New windows into a broken construct: A multilevel factor analysis and DIF assessment of perceived incivilities. (July 2017)

• Record Type: Journal Article
• Authors: Ward, Jeffrey T.; Link, Nathan W.; Taylor, Ralph B.
• Is Part Of: Journal of criminal justice. Number 51(2017)
• Page Start: 74
• Page End: 88
• Publication Date: 2017-07
• Subjects: Incivilities -- Disorder -- Neighborhood -- Broken windows -- Differential item functioning -- Multilevel CFA -- Multilevel MIMIC; Criminal justice, Administration of -- Periodicals; Justice pénale -- Administration -- Périodiques; 364.05
• DOI: 10.1016/j.jcrimjus.2017.06.004
• Languages: English
• ISSNs: 0047-2352

Abstract

Abstract: Objectives: (1) To determine whether perceived physical and social incivilities are distinguishable at the individual and/or neighborhood levels and, if so, whether there are differences in effects on fear of crime. (2) To identify, characterize, and account for differential item functioning (DIF) to understand differences in subjective perceptions of incivilities across demographic groups. Methods: This study uses data from a probability sample of 1622 residents nested within 66 ecologically valid neighborhoods and employs multilevel SEM to identify factor structure, assess DIF, and examine structural relations at individual and neighborhood levels. Results: Physical and social incivilities are distinguishable at the individual level but not at the neighborhood level. Three physical incivilities items exhibit DIF for race and three social incivilities items exhibit DIF for age. Residents in neighborhoods with higher concentrations of African Americans report greater levels of combined incivilities, but, within neighborhoods, African Americans perceive lower levels of physical and social incivility. Within neighborhoods, social incivilities link to fear of crime. Conclusions: Demographic factors affect how individuals use response categories for gauging perceived incivilities in their locale. Discriminability of underlying separate physical and social components only at the individual level points to needed areas of theoretical elaboration in incivilities models. Highlights: Individual level social and physical incivilities perceptions are separable. Neighborhood level social and physical incivilities perceptions are not separable. Three items exhibit DIF across race and three items exhibit DIF across age. Race influences incivilities perceptions across levels of analysis differently. Combined neighborhood incivilities and individual social incivilities link to fear.