Electronic Thesis and Dissertation Repository

Degree

Master of Science

Program

Psychology

Supervisor

Paul Tremblay

Abstract

Prominent measures of masculinity focus on traditional masculine norms, such as high aggression, low emotional expression, and heteronormativity. However, recent qualitative research has indicated that a variety of men embrace alternative forms of masculinity that include unique characteristics not represented by traditional norms. I developed the Alternative Masculinity Measure (ALT-M) to address this gap. The ALT-M was designed to measure individual differences on constructs derived from a modern, socially progressive representation of masculinity. Concepts, scales, and items were developed primarily from readings of qualitative research on alternative masculinities. Nine dimensions with 14 items each was sent to 15 experts for content validity assessment. A final pool of 101 items distributed across 9 constructs (10 -12 items per scale) was distributed to undergraduate males at Western University (N = 497). Participants also completed the Conformity to Masculine Norms Inventory (CMNI-46; Parent & Moradi, 2009), the Aggression Questionnaire (Buss & Perry, 1992), the Trait Emotional Intelligence Questionnaire – Short Form (TEIQue-SF; Petrides, 2009) and the HEXACO Personality Inventory-short version (Ashton & Lee, 2009) for construct validation purposes. Item descriptive statistics and separate factor analyses were inspected for each construct to evaluate unidimensionality. One scale was removed due to its inability to adequately represent a homogeneous set of items. The other 8 scales displayed evidence of unidimensionality and were analysed together using a combination of Exploratory Structural Equation Model (ESEM) with a target rotation and CFA. Weak items, defined as items with loadings below .30 on their hypothesized scales or high loadings on other scales were removed one at a time using an iterative process. During this process, another full scale and multiple items were removed. In the end, I retained 7 scales with 6 items each, for a total of 42 items. All items load strongly on their scales (range .414 - .823) and have low cross-loadings on other scales. The seven remaining scales are labelled Homophilia, Emotional Openness, Cooperation, Avoidance of Physical Aggression, Reticence, Gender Egalitarian, and Intimacy Orientation. A higher order factor structure reveals all scales load significantly onto a latent “progressive masculinity” construct, loadings range from .506 - .771. Final CFA and ESEM models show good level of fit. Average scores for the scales and full measure were created and compared to the scale and full measure scores of the CMNI, revealing a good level of convergent and discriminant validity. In order to evaluate the distinctiveness of the ALT-M from the CMNI, these two instruments were compared in terms of their relationships with the HEXACO, Aggression Questionnaire, and Emotional Intelligence Questionnaire. Strong evidence of construct distinctiveness is provided. Finally, a Latent Profile Analysis was conducted on the measure, revealing four unique masculinity profiles that differ in shape, ethnic composition, and trait aggression levels. Overall the ALT-M has sound psychometric properties and can be used to delineate multiple unique masculinities. It will benefit from future cross validation in terms of its factorial structure and profiles.

Share

COinS