Designing an Experimental Action Research RICHELLE CONDE- DELOS SANTOS MT-1 26-29 April 2016 Civil Service Commission, Quezon City Session Objectives • Describe experimental research design • Differentiate three different types of experiments • Determine the appropriate type of experiment for given Action Research topics Action Research Template CONTEXT AND RATIONALE RESEARCH QUESTIONS INNOVATION, INTERVENTION RESEARCH METHODS WORK PLAN & COST ESTIMATES PLAN FOR DESSIMINATION REFERENCES Definition of Terms Variable – a characteristics that have different values Independent Variable – the variable that is applied on purpose (manipulation) by the Action Researcher, also known as the intervention, treatment, or cause Dependent Variable – the variable that is expected to change as a result of the implementation of the intervention or treatment, also known as the effect Definition of Terms Extraneous Variable – the variable that might differentially affect the dependent variable or the relationship between independent and dependent variable if not controlled Experimental Group – the group that receives the intervention or treatment, also known as treatment group Control Group – the group that does not receive the intervention or treatment Definition of Terms Randomization – the process of selecting participants for a study such that all members of the sampling space or pool of participants have equal chance of being included. Group Task Independent – Dependent Variable Chart 1. Complete the Chart by specifying the missing dependent or independent variables 2. Choose a presenter to present the output 3. Work within 5 minutes Group Task Pairs Independent Variables 1 Instructional method: Cooperative learning vs. Traditional 2 3 Dependent Variables Quarterly grades Type of in-school breakfast: Protein-rich vs. High carbohydrates 4 Length of attention span Note: Responses may vary Group Task Guide Questions: 1. What do you see as the common feature among the two(2) variables? 2. What then is the major difference of variables? Group Task SAMPLE RESPONSES Pairs 1 2 3 4 Independent Variables Instructional method: Cooperative learning vs. traditional Type of learning material: Digital vs. Print Type of in-school breakfast for Grade 1: Protein-rich vs. High carbohydrates Teaching aid: Videos vs. Photos Dependent Variables Achievement test scores Quarterly grades Weight after 3 months Length of attention span WORD BANK ACTIVITY! PRE-EXPERIMENTAL QUASI-EXPERIMENTAL TRUE-EXPERIMENTAL Group Task Word Bank Activity! Choose from the word bank on the board the name of the experiment described on the worksheets distributed. Select the type of experiment used given the purpose and condition in each situation. Group Task Guide Questions: 1. Compare the three situations and decide which one would be considered a preexperiment, a quasi-experiment, and a true experiment. 2. What do is/are the common feature/s among the three studies that makes them all experiments? 3. What then is the major distinction of experiments? Group Task Item No. 1 Purpose of Experiment to determine if the use of manipulatives will help Grade 8 students understand concepts in Geometry Details 2 to study the gains in performance of a class of Grade 10 student volunteers after an intensive 20-day summer workshop on English grammar 3 to determine if the integration of video clips in teaching Science is effective in improving the academic performance of one Grade 5 section, which is selected randomly from 10 heterogeneous sections, in comparison with that of another Grade 5 section, also selected randomly, that receives only the traditional lecture-discussion instructional approach application of the intervention (manipulatives) only two sections in Grade 8 Section Mahogany is assigned as experimental group; students will be using manipulatives in Geometry class Section Narra is assigned as control group; students will be taught Geometry without manipulatives only one group is involved composed of students who wanted to join the workshop and who lived close to the school implementation of the intervention (intensive workshop on English grammar) employment of video clips in instruction for one section randomly assigned as the experimental group; classroom lecture and discussion for the other section also randomly assigned as the control group Type of Experimental Research Word Bank Activity! Item No. 1 Purpose of Experiment to determine if the use of manipulatives will help Grade 8 students understand concepts in Geometry Details 2 3 to study the gains in performance of a class of Grade 10 student volunteers after an intensive 20-day summer workshop on English grammar to determine if the integration of video clips in teaching Science is effective in improving the academic performance of one Grade 5 section, which is selected randomly from 10 heterogeneous sections, in comparison with that of another Grade 5 section, also selected randomly, that receives only the traditional lecture-discussion instructional approach application of the intervention (manipulatives) only two sections in Grade 8 Section Mahogany is assigned as experimental group; students will be using manipulatives in Geometry class Section Narra is assigned as control group; students will be taught Geometry without manipulatives only one group is involved composed of students who wanted to join the workshop and who lived close to the school implementation of the intervention (intensive workshop on English grammar) employment of video clips in instruction for one section randomly assigned as the experimental group; classroom lecture and discussion for the other section also randomly assigned as the control group Type of Experimental Research Manipulatives-Quasi Experimental -NO CONTROL GROUP -NOT RANDOMLY ASSIGNED Summer Workshop-PreExperiment -NO CONTROL GROYP -NO RANDOMINATION Video Clips-True Experiment -RANDOMNIZATION -CONTROL GROUP IS PRESENCE TYPES OF EXPERIMENTAL RESEARCH DESIGN PRE-EXPERIMENTAL QUASI-EXPERIMENTAL TRUE-EXPERIMENTAL EXPERIMENTAL RESEARCH DESIGN • Is an Action Research design that determines causation or cause-and-effect relationship between variables. Its distinctive feature is the manipulation of an independent variable to see its effect on a dependent variable. Legend of the Procedures Where: R stands for random selection O1 stands for pretest O2 stands for posttest X stands for intervention PRE-EXPERIMENTAL • Pre-Experimental Studies follow the basic feature of an experimental research design, wherein an independent variable is manipulated to see its effect on a dependent variable. However, randomization is not applied in the selection of participants. There is also no control group. Pre-experimental design a. One shot case study. The procedure is as follows: X O Pre-experimental design b. One-group pretest-posttest design. The procedure is as follows: O1 X O2 QUASI-EXPERIMENTAL • Quasi-Experimental studies also involve the basic characteristic of experimental research design; an independent variable is manipulated to examine its influence on a dependent variable. Like pre-experimental studies, there is no random selection of participants in quasi-experimental studies. However, there is a designated control group . Quasi-experimental design a. Non-equivalent controlled group design. The procedure is as follows: O1 X O2 (experimental group) O1 O2 (control group) Quasi-experimental design b. Time-series design. The procedure is as follows: O1 O2 O3 X O4 O5 O6 Where: O1,O2,O3 stand for pretest (multiple observations) O4,O5,O6 stand for posttest (multiple observations) TRUE-EXPERIMENTAL • True experimental studies introduce an independent variable to assess its impact on a dependent variable. Like quasiexperimental studies, a control group is included in true experimental studies. In contrast though, randomization is utilized in the selection of participants. True experimental designs a. Pretest-posttest controlled group i. Subjects are randomly assigned to groups. ii. A pretest is given to both groups. iii. The experimental group receives treatment while the control group does not. iv. A posttest is given to both groups. Pretest-posttest controlled group design The procedure is summarized as follows: R O1 X O2 (experimental group) R O1 O2 (control group) b. Posttest only controlled group design i. Subjects are randomly assigned to groups. ii. The experimental group receives the treatment while the control group does not receive the treatment. iii. A posttest is given to both groups. Posttest only controlled group The procedure is summarized as follows: R X O2 (experimental group) R O2 (control group) c. Solomon four-group design It is considered as the most prestigious experimental design. It minimizes threats to both internal and external validity. i. Subjects are randomly assigned to one or four groups. ii. Two of the groups (experimental group 1 and control group 1) are pretested. iii. The other two groups (experimental group 2 and control group 2) receive routine or no treatment. Solomon four-group design The procedure is summarized as follows: R O1 X O2 (experimental group) R O1 O2 (control group) R X O2 (experimental group) R O2 (control group) TYPES OF EXPERIMENTS Pre-experiment Quasi-experiment True Experiment Manipulation of independent variable No random assignment of participants Manipulation of Manipulation of independent independent variable variable No random Random assignment assignment of of participants participants No control group Either there is no control group Has control group Individual Task Complete Me! 1. Read carefully the description study and complete the table distributed on the activity sheet. 2. Work within 5 minutes. GROUP TASK Description of Study Teacher A used advanced organizers in Grade 6 Science to improve student achievement in Section Mabini; Section Quezon was not given advanced organizers. Scores of the two sections in long quizzes were compared. Variables Control Group? Randomization? Type of Experiment GROUP TASK Description of Study Teacher B used PeerAssisted Learning Strategies (PALS) to improve the performance of the bottom 33% of her class. To assess the effectiveness of PALS, scores before and after their implementation were examined. Variables Control Group? Randomization? Type of Experiment GROUP TASK Description of Study To find the effectiveness of scaffolding, English Coordinator C randomly selected two of twelve heterogeneous sections of Grade 7 in a large school and assigned them to one of two conditions: one where scaffolding was used for every lesson and another where no scaffolding was applied. Their scores in the long examinations were evaluated. Variables Control Group? Randomization? Type of Experiment GROUP TASK Description of Study Variables Teacher A used advanced organizers in Grade 6 Science to improve student achievement in Section Mabini; Section Quezon was not given advanced organizers. Scores of the two sections in long quizzes were compared. Independent variable: Use or non-use of advanced organizers Control Type of Randomization? Group? Experiment Yes Dependent variable: Scores in long quizzes None Quasiexperiment GROUP TASK Description of Study Teacher B used PeerAssisted Learning Strategies (PALS) to improve the performance of the bottom 33% of her class. To assess the effectiveness of PALS, scores before and after their implementation were examined. Variables Independent variable: Application or nonapplication of PALS Dependent variable: Scores before and after PALS Control Type of Randomization? Group? Experiment None None Preexperiment GROUP TASK Description of Study Variables To find the effectiveness of scaffolding, English Coordinator C randomly selected two of twelve heterogeneous sections of Grade 7 in a large school and assigned them to one of two conditions: one where scaffolding was used for every lesson and another where no scaffolding was applied. Their scores in the long examinations were evaluated. Independent variable: Implementatio n or nonimplementatio n of scaffolding Control Type of Randomization? Group? Experiment Yes Dependent variable: Scores in long examinations Yes True Experiment THANK YOU! SAMPLING Sampling Method It is desirable that sampling method for experiments be random. However, since experimental Action Research are conducted to solve specific classroom problems, the sampling may have o be purposive and accessible. Data Collection Method The most common data collection methods in experimental Action Research are paper andpencil instruments that yield numerical data. These include tests, rating scales, and checklists. Observation and documentation, such as portfolios, may also be used as long as the data derived from these can be transformed to numerical values. Data Analysis Techniques Descriptive Statistics -Measures of central tendency or location, such as mode, median, and mean -Measures of variability or dispersion, such as range, variance, and standard deviation Data Analysis Technique Inferential Statistics a. Independent t-test – when comparing two means from two groups (e.g. Experimental vs. control) b. Paired t-test – when comparing two means from one group measured twice (e.g., pretest vs. posttest) c. Analysis of Variance – when comparing three or more groups or means Group Task 3 1. Formulate an Action Research Question using the following topics: -effectiveness of buddy system in improving test scores in Math -efficacy of self-eavaluatio in developing self-directed learning -usefulness of feedback through SMS in increasing the achievement levels of at-risk students 2. Using the research questions, complete the table Research Question Hypotheses: Type of Null and Experiment Research Independent Variable and Dependent Variable Group/s: Experimental or Experimental & Group Section of Participants: Random or Nonrandom References Center for Innovation in Research and Teaching. n.d. Type of Experimental Research. Arizona: Grand Canyon University. Accessed May12, 2018. https://cirt.gcu.edu/research/developmentresources/research_ready/ex perimental/design_types. Complete Dissertation.n.d.Experimental Research Designs.Statistics Solution. AccessedMay12,2018. http://www.statisticssolutions.com/experimentalresearchdesigns/. Trochim,WilliamM.K.(1999)2008.TheResearchMethodsKnowledge Base. Mason,OH:AtomicDogPublishing,Inc. “The heart and soul of good writing is research; you should not what you know but you can find out about.” - ROBERT J. SAWYER