Factor Structure of the Urban Hassles Index

Background: This study examined the factor structure and psychometric properties of the Urban Hassles Index (UHI), a self-report measure of social and environmental stressors that may affect urban adolescents. Previous research has reported inconsistent findings of its underlying factor structure, with limited examination of its construct validity.

Methods: Participants were 341 adolescents (158 boys, 183 girls), primarily African American (82%), of low socioeconomic status, participating in a longitudinal, prospective study from birth. The UHI was administered at age 15 and includes 31 items describing hassles experienced in the last 3 months, with frequency rated using a 4-point Likert scale.  Higher scores indicate greater hassles.  Exploratory factor analyses (EFA) were conducted to identify the underlying factor structure of the UHI via Principal Axis Factoring extraction method.  Number of factors was determined based on parallel analysis, a scree plot, and magnitude of factor loadings (≥ .40). Both orthogonal and oblique rotations were performed.  Confirmatory factor analyses (CFA) were conducted using Mplus 7.11 to evaluate the model fit of the factor structure derived from EFA.  Due to the polytomous nature of the data and skewness in some variables, the weighted least squares estimator with mean and variance adjustments (WLSMV) was used. Composite/scale scores were created for each of the extracted factors. Construct validity was examined using hierarchical linear or logistic regressions on the following variables as criteria: neighborhood safety from the Developmental Assets Profile and internalizing and externalizing problems from the Youth Self Report assessed at age 15, and substance use and mental health problems from the Problem Oriented Screening Instrument for Teenagers at age 17.

Results: EFA yielded a final model comprising 16 items loading on four factors, accounting for 43% of the total variance.  Orthogonal rotation was selected due to low-to-moderate factor correlations (r = 0.15-0.35).  The first factor (5-item; α= .81) was named Environment (e.g., abandoned buildings and lots), contributing 15% of the total variance.  The second (6-item; α= .68) and third factors (3-item; α= .71) were named Safety (e.g., worried about the safety of friends) and Drug (e.g., asked to sell drugs), respectively, each explaining 11% of the variance.  The fourth factor (2-item; α=.54) was named Coercion (e.g., pressured for sex by boyfriend/girlfriend) and contributed 6% of the variance.  CFA yielded excellent model fit, Chi-square=126.94 (99), p=.03; CFI=0.98; TLI=0.98; RMSEA=0.029 (90% CI=0.009-0.043); WRMR= .756, supporting the 4-factor structure of the UHI. After controlling for socio-demographic variables, higher levels of Environment and Safety were related to poor neighborhood safety; Safety and Coercion were associated with more internalizing problems; Environment, Safety, and Coercion were associated with more externalizing problems and mental health problems; and Drug was associated with substance use problems, resulting in 10-25% increases in R-square in each criterion.

Implication: The UHI captures contextual stressors and challenges germane to urban, inner-city adolescents. Although additional items for Coercion may improve reliability and validity of the UHI, the shortened 16-item UHI will be a sound rapid assessment tool assessing stressors unique to urban adolescents.