Predicting Science Literacy: A Multiple Regression Model of Factors That Influence Science Literacy

Document Type


Degree Name

Doctor of Education (Ed.D)


Curriculum and Instruction

Date of Award

Summer 2020


Academic scholarship has focused on the development of an operational definition (Bybee, 1997; Laugksch, 2000; Miller, 1998), classification system (Roberts, 2007), or intervention (Allchin, Andersen, & Nielsen, 2014) of science literacy. The present study investigated the relationship between science literacy, as measured by the Test of Science Literacy Skills, and predictive factors commonly collected in secondary schools. This quantitative correlation study examined the relationship between variables commonly collected by schools with science literacy performance to quantify the impact of and target intervention at variables most likely to impact science literacy. Predictive factors included Gender, Ethnicity, Economic Need, English Proficiency, Number of Science Courses Completed, Biology Content Knowledge, English Language Arts Content Knowledge, Algebra Content Knowledge, and Attitude Toward Science. Participants were selected through convenience sampling from a single suburban high school in the southwestern United States. A hierarchal, multiple regression model was developed using the Theory of Reasoned Action and Planned Behavior framework (Fishbein & Ajzen, 2010). The model predicted science literacy based on distal, opportunity, propensity, and demographic factors. A significant regression equation was found. Results suggest that students’ science literacy scores can increase by increasing the number of science courses complete, facilitating interventions via English Language Arts, and supporting students’ development of a positive attitude toward science.


Gilbert Naizer

Subject Categories

Education | Elementary Education