Analysis of lst Semester of Classtalk use at McIntosh Elementary School by Fred Hartline, bE Inc. Overview: Classtalk was used in a multi-strategy effort to boost reading performance of 5th graders prior to the administration of a state-mandated" Literacy-passport" test of reading comprehension. The strategy succeeded far beyond expectations. The school was expecting 56% of the students to "pass" (i.e. test "at or above grade level'). They got 90%. The school has been a bit besieged with visitors trying to figure out how they did it. Quick profile of McIntosh Located about 5 miles from bE's offices; Part of the Newport News public school system;. K-5th grades; Greater than 70% minority; 79% of the students are "At risk"...which means -single-parent family, under poverty line, free meals; "At-risk" kids are street-smart, but tend to have great difficulty in school; Class size: 25-28 students (5th grade), 20 students (4th grade), 18 students (lst-3rd). McIntosh is a good model of an urban elementary school. Special reasons for working with McIntosh, Exceptional atmosphere for kids-- my judgment is that McIntosh tries harder and succeeds better at creating a special environment for learning than other schools I've visited. They're Blue Ribbon finalists--we'll see if they make it all the way this Spring; I have a 5-year working relationship with administration, teachers & students. This is an honest, open relationship w/ lots of trust, rapport, and good communication; Characteristics of the State "Literacy Passport" assessment Test is administered: mid-March '96, mid-May '97; Nationally normed test. State Law requires: -students must "pass" at 6th grade level or better to graduate from high school; -all students are tested in 4th and 6th grades; Newport News school district requires testing at 3rd and 5th grad s, Test includes Math, Reading, and Writing components...Reading test is called the "DRP" -- Degrees of reading Power; DRP consists of-- 7 short selections on novel subjects are to be read & analyzed in N-hours Each selection contains 7- fill-in-the-blank sentences. The student is to chose which one of 5 words best completes each sentence. DRP Test Score Norms (for 5th grade Mar '96)  Precentile  Year relative to grade level  88%  3-years above  77%  2-years above  65%  1-year above  50%  At-grade-level  41%  1-year behind  33%  2-years behind Classtalk at McIntosh-- Spring '96 Teacher: Mrs. Carol Wiatt (used Classtalk in her own AND other 5th grade sections) Subject: Reading proficiency (also teaches Language arts / Social Studies / Technology) Grade: 5th Classtalk intervention Details The goal is to help students successfully complete the DRP test w/ score of 50 or above; This is "teaching to the test"--the students NEED test-taking skills AND reading skills; The test environment (especially time-on-task) is difficult for many 5th graders (esp. "at risk"); The test is a reasonable assessment of reading comprehension; The intervention strategy relies heavily on Socratic methods of teaching & learning.   In-class Learning Strategies using Classtalk-- Students complete CLOZE worksheet (a DRP-like reading assignment); On completion, the class reviews answers and analyzes "why" some choices are better. -Students exchange papers for peer-grading; -Students send in their own answers using Classtalk; -Class looks at answer histogram; -Teacher asks "what's the best choice and WHY?" Students explain. Students keep individual charts of their performance over the semester; Students receive a "reward" (candy) if they've missed one or fewer answers; Teacher assigns more worksheets per session as the semester progresses to simulate the DRP test environment; Students learn-- To determine what reading style works best for each of them; To concentrate on reading of factual material and to work for extended periods; To understand how the sentence they are analyzing relates to the paragraph; To analyze which word best completes the sentence, relating it to the paragraph. Highlights of the Data-- Enormous improvements were seen in reading performance for students in instructional groups where Classtalk was used as part of the reading-skills development program; 84 of 91(92%) of 5th grade students "Passed" the DRP--the school had projected 60% would pass); Students in groups 1 & 2 (using Classtalk) advanced their reading skills more than those in Group 1: 52 of 60 (87%) advanced 2 or more years w/ Classtalk, compared to 8 of 15 (53%) w/out Classtalk;    2 day CT  1 day CT  no CT  Growth  Group 1  Group 2  Group 3  -1 yr  1  1    none  1      1 yr  5    7  2 yr  12  4  5  3 yr  11  5  3  4 yr  10  3    5 yr  4  3   Caveats this is NOT a true "experiment"--groups were selected to maximize educational gains for the neediest students. It would be unethical (and immoral) to deny a promising treatment for diagnosed reading problems for kids with this high a level of educational "risk"; Classtalk was one component of the school's effort to boost the reading scores of their "at risk" 5th grade students. Other intervention strategies and the skill of the teacher undoubtedly contributed substantially to the achieved gains. The Non-Classtalk group was not taught by Mrs Wiatt. DRP Outcomes for three groups of 5th graders Group I had Classtalk two or more days per week over the Spring semester; This group (2 classes) was not separated based on 4th grade performance;   Group 2 had Classtalk one day per week over the Spring semester This group was pulled for special treatment due to poor prognosis for DRP testing;   Group 3 had NO Classtalk This group was prejudged to have adequate prognosis for successful DRP testing;   Before Intervention (based on 4th grade DRPS)    Group 1  Group 2  Group 3  Selection  No special criteria  Poor prognosis  Good prognosis  Group size  44  16  15  Below grade level  20 (45%)  15 (94%)  0 (0%)  On-or-above grade level  12 (55%)  1 (6%)  15 (100%)  2-or-more yrs above grade  7 (16%)  1 (6%)  7 (47%) After Intervention (based on 5th grade DRPs)    Group 1  Group 2  Group 3  Selection  No special criteria  Poor prognosis  Good prognosis  CT use  2 or more days/wk  1 day/wk  NO Classtalk  Group size  44  16  15  Below grade level  4* (9%)  3* (81%)  0 (0%)  On-or-above grade level  40 (91%)  13 (81%)  15 (100%)   2-or-more yrs above grade  27 (61%)  7 (44%)  11 (73%)  No growth or regressed  2 (5%)  1 (6%)  0 (0%)  "Grew" 1 year  5 (11%)  0 (0%)  7 (47%)  "Grew" 2-3 years  23 (52%)  9 (56%)  11 (73%)  "Grew" 4-5 years  14 (32%)  6 (38%)  0 (0%) * just barely did not "Pass" DRP--but "grew" 2 years in reading ability The 5th grade included an additional 16 students who were excluded from the above analysis because the school did not receive 4th grade DRP scores for them:  Selection:  No special criteria  Poor prognosis  Good prognosis  CT use  2 or more days/wk  1 day/wk  NO Classtalk  Subgroup size  4  8  4  Finished below grade  0  0  0  Finished on or above grade  4  8  4  2-or-more yrs above  2  3  4 Publications List