DescriptionPurpose: The goal of the study was to test an intervention using a brief essay as an instrument for evaluating higher-order information literacy skills in college students, while accounting for prior conditions such as socioeconomic status and prior academic achievement, and identify other predictors of information literacy through an evaluation of student behavior and attitude. Design/Methods/Approach: An instructional intervention was evaluated using a brief essay as a pre- and posttest of learning in a course in technical communication. Multiple readers rated essays on five criteria to measure higher-order skills. Interrater reliability and internal consistency of the measures were tested. Analyses of variance and covariance were used to measure academic gains and to partial out the effects of confounding variables. Student behavior was measured by level of activity in the course management system and essay length. Student attitude was measured through a content analysis of their reflective statements. A control group of students who took the same course without the intervention, but who did not take the pretest, also took the posttest. Findings: 1) The method used for measuring information literacy was found to be reliable and valid. 2) The use of the brief essay as a pre- and posttest showed that the students in the treatment group achieved impressive gains in higher-order skills associated with information literacy. 3) The students in the treatment group significantly outperformed students in the control group with substantive effect sizes explaining results. 4) Socioeconomic status had no significant impact on information literacy. 5) Student use of online instructional materials had no significant impact on information literacy. Originality/Value: A model of information literacy assessment in higher education was proposed to isolate important classes of variables affecting learning. An experimental design using multivariate methods to account for the multiple influences of variables on information literacy allowed for the determination and partitioning of the influence of each variable and sets of variables. This knowledge allows for efficient and systematic progress to be recorded where less productive variables can be dropped from the model and significant and important variables are kept in the model to increase the amount of variability explained in information literacy outcomes.