Validity and Reliability Study of the Violence Prevention Climate Scale
Salih Dursun, Oğuz Başol, Serpil AytaçExposure to workplace violence increases employees’ levels of stress, depression, anxiety and burnout and decreases the productivity of the organization. The violence prevention climate refers to the perceptions of employees about organizational policies, procedures and practices relevant to eliminating and controlling violence or aggression in the workplace. The violence prevention climate is an important concept both in the emergence of different types of violence in the workplace and its effects on employees. Therefore, it is important to measure the violence prevention climate. In this context, this research aims to adapt the Violence Prevention Climate Scale, developed by Kessler et. al., into the Turkish language. According to the results of the research conducted with the participation of 279 private-sector employees, the total explained variance was determined as 72.45%. According to the confirmatory factor analysis results, the scale consists of 18 items and 3 factors as in the original (Chi-square/df: 3.78; RMSEA: 0.073; NFI: 0.95; NNFI: 0.96; CFI: 0.96; GFI: 0.90 and AGFI: 0.85). The internal consistency of the scale was calculated as 0.924 and item-total correlations ranged between 0.375 and 0.752. As a result, it was concluded that the violence prevention climate scale is valid and reliable for Turkish society.
Şiddet Önleme İklimi Ölçeğinin Türkçe Geçerlik ve Güvenirlik Çalışması
Salih Dursun, Oğuz Başol, Serpil Aytaçİşyerinde şiddete maruz kalmak bir yandan çalışanların stres, depresyon, kaygı ve tükenmişlik düzeylerini arttırmakta diğer yandan ise örgütün verimini düşürmektedir. Şiddet önleme iklimi, işyerinde şiddet veya saldırganlığın ortadan kaldırılması ve kontrol edilmesine yönelik, örgütsel politika, prosedür ve uygulamalarla ilgili çalışanların algılarıdır. Şiddet önleme iklimi gerek işyerinde farklı şiddet türlerinin ortaya çıkmasında gerekse de çalışanlar üzerindeki etkileri nedeniyle önemli bir kavramdır. İş ortamında şiddet önleme ikliminin ölçülmesi bu açıdan oldukça önemlidir. Bu bağlamda mevcut araştırma Kessler ve diğerleri tarafından geliştirilen şiddet önleme iklimi ölçeğinin Türkçe geçerlik-güvenirlik çalışmasını gerçekleştirmeyi amaçlamaktadır. 279 özel sektör çalışanının katılımıyla gerçekleştirilen araştırma sonuçlarına göre, ölçekte bulunan maddelerin faktör yükleri 0,671 ile 0,846 arasında değişmektedir ve açıklanan toplam varyans %72,45 olarak tespit edilmiştir. Gerçekleştirilen doğrulayıcı faktör analizi sonuçlarına göre, şiddet önleme iklimi ölçeği (Ki-kare/serbestlik değeri: 3,78; RMSEA: 0.073; NFI: 0.95; NNFI: 0.96; CFI: 0.96; GFI: 0.90 ve AGFI: 0.85) orijinal ölçekte olduğu gibi 18 madde ve 3 faktörden oluşmaktadır. Ölçeğin içsel tutarlığı 0,924 olarak hesaplanmıştır ve madde-toplam korelasyonlarının 0,375 ile 0,752 arasında değiştiği görülmüştür. Sonuç olarak analizler, şiddet önleme iklimi ölçeğinin bu alandaki araştırmalarda, araştırmacılar tarafından kullanılabilecek geçerli ve güvenilir bir ölçme aracı olduğunu göstermektedir.
One of the reflections of the concept of violence, which is as old as the history of humanity, is workplace violence. Workplace violence, defined as all kinds of actions or events aimed at physical or psychological harm (Wiskow, 2003, p. 7), affects employees negatively. Workplace violence affects not only those who are exposed to violence but also the ones who witness workplace violence. Workplace violence also causes significant problems for organizations. The main organizational consequences of violence are the increase of illness-related absences, job turnover rates, insurance costs and the reduction of productivity (Di Martino, Hoel and Cooper, 2003, p. 63). Moreover, workplace violence affects employees’ motivation and commitment, quality and timely production and competitiveness, corporate loyalty, creativity, working climate and causes indirect costs on organizations (Di Martino, 2002, p. 27). The main social cost of workplace violence is the increase in health/treatment costs, early retirement and loss of production (Hoel, Sparks and Cooper, 2001, p. 51).
As can be seen, workplace violence causes serious costs for employees, organizations and society. Therefore, it is important to identify organizational factors to reduce violence. One of the important organizational factors for reducing or preventing violence is the climate of preventing violence (Spector, Yang and Zhou, 2015, p. 325-326).
The violence prevention climate is the perception of employees about organizational policies, procedures and practices for eliminating and controlling violence or aggression in the workplace (Spector, Coulter, Stockwell and Matz, 2007, p. 119-120). A positive violence climate is the establishment of policies and procedures to manage and prevent physical violence and verbal aggression in the workplace by the organization. In this respect, having a positive violence climate helps to eliminate and identify the main risk factors related to violence and aggression in the workplace as well as reducing existing violence and aggression (Kessler, Spector, Chang and Parr, 2008, p. 110)
In this context, this research aims to adapt the Violence Prevention Climate Scale developed by Kessler et. al. in 2008 into the Turkish language. The research was conducted voluntarily with a total of 279 people working in different companies operating in the service sector in Bursa, Turkey. There are 2 sections in the questionnaire form used in the research. The first part aims to determine the demographic characteristics of the participants (gender, marital status, educational age and work experience). In the second part, the Violence Prevention Climate Scale, which was developed by Kessler et. al. (2008), consisting of 18 items and 3 factors (policies and procedures, practices and response, pressure for unsafe practices) has been conducted. In the evaluation of the scale, a 6-point Likert-type assessment was used (1: strongly disagree; 6: strongly agree) and SPSS 22, LISREL 8.71 programs were used for data analysis.
According to the results, 53.8% of the participants were female, 55.2% were single and 45.5% had a high school education. The ages of the participants ranged between 17 and 54 and the mean age was 28±6.43. The duration of the work experience of the participants ranged from 1 to 35 years and the average experience duration was 5±5.33 years.
After the analysis, the data set was found to be suitable for explanatory factor analysis (KMO: 0.899; Barlett’s p: 0.00). According to the results of the exploratory factor analysis (the Extraction method was principal component analysis, the rotation method was Varimax with Kaiser Normalization), 18 items with factor loads ranging from 0.671 to 0.846 were collected in the 3 factors and these factors explained 72.45% of the total variance. In this case, the items on the scale exactly matched the factor structure in the original study.
According to confirmatory factor analysis, all paths between the item and the factor were significant (t>1.96) and the standardized factor loadings of all items were greater than 0.50. When the goodness of fit statistics are examined (Chi-square / freedom value: 3.78; RMSEA: 0.073; NFI: 0.95; NNFI: 0.96; CFI: 0.96; GFI: 0.90 and AGFI: 0.85), all of the values are above the acceptable goodness of fit statistics, so it is appropriate to say that the scale is validated with 3 sub-factors. In other words, it can be said that the scale for violence prevention scale is valid.
Cronbach’s Alpha was used for the reliability of the violence prevention climate scale. As a result of the analysis, the Cronbach’s Alpha value of the policy and procedures factor was 0.91; the applications and response factor was 0.93; the pressure factor for insecure practices was 0.91 and finally, the whole scale was 0.92. These values indicate that the scale is reliable.
According to the results of the analysis, the internal consistency coefficient does not decrease significantly if any of the items in the scale are deleted. Therefore, the 18-item scale was found to have a high level of reliability. Furthermore, item-total correlations were greater than 0.30 (item-total correlations ranged from 0.375 to 0.752).