1 / 19

Reading First: An Evidence-based school-wide r eading model

Analyzing the Reading Gap: Using MANOVA with discriminant group design to explore reading differences between young males and females. Reading First: An Evidence-based school-wide r eading model. Intervention Components Establish scientifically based K–3 reading programs

viho
Download Presentation

Reading First: An Evidence-based school-wide r eading model

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Analyzing the Reading Gap: Using MANOVA with discriminant group design to explore reading differences between young males and females

  2. Reading First: An Evidence-based school-wide reading model Intervention Components • Establish scientifically based K–3 reading programs • Provide teacher professional development • Select and administer screening, diagnostic, and classroom-based instructional reading assessments • Select and implement effective instructional materials, programs, learning systems, and strategies that are scientifically based and proven to prevent or remediate reading failure

  3. The Purpose • To explore how K-2 students perform on different reading components as differentiated by gender and grade level • How any exploratory analysis results can be considered in conjunction with research on reading development to make informed statements on reading instruction and assessment

  4. The Sample • RMC Research conducted the impact evaluation of a Reading First grant in a Mid-Atlantic State • In 2009 the grant served 21,310 K-3 grade students in 86 schools • Our sample includes 15,878 K-2 grade students • Grades each comprise 1/3 of the sample • 50% white, 42% African American, 8% either Hispanic, Asian, or undefined due to missing data • 61% of students were identified as economically disadvantaged • 9% Special Education

  5. What is MANOVA • Multivariate Analysis of Variance • Used to investigate differences among two or more groups on a set of two or more dependent variables

  6. Why use MANOVA? • Correlation among the dependent variables may be of substantive value • Determine what subsets of these dependent variables serve to separate groups • What are the underlying constructs for these subsets? • This analysis was run in SPSS, syntax is available on request

  7. MANOVA Assumptions • The subjects are randomly sampled from the target population. • The observations are statistically independent of one another. • The dependent variables follow a multivariate normal distribution within each group. • All groups have the same variance on each dependent variable. • The correlations between any two dependent variables are the same in all groups.

  8. Wilk’s Lambda - Λ • Wilk’s Lambda is an omnibus statistic that indicates if the groups differ on one or more linear combinations of dependent variables • Λ = |W|/|T| • W = within group variability • T = total group variability

  9. Linear Discriminant Functions (LDFs) • The linear combination of the dependent variables that the groups differ on • Groups will differ on at least one if Wilk’s Lambda is significant • An LDF is significant if determined by a chi-squared (χ2) test

  10. Research Question • How do males and females in younger grades (Kindergarten, 1st, and 2nd) differ on foundational reading skills?

  11. Analysis of the Reading Test 4 Dependent Variables Phonemic awareness, reading comprehension, reading fluency, and vocabulary development 6 Groups Males and females of the following grades: Kindergarten, 1st, and 2nd

  12. MANOVA statistics – Wilk’s Lambda • Λ= .68 • F(20, 52632) = 321.56, p < .001 • Therefore one or more linear combinations of the dependent variables differ across groups

  13. MANOVA – Linear Discriminant Functions • LDF1 – phonemic awareness • Statistically significant at Λ = .68, χ2 (20) = 6068.72, p < .001 • LDF2– comprehension and vocabulary • Statistically significant at Λ = .88, χ2 (12) = 2018.31, p < .001 • LDF3– reading fluency • Statistically significant at Λ = 1.0, χ2 (6) = 75.60, p < .001

  14. MANOVA with discriminant group analysis LDF3 MALES FEMALES LDF2 LDF2 - Comprehension and Vocabulary LDF3- Reading Fluency

  15. Questions Implicated by Results • Is this disparity in performance on comprehension and vocabulary vs. fluency supported by research and/or other reading assessment data? • If so, what are the implications for reading instruction and assessment? • If not, can we cite limitations of the assessment, possible deficiencies in instruction, or unique characteristics of this sample’s reading skills?

  16. Research and data on disparity of reading performance between males and females Findings • Biological/physiological factors • Environmental/educational factors • Reading skill development • Intelligence tests • National Assessment of Educational Progress (NAEP) • State Test Score Trends Through 2007-2008 Center on Education Policy, 2010; Chatterji, 2006; Mead, 2006

  17. Research and data on disparity of reading performance between males and females Implications • Kindergarten teacher awareness of male deficits in early reading skill development, but differences are small, temporary, and surmountable. • Implementation of procedures in 4th grade, and maybe 3rd grade, which enhance reading fluency in males. • Use of procedures such as altering reading material, providing choices and more reinforcement for reading to enhance males’ motivation to read during these grades. Below, Skinner, Fearrington, & Sorrell, 2010

  18. Possible limitations of the reading assessment • Measurement tasks • Format • Teacher subjectivity or bias

  19. MANOVA with discriminant group analysis Questions?

More Related