Randomized experimental design
From WikEd
Select some research issue that you care about that could be addressed with a randomized experimental design and describe the design you might use. Sign your posting and separate it from the others with a horizontal line. jal
Possible focus-- Designing an experiment to evaluate the effects of using a constructivist approach to teacher professional development (TPD)
Possible design-- Randomly select a group of approx. 40 teachers. Among those teachers, randomly assign teachers to two groups-- A control group and an experiment group. Both groups would receive tpd around a particular instructional approach or academic content area. The control group would receive the relevant information through a traditional workshop in which the presenter stands and delivers information. The experiment group would participate in a professional development opportunity that uses a constructivist approach to immerse participants in building their own understanding of the focus area. Evaluation of this type of experiment often takes place through attitude assessments which might be done through a pre- and post-questionaire or pre- and post-interviews. Although this type of evaluation is valuable, I would like to see two changes. (1.) Questionaires and/or interviews should be arranged in a time series so that assessment occurs multiple times as the learning (or lack thereof) is implemented in the classroom. (2.) Assessment of student learning also needs to occur. This could be done through pre- and post-tests of students in the participating teachers classes. Design of these assessments would need to be carefully considered.
Heather
Research objective: To determine the affects on 3rd grade deaf and hard of hearing children accessing academic content through an interpreter who has worked for five years in an educational setting certified by RID versus one who is certified by the ESSE at a level 4.
Randomized experimental design: Using a list of all public schools in California that employ interpreters, randomly select approximately 20 schools. Randomly select two 3rd grade students from each school to participate in the study. Randomly assign twenty students to the control group and twenty to the experimental group. Deaf and hard of hearing students in both groups would watch a pre-recorded lecture showing both a hearing instructor and an interpreter. The control group would watch the lecture interpreted by an RID certified interpreter and the experimental group would watch the lecture interpreted by the ESSE certified interpreter.
Written pre and post tests on content could be administered, as well as surveys asking students about their level of comprehension about the content and comfort watching the interpreter.
Melissa
Randomized experiment – biliteracy, genre instruction & reading achievement
Testing hypothesis that reading achievement for ELL in L2 should be higher in dual language program plus genre instruction than in SEI plus genre instruction using same curriculum and instructional delivery (isolating variable of additive bilingualism).
Assumes: 1-dual language model builds on additive bilingualism 2-previous studies establishing efficacy of genre-focused instruction over regular curriculum alone in similar settings (single studies before comparative)
Selection: Use data on “alternative programs�? (CDE; CAL) to generate list of self-described dual language programs.
Randomly select X number of dual language programs for study and X number of SEI classrooms at same school site to keep same demographics.
Measures: • Pre and post-test reading assessments (from reading series), English & Spanish • Interviews with students to elicit resources they draw on (describe additive bilingualism, knowledge of genres in L1), attitudes toward language use (L1 & L2) • Interviews with teachers to elicit attitudes toward languages, knowledge about students
Implementation: • Professional development on program models, genre-based instruction • Common curriculum • Observational protocols to track fidelity to program model (language use time allocations; instructional practices – not mixing languages, etc.) and instructional practices (eg., genre-based instruction)
Constraints on sample: In CA, ELL students can only receive bilingual education through parent waiver under Prop. 227 – “self-selected�??
SES groups tend to be distinct in dual language classrooms (high SES for native English + low for ELL) and similar in SEI (all low SES, all English learners).
Size, willingness of districts/schools to participate in accountability context
Cheryl
