Uji Validitas Dan Reliabilitas Alat Ukur Organizational Citizenship Behavior (OCB)

Abstract

Organizational Citizenship Behavior (OCB) merupakan perilaku sukarela yang tidak tercantum dalam deskripsi kerja formal, namun berkontribusi penting terhadap efektivitas organisasi. Instrumen OCB dua dimensi yang dikembangkan Williams dan Anderson (OCB-I dan OCB-O) telah banyak digunakan, termasuk penelitian adaptasi di Indonesia. Namun, hasil sebelumnya menunjukkan perlunya pengujian ulang agar instrumen ini sesuai dengan konteks budaya kerja Indonesia. Penelitian ini bertujuan menguji kembali validitas dan reliabilitas instrumen OCB dua dimensi pada pekerja di Indonesia. Penelitian menggunakan pendekatan kuantitatif dengan desain cross-sectional survey. Sebanyak 240 pekerja aktif berpartisipasi melalui teknik purposive sampling. Analisis data meliputi analisis item, uji reliabilitas menggunakan Cronbach’s Alpha dan item-rest correlation, serta Confirmatory Factor Analysis (CFA). Hasil menunjukkan reliabilitas keseluruhan cukup (α = 0.734), dengan variasi reliabilitas per dimensi yaitu OCB-I (α = 0.555) dan OCB-O (α = 0.698). Item-rest correlation berada dalam kategori memadai. Namun, model belum memenuhi kriteria model fit (CFI = 0.716; TLI = 0.660), dan beberapa item memiliki muatan faktor rendah (<0.30), sehingga memerlukan revisi atau penghapusan. Penelitian ini menyimpulkan bahwa instrumen OCB dua dimensi memiliki potensi digunakan, namun tetap memerlukan modifikasi agar lebih sesuai dengan karakteristik budaya Indonesia.

Organizational Citizenship Behavior (OCB) refers to voluntary employee behaviors that are not formally included in job descriptions but significantly contribute to organizational effectiveness. The two-dimensional OCB scale developed by Williams and Anderson (OCB-I and OCB-O) has been widely applied, including in Indonesian research. However, previous findings indicate the need to reassess its psychometric properties to ensure suitability within the Indonesian cultural context. This study aims to re-evaluate the validity and reliability of the two-dimensional OCB scale among Indonesian employees. A quantitative design with a cross-sectional survey was used. A total of 240 full-time employees participated using purposive sampling. Data were analyzed through item analysis, reliability testing using Cronbach’s Alpha and item-rest correlation, and Confirmatory Factor Analysis (CFA) to assess construct validity. The results showed acceptable overall reliability (α = 0.734), with variations across dimensions, namely OCB-I (α = 0.555) and OCB-O (α = 0.698). Item-rest correlations were within an acceptable range. However, the CFA results indicated that the model did not meet recommended fit criteria (CFI = 0.716; TLI = 0.660), and several items had low factor loadings (<0.30), requiring revision or removal. In conclusion, the two-dimensional OCB scale demonstrates potential for use in Indonesia, yet requires refinement to enhance cultural fit and psychometric quality