Porthole
Scale Trimming and Validating: An Effective Short-form of the UCLA Loneliness Scale (307854)
*Jinyuan Liu, University of California, San DiegoKeywords: scale development, UCLA Loneliness Scale, item reduction analysis
Scales are typically used to capture multifaced constructs such as a behavior, or an action that cannot be captured in a single variable or item. The use of multiple items helps to measure an underlying latent construct to account for item-specific measurement error, which leads to more accurate research findings.
Loneliness has been considered the latest global health epidemic, affects nearly two-thirds of older adults and has serious health implications, including depression, cognitive impairment, and frailty.
There is no consensus definition of loneliness across the world, and measures of loneliness have been based on the phenomenological perspective of the individual. The most commonly used multiple-item scale is the UCLA Loneliness Scale, which includes 20 items and has high internal consistency, and validity. However, the response burden is a concern, and some short forms of this scale will be as valid as the full length original one.
In this paper, we will adopt both Classical Test Theory (CTT) and Item Response Theory (IRT), the two major theories in scale development, and conduct item reduction analysis to trim the original scale into an effective short-form.