P. 1
Presenting Research in Education

Presenting Research in Education

|Views: 0|Likes:
Published by yusuf hidayat

More info:

Published by: yusuf hidayat on Jan 01, 2013
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PPT, PDF, TXT or read online from Scribd
See more
See less

06/12/2015

pdf

text

original

EXPERIMENTS, QUASIEXPERIMANTS, SINGLE-CASE RESEARCH AND META-ANALYSIS

BY YUSUF HIDAYAT (2003512120)

POSTGRADUATE PROGRAM STATE UNIVERSITY OF SEMARANG

Introduction
The essential features of experimental research are:  The investigators deliberately control and manipulate the conditions which determine the events;  Making a change in the value of one variablecalled the independent variable;  Observing the effect of that change on another variable-called the dependent variable.;  Seeking to support or not support a null hypothesis;

Designs in educational experimentation There are several different kinds of experimental design. (Cohen et al.  the natural experiment. for example:  the controlled experiment (the ‘true’ experiment): two or more groups..  The field or quasi-experiment. 2007:274) .

 the two control groups and one experimental group pretest-posttest design.  repeated measures designs.  the posttest two experimental groups design.  the matched pairs design. (Cohen et al.  the factorial design..  the parametric design.  the posttest control and experimental group design.True experimental designs There are several variants of the ‘true’ experimental design. 2007:275) . and we consider many of these below:  the pretest-posttest control and experimental group design.  the pretest-posttest two treatment design.

 pretest of the groups to ensure parity.  posttest of the groups to see the effects on the dependent variable.  random allocation to control and experimental groups.A ‘true’ experiment includes several key features:  one or more control groups.  one or more interventions to the experimental group(s).  isolation.  non-contamination between the control and .  one or more experimental groups. control and manipulation of independent variables.

Quasi-experiment designs.  the time series design.  the non-equivalent control group design. include:  the one-group pretest-posttest. .

The pretest-posttest control and experimental group design The pretest-posttest control group design can be represented as: Experimental RO1 χ O2 Control RO3 O4 .

The two control groups and one experimental group pretest-posttest design This design can be illustrated as follows: Experimental RO1 χ O2 Control1 RO3 O4 Control2 χ O5 .

The posttest control experimental group design The design is: Experimental R1 χ O1 Control R2 O2 and .

The post-test two experimental groups design The design is: Experimental1 R1 χ1 O1 Experimental2 R2 χ2 O2 .

The pretest-post-test two treatment design The design is: Experimental1 RO1 χ1 O2 Experimental2 RO3 χ2 O4 The other design might be: Experimental1 RO1 χ1 O2 Experimental2 RO3 χ2 O4 Control RO5 O6 .

if appropriate. Ensure that the control group is not subject to the intervention. 5. based on the scores and measure established from step 1. Randomly assign one person from each pair to the control group and the other to the experimental group. 4. a placebo to control group. Carry out a measure of the dependent variable. 2. Carry out a measure of the dependent variable with both groups and compare/measure them in order to determine the effect and its size on the dependent . 3.Borg and Gall (1979:547) as quoted in Cohen et al. Assign participant s to matched pairs. (2007:279) set out a useful series of steps in the planning and conduct of an experiment: 1. Administer the experimental treatment/intervention to the experimental group and.

Or the effect on motivation for learning mathematics (see p. It also has to take account of the interaction of the independent variables.The factorial design Factorial design is useful for examining interaction effects. . For example: On factor (independent variable) may be ‘sex’ and the other ‘age’.281).

.The parametric design Parametric designs are useful if an independent variable is considered to have different levels or a range of values which may have a bearing on the outcome (confirmatory research) or if the researcher wishes to discover whether different levels of an independent variable have an effect on the outcome (exploratory research).

So. a member of the experimental group may receive more than one ‘intervention’. for example. (See page 281). which may or may not include a control condition. .Repeated measures designs Here participants in the experimental groups are tested under two or more experimental conditions.

A pre-experimental design: the posttests only non-equivalent groups design 4. A pre-experimental design: the one group post-tests only design 3. A quasi-experimental design: the pretest-post-test non-equivalent group design . A pre-experimental design: the one group pretest-post-test 2.A quasi-experimental design 1.

and it is given more than one pretest and more than one post-test. .The one-group time series The one group is the experimental group. The time series uses repeated tests or observing both before and after the treatment.

This design has the following characteristics:  They involve the continuous assessment of some aspects of human behavior over a period of time. medicine.Single-case research Single-case research is an experimental methodology which has been extended to such diverse fields as clinical phychology. education. psychiatry and counseling. . requiring on the part of the researcher the administration of measures on multiple occasions within separate phrases of a study. social work.

Researchers must select appropriate levels at which to test the independent variables.Procedures in conducting experimental research An experimental investigation must follow a set of logical procedures. those are: 1. Researchers must identify and define the research problem. 3. 4. . Researchers must decide which kind of experiment they will adapt. 2. Researchers must formulate hypotheses that they wish to test.

During the experiment itself. researchers must take account of the population to which they wish to generalize their result. researchers must select instruments. 7. researchers must endeavor to follow tested and agreed-on procedures to the letter. . In planning the design of the experiment.5. 6. With problems of validity in mind. choose tests. 8. researchers must pilot test the experimental procedures to identify possible snags in connection with any aspect of the investigation. and decide upon appropriate methods of analysis. Before embarking upon the actual experiment.

Select the relevant variables. Administer the pretest.. Select the appropriate experimental design. Control the experimental conditions and environment.g. Identify the purpose of the experiment.A ten-step model to conduct the experiment research can be suggested: 1. . low. High intervention). 2. 5. Specify the level(s) of the intervention (e. 6. 3. medium. 4.

10. Conduct the post-test. 8.Analyze the results. Conduct the intervention. Assign the participants to the group(s). 9.7. .

.Evidence-based research Evidence-based research is a method of investigation. 2007:289290). bringing together different studies to provide evidence to inform policy-making and planning (Cohen et al..

. This is often done statistically. 2007:291)..Meta-analysis Meta-analysis is the analysis of other analysis. It involves aggregating and combining the results of comparable studies into a coherent account to discover main effects. though qualitative analysis is also advocated (Cohen et al.

The advantages of using metaanalysis. Fitz-Gibbon (1985) cites the following: o Humble. small-scale reports which have simply been gathering dust may now become useful. a whole new genre of . o Small-scale research conducted by individual students and lectures will be valuable since meta-analysis provides a way of coordinating results drawn from many studies without having to coordinate the studies themselves. o For historians.

o Coding the study characteristics (e.g. . publication status.g. o Correlating effect sizes with context variables (e.g. published. controlling for ‘lumphy data’ (non-independent data from a large data set). o Measuring the effect sizes (e.g. quality of design. unpublished review) from which effect sizes can be computed. status of researcher). date. design characteristics. locating the experimental group as a z-score in the control group distribution) so that outcomes can be measured on a common scale.There are four steps in conducting a meta-analysis: o Finding studies (e.

(1981):3544) as cited in Cohen et al. the 725 comparisons were integrated into a single curve showing the relationship between class size and achievement in general. 2007:295) identified 77 empirical studies of the relationship between class size and pupil learning. This curve revealed a definite inverse .000 pupils of all ages and aptitudes studying all manner of school subjects.An example of meta-analysis in educational research Glass & Smith (1978) and Glass et al. These studies yielded 725 comparisons of the achievements of smaller and larger classes. Using regression analysis. the comparisons resting on data accumulated from nearly 900.

Routledge. (2007). Education. USA: . Research in sixth edition.References: Cohen et al.

Thank You .

You're Reading a Free Preview

Download
scribd
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->