Replicating future orientation: Investigating the constructs of hope and optimism and their subscales through replication and expansion

- Research has supported hope and optimism as uni- and bi-dimensional constructs.
- Unique sampling method collected data through Amazon's Mechanical Turk.
- Six competing, theoretically supported models compared.
- Hope and optimism are best explained by a bifactor model.

Recent decades have seen considerable debate as to how the constructs of hope and optimism are conceptualized, with some researchers questioning if they represent one global trait or two distinct traits. Past research has supported both the unidimensional view (Rand, 2009), and the claim they are two distinct entities (Bryant & Cvengros, 2004). The primary aims of the present study were to replicate the work of 1) Bryant and Cvengros (2004), who constructed several competing models and found that hope and optimism are best explained as two separate constructs, and 2) Rand (2009), who found hope and optimism to be part of one larger trait (“goal attitude”). Six distinct, theoretically-meaningful models were constructed and compared using confirmatory factor analysis to determine which model best fit data collected from 417 participants using Amazon's Mechanical Turk: (1) a one-factor hierarchical model, (2) a one-factor non-hierarchical model, (3) a two correlated factor model, (4) a four correlated factor model, (5) a bi-factor model, and (6) Rand's (2009) hierarchical model. Findings suggest that a bi-factor model, which includes both a global and a distinct two-factor component, best explains the constructs of hope and optimism. Implications of these findings are discussed.