Statistical Literacy in Adult College Students

Open Access
Author:
Wade , Barbara
Graduate Program:
Adult Education
Degree:
Doctor of Education
Document Type:
Dissertation
Date of Defense:
February 24, 2009
Committee Members:
  • Dr Ed Taylor, Dissertation Advisor
  • Edward W Taylor, Committee Chair
  • Patricia Angelica Cranton, Committee Member
  • Pamela Preston, Committee Member
  • John Basil Urenko, Committee Member
Keywords:
  • statistical literacy
  • adult education
  • statistics
  • attitudes and statistics
  • beliefs and statistics
  • research methods and statistics
Abstract:
ABSTRACT Currently statistics is used in society as social indicators that measure, for example crime rates, or research concerning health and medical issues, which are reported by the media. Educated citizens need to be able to understand statistics; it is assumed that adult students who graduate from college will have this ability. Knowing how important statistical literacy is, the purpose of this research was to measure statistical literacy in adult learners before and after they have completed a statistics class, or a research methods class with no prior statistics, or a research methods class with prior statistics. Participants were 110 adult students, 74 females and 37 males, 72% Caucasian, 26% African American, 1% Native American and 1% Asia/Pacific Islander; ages ranged from 18 to 40 years old (M = 21, S.D. = 3.25), with most reporting marital status as single and being full-time students. Using a quasi-experimental research design, adult students completed pre- and post-test surveys, which measured statistical literacy, based on Gal’s (2004) Model of Statistical Literacy that embraced both knowledge and dispositional elements. Fisher’s LSD post-hoc test results showed a statistically significant difference among class types for the knowledge elements, statistical thinking, reasoning, and literacy, but no statistically significant differences for the critical questions. Results from a MANOVA on post-test scores among class types showed no significant differences on the dispositional elements, affect, cognitive competence, difficulty and value. For beliefs, a Chi-square analysis showed that 90% of adult students in the research methods class with prior statistics believed that statistics was relevant in their lives. More importantly, further tests that examined pre- to post-test scores revealed significant differences among class types. Implications for policy include a re-examination of prerequisites for statistics courses, and requiring completion of a statistics course before a research methods course. Many teaching implications are indicated in this research; however, the most important is the use of constructivist perspectives in the classroom; accordingly, future research should examine teaching methodologies. In addition, statistical literacy needs to be examined again; however, using modifications to the model that separates statistics into its two main branches, descriptive and inferential.