A categorical examination of strain on college campuses



Journal Title

Journal ISSN

Volume Title



Previous research has demonstrated a link between academic strain and intimate partner violence, specifically psychological aggression (Mason & Smithey, 2011). Though this test of Merton’s Strain Theory (1938) on intimate partner violence on a college campus measured strain in several ways, including cumulative strain as the passage of time in potentially stressful situations and economic strain, the primary tool relied on for the measurement of academic strain was the College Undergraduate Stress Scale (Renner & Mackin, 1998). This item yielded an aggregated measure of general strain; however, this general strain score failed to discuss potential categorical differences in types of strain. As a part of continued research into the effect of strain upon intimate partner violence on college campuses, this thesis aims to further explore and develop a diverse, multi-dimensional model of strain, rather than presenting an inaccurate portrayal of academic strain as a single, aggregate variable. In order to analyze the degree to which strain-causing items fall into theoretically distinct categories, additional survey data were collected from 352 Texas Tech students in a mixture of upper and lower-division classes. The College Undergraduate Stress Scale (CUSS) was administered to measure academic strain, while the Conflict Tactics Scales II (Straus et al. 1996) is used to measure the perpetration of intimate partner violence. Factor analysis is performed in order to disaggregate the CUSS into a number of unique strain categories. Findings reveal three primary categories of strain: 1.) Academic Strain, or strain associated with a difficult semester, such as “Finals Week” or “Two exams in one day”, and 2.) Peer Strain, such as “Drinking or use of drugs” and “Peer pressures”, and 3.) Unanticipated Strain, or strain that was not expected, such as “Lack of Sleep” or “Getting sick” (Renner & Mackin, 1998). Cronbach’s alpha reveals a fair amount of internal consistency within these subscales, with scores at 0.75, 0.65, and 0.57, respectively. Though the Unanticipated Strain subscale shows a lower than desirable degree of internal consistency, there is a qualitative similarity between the items “Lack of sleep,” “Change in housing situation (hassles, moves),” “Difficulties with a roommate,” “Commuting to campus or work, or both,” “Getting sick,” and “Attending an athletic event (e.g., football game).” Combined with a reasonably high degree of internal consistency, this qualitative consistency justifies the inclusion of these items into a unique category of strain as measured by this scale. The unique ways in which these categories of strain correspond with an individual’s race, religion, belief in traditional gender roles, and gender are discussed. In addition, the ways in which these different categories of strain interact with the perpetration of intimate partner violence are explored. After controlling for outside influences, analysis using OLS regression reveals no relationship between strain and intimate partner violence; however, strain is positively associated with increased negotiation within the relationship. In addition, the number of years spent in the relationship are positively related to both the perpetration of psychological aggression and increased rates of negotiation. Results are discussed.