Learning Objective #1: The learner will be able to identify statistical tests for psychometric properties of an instrument.

Learning Objective #2: The learner will be able to critic the relevance of an instrument before applying it to a different population.

---

Organizational climate (OC) is an important measure to predict nurses' intention to leave their current job. The Litwin and Stringer's Organizational Climate Questionnaire (LSOCQ) has been widely used to measure OC in industrial settings, but not in healthcare settings. The purpose of this study was to test psychometric properties of the Chinese LSOCQ (C-LSOCQ) when used on hospital nurses in Taiwan. Research question was “what are the reliability and validity of the C-LSOCQ?” The C-LSOCQ (50 items) was tested with 492 nurses in 8 hospitals in Taiwan. Three nurses' data were excluded due to high missing rate. Of 489 participants, 75.5% were younger than 36, 51.6% were single, 48.1% worked within 4 years, and 64% worked in public hospitals. The reliability of the C-LSOCQ was tested with Cronbach's alpha and item-total correlation. Factor analysis was used for validity. Results showed a satisfactory reliability (alpha=.87). The item-total correlation coefficients were satisfactory (.23-.56) except for 12 items (<.20). Fifteen factors were emerged and explained 42.18% of the variance of OC. Four factors contained only 1 item. Excluding those 12 items with low item-total correlation, 11 factors were emerged, explained 42.40% of the variance of OC, and the alpha=.89. These factors were named affective tone toward organization (alpha=.75, 9 items), policy/authority (alpha=.71, 6 items), interpersonal milieu (alpha=.69, 4 items), responsibility (alpha=.70, 3 items), organizational identity (alpha=.73, 2 items), openness (alpha=.58, 2 items), structure/proceeds (alpha=.64, 3 items), reward (alpha=.51, 2 items), affective tone toward people (alpha=.58, 3 items), belonging (alpha=.48, 3 items), and conflict (1 item). Factors emerged in this study were different from those in the original instrument. Different population and cultural background may cause this difference. Items that showed low item-total correlation and other types of validities should be examined further. The C-LSOCQ may need to be modified to be culturally relevant.