they live (Onurlubaş, 2019). Venkatesh and Davis (2000) stated another definition of social impact to be the influence from the person or groups that one attaches importance to in their environment. Compliance behaviors occur in individuals as a result of social impact. Studies from past to present have examined how group norms affect people’s attitudes and behaviors (Arkonaç, 2005). However, social media, having recently gained importance in people’s lives, has had an undeniable effect on the formation of social norms. In line with these developments, social psychologists in particular are studying how the virtual world affects the compliance behaviors individuals exhibit (Kağıtçıbaşı & Cemalcılar, 2014). One study observed the obedience behaviors exhibited by anonymous accounts in the virtual world to be greater than the obedience behaviors exhibited by accounts opened using one’s actual name (Lee, 2006). Another study revealed that, as the questions asked in the virtual environment become more difficult, people in the online communities of which they are members take the answers given to questions as examples (Rosander & Eriksson, 2012).
Today, the most important issue parents and educators in particular complain about is the level of social impact young individuals encounter. Moreover, this social impact can sometimes lead to negative consequences. Meanwhile, although social impact is prominent on both individuals and social life, a reliable and valid measurement tool for directly determining individuals’ level of social impact does not exist in any country. In this context, this study aims to develop a scale for determining university students’ social impact levels and is considered to contribute to the literature by fulfilling the need for such a measurement tool.
Method
This study aims to bring a measurement tool for determining the social impact levels of people to the literature and includes the steps for developing the Social Impact Scale. The study group of this research consists of 316 university students between the ages of 18-32 years studying in Turkey’s Istanbul, Elâzığ, and Muş Provinces. The study uses the convenience sampling method because it facilitates access to students in all three universities and the comparison among the selected sample groups. During the formation of the study group, the participants were informed about the purpose of the research and participated voluntarily. The average age of the students in the study group is 23.08 years (SD = 2.68).
In the literature review, no measurement tool was found to have measured university students’ social impact levels. In this respect, the aim is to develop a measurement tool for gauging university students’ social impact levels. The obtained 23-item scale has been applied to 316 university students studying in different provinces (Istanbul, Elazığ, and Muş) in the 2020 spring semester. The data were initially collected face to face. However, due to the pandemic process, the remaining (approximately 200 pieces) data were collected online. Meticulous care was taken during the implementation process, and no data was lost. Thus, the scale development analyses began by entering the data obtained from applying the scale to 316 participants.
Results
This heading includes the psychometric findings obtained as a result of the processes carried out while developing and introducing the Social Impact Scale (SIS). The data were analyzed by considering the procedures followed in the literature for scale development. After performing the procedures for exploratory factor analysis (EFA), the obtained scale was concluded to have a single factor consisting of 13 items. Accordingly, the highest item on the scale was found to have a load value of .783 and the lowest item to have a load value of .594. A Cronbach’s alpha of internal consistency greater than .70 was obtained for the overall scale, and the total internal consistency coefficient was found to be .90. After this process, analyses were made regarding the distinctiveness of total scores for the scale, and the independent samples t-test was conducted to determine the difference between the upper and lower 25th percentiles of scores. According to the results from the independent samples t-test performed to determine the difference between the upper and lower percentile groups, a significant difference is seen to exist between the groups (p < .000). When examining the average scores of the groups, significant differences are seen between the scores for the upper and lower percentile groups. As a result of the confirmatory factor analysis (CFA), all items in the Social Impact Scale are understood to be contained in one dimension and to have significant relationships. This result confirms the previously performed EFA and shows that all the scale items can be considered as elements of a single construct. Thus, the results from the CFA have confirmed the 13-item, one-dimensional structure of the Social Impact Scale (SIS). Looking at the analyses made as a result of the CFA, all item load values are seen to be greater than .50. This situation shows all items on the scale to have sufficient load levels.
Conclusion and Discussion
The exploratory factor analysis was performed first by considering the responses of 316 participants. According to the results from the EFA, the conclusion has been reached that the first dimension explains 46.637% variance when the eigenvalue is taken as 1. This amount is considered acceptable in the context of social sciences (Scherer et al., 1988). The results from the factor analysis reveal the scale to have one dimension consisting of 13 items. Confirmatory factor analysis was conducted to verify this result. As a result of the analysis, the single-factor structure of the scale was confirmed and the goodness-of-fit values were seen to be within acceptable limits (χ2 / df) = 2.786; RMSEA = 0.076; SRMR = 0.09; CFI = 0.929; GFI = 0.915). According to the results from this analysis, the scale has sufficient values for its validity and reliability. The scale’s Cronbach alpha of internal consistency, difference in scores between the upper and lower 25th percentile groups, and criterion validity tests were also examined by carrying out the required tests. As a result, the Social Impact Scale (SIS) is seen to have been developed in line with all these analyses and to provide sufficiently valid and reliable scores.