Reporting Confidence Intervals and Effect Sizes: Collecting the Evidence
Confidence intervals (CIs) and effect sizes are essential to encourage meta-analytic thinking and to accumulate research findings. CIs provide a range of plausible values for population parameters with a degree of confidence that the parameter is in that particular interval. CIs also give information about how precise the estimates are. Comparison of CIs across various samples cumulates evidence for more accurate estimates of population parameters than an estimate from a single sample. Effect sizes, on the other hand, provide information about the magnitude, or the substantive significance. That is, effect sizes provide information about sizes of differences between groups or strengths of relationships. The current paper's purpose is to inform researchers about the importance of reporting and interpreting CIs and effect sizes using examples from studies published in the Career and Technical Education Research journal. In addition, calculations and interpretations of CIs around means and effect sizes for some commonly used statistical analyses are explained.
Document Type: Research Article
Publication date: January 1, 2012
More about this publication?
- (CTER) publishes refereed articles that examine research and research-related topics in vocational/career and technical education, career development, human resource development, career issues in the schools (Grades K-12), postsecondary education, adult and lifelong learning, and workforce education. The CTER Editorial Board is committed to publishing scholarly work that represents a variety of conceptual and methodological bases. Submission of manuscripts representing one of the following styles is encouraged: (a) empirically-based manuscripts that report results of original research, either quantitative or qualitative, (b) reviews or synthesis of empirical or theoretical literature, (c) essays derived from original historical or philosophical research, (d) reviews of recently published books, and (e) rejoinders to articles recently published in CTER. CTER will consider for publication papers initially presented at conferences, including those disseminated through conference proceedings.
To view issues before 2005, please click here.
- Editorial Board
- Submit a Paper
- Membership Information
- Ingenta Connect is not responsible for the content or availability of external websites
- Access Key
- Free content
- Partial Free content
- New content
- Open access content
- Partial Open access content
- Subscribed content
- Partial Subscribed content
- Free trial content