1. Brian Belland
  2. Associate Professor
  3. Impact of Scaffolding Characteristics and Study Quality on Learner Outcomes in STEM Education: A Meta-analysis
  4. http://itls.usu.edu/bbelland/grants/reese.php
  5. Utah State University
  1. Lindi Andreasen
  2. Impact of Scaffolding Characteristics and Study Quality on Learner Outcomes in STEM Education: A Meta-analysis
  3. http://itls.usu.edu/bbelland/grants/reese.php
  4. Utah State University
  1. Nam Kim
  2. Impact of Scaffolding Characteristics and Study Quality on Learner Outcomes in STEM Education: A Meta-analysis
  3. http://itls.usu.edu/bbelland/grants/reese.php
  4. Utah State University
  1. Mason Lefler
  2. Impact of Scaffolding Characteristics and Study Quality on Learner Outcomes in STEM Education: A Meta-analysis
  3. http://itls.usu.edu/bbelland/grants/reese.php
  4. Utah State University
  1. Andy Walker
  2. Impact of Scaffolding Characteristics and Study Quality on Learner Outcomes in STEM Education: A Meta-analysis
  3. http://itls.usu.edu/bbelland/grants/reese.php
  4. Utah State University
  1. Brett Whitney
  2. Impact of Scaffolding Characteristics and Study Quality on Learner Outcomes in STEM Education: A Meta-analysis
  3. http://itls.usu.edu/bbelland/grants/reese.php
  4. Utah State University
Public Discussion

Continue the discussion of this presentation on the Multiplex. Go to Multiplex

  • Icon for: Elizabeth Hassrick

    Elizabeth Hassrick

    Facilitator
    Research Scientist
    May 11, 2015 | 01:10 p.m.

    This study addresses a very key question about how computer assisted learning might be designed to support rigorous, ambitious STEM instruction. To begin this effort, this project is providing a much needed meta-analysis about the effectiveness of computer-based scaffolding. More details about how you conducted meta-network analysis would be useful. Could you provide some preliminary findings to your exciting questions? (What forms are best? Quality of research? Being applied as originally intended?)

  • Icon for: Brian Belland

    Brian Belland

    Lead Presenter
    Associate Professor
    May 12, 2015 | 11:21 a.m.

    at a global level, in a network meta-analysis, every included study needs to have pretest and posttest scores that results from tests that measure the same construct. So if your pretest is about botany knowledge and facts and then your posttest is about problem solving, then that does not work. then when you run the stats, you are not comparing all of your interventions to control conditions on the basis of a posttest, but you are making comparisons on the basis of pre-post gains among all of the interventions and control conditions. this is oversimplified to be sure, because my co-PI Andy Walker and one of my GAs NamJu Kim are more knowledgeable about the nitty gritty details of network meta-analysis.

    for our traditional meta-analysis to date, we found that scaffolding actually has the highest effect sizes among college and graduate level learners, there is no difference in effect sizes on the basis of the presence or lack thereof of scaffolding customization (fading, adding, or fading/adding), there is no difference between generic and context-specific scaffolding, and scaffolding’s effect size is higher when assessed at the principles level, but had substantial effect sizes that were statistically greater than zero at the concept, principles, and application levels.

  • Icon for: Lisa Hogan

    Lisa Hogan

    Facilitator
    Technology Integrator
    May 11, 2015 | 08:18 p.m.

    This study of a meta-analysis of computer-based scaffolding to support complex problem solving seems to incorporate a very wide range of learners, i.e. K-12, college, graduate, and adult learners. Do your findings show effective ways to use computer assisted learning have consistent characteristics through a very wide range of learners, i.e. K-12, college, graduate, and adult learners? Or, do your findings show there are specific effective ways to use computer assisted learning for specific learners such as k-3 learners versus college level learners?

  • Icon for: Brian Belland

    Brian Belland

    Lead Presenter
    Associate Professor
    May 12, 2015 | 11:24 a.m.

    we have not had the opportunity to drill down to make comparisons akin to what you asked about among the different populations. We are currently coding studies from 2014, and once that is done, we should be able to conduct some additional analyses.

  • Icon for: Avron Barr

    Avron Barr

    Facilitator
    Consultant
    May 12, 2015 | 11:05 a.m.

    I agree that computer-based scaffolding is an important component of our efforts to improve STEM education. However, our “general public” audience might need some help understanding the issues, including some background on words like scaffolding, meta-analysis, and effect size. The voice-over-PowerPoint type of video also seems rather academic and less suited for a general audience. How could you explain the importance of this work to non-researchers?

  • Icon for: Brian Belland

    Brian Belland

    Lead Presenter
    Associate Professor
    May 12, 2015 | 12:14 p.m.

    Scaffolding assists students as they generate solutions to ill-structured tasks. It helps extend student capabilities such that they can meaningfully participate in solution generation. Computer-based scaffolding is scaffolding that is delivered via a computer tool.
    Meta-analysis is a strategy for systematically reviewing quantitative studies that meet certain criteria. It helps one see patterns in the pool of research studies more effectively than just looking at each study individually.
    Effect size can be either the standardized difference in a posttest between an experimental and control group, or standardized pre-post gains. Standardized means that the difference is in some common metric, like standard deviation units

  • Icon for: Kevin Brown

    Kevin Brown

    Senior Research Scientist
    May 13, 2015 | 09:47 a.m.

    Very timely project and crystal-clear presentation! I’m very interested in future findings about the kinds of scaffolding that work best.

  • Icon for: Gregory Moore

    Gregory Moore

    Doctoral student
    May 14, 2015 | 05:23 p.m.

    While it was probably out of the scope of a 3-minute video, I am wondering how you chose studies to include in your meta-analyses. What were your inclusion and exclusion criteria? It sounds like you were casting a wide net (e.g., Kindergarten to adult), but I would like to know for sure. Scaffolding is definitely an important topic, so I’m glad you are approaching it from a slightly different perspective.

  • Icon for: Brian Belland

    Brian Belland

    Lead Presenter
    Associate Professor
    May 15, 2015 | 04:41 p.m.

    Inclusion criteria were:
    To be included in this meta-analysis, studies had to: (a) be published between January 1, 1993 and December 31, 2014, (b) provide cognitive outcome statistics for both a control and an experimental condition, © provide sufficient information for calculating (the) effect size (s), (d) describe a scaffolding condition (Wood, Bruner, & Ross 1976) that was substantially different from that of the control, and (e) deliver scaffolding directed towards problem solving in STEM.
    Studies were excluded if they did not meet the inclusion criteria; if two or more studies included the same data, we kept the one with the most data (e.g., dissertation).

  • Further posting is closed as the showcase has ended.