METHODS EXPERIMENTS

Methods of experimental research are: methods of research used to search for a specific treatment pengaru against another in an uncontrolled condition. For example in the field of physics studies can use experimental design because the variables can be selected and other variables can affect the process of experimentation and can be precisely controlled, while that for Example in the field of physics to find the effect of heat on the long expansion of an object. In this case the variation of heat and the long expansion can be measured accurately, and conducted laboratory research, so that the effects of other variables from the outside can be in control. Whereas in social studies education in particular, the experimental design used for the study will be difficult to obtain accurate results, because many external variables that affect and difficult to control as for example to find the influence of contextual methods to speed the understanding of students in math.

Some form of experimental design
A. Pre-Experimental Designs (Nondesigns)
Pre-Experimental Designs (nondesigns) is not really an experiment because there are external variables that also affected the formation of the dependent variable. Pre-Experimental Designs forms (nondesigns) there are several kinds :

1. One-Shot Cose Study

Paradigm in experimental studies in this model can be described as follows there is a group given trikmen or treatment and subsequent observation of the results.

2. One- Group Pretest-Posttest Design

When the one-shot case study are not given a pretest, the pretest contained on this paradigm before been treated so that the treatment can be determined more accurately, because can compare with the state before being treated.

3. Intact-Group Comparison

There is a group that is used for research but divided by 2 is half the experimental group and half to the control group

B. Tru-Experimental design
Presented two forms, namely :

1. Pottest-Only Control Design

In this design there are two groups, each of which is selected by randum (R). The first group was given the treatment (X) and the other is not. Groups in the given treatment is called the experimental group and the group that was given (treatment) is (O1: O2). In real research analyzed the influence of treatment with different test, wear t-test statistics for example. If there are significant differences between the experimental and control groups, the treatment given them significant influence.

2. Pretest-posttest control group design

There are two groups selected by randum, then given a pretest to determine the initial state is the difference between the experimental and control groups. Pretest results are better when the experimental groups did not differ significantly. Treatment effect is (O2 - O1) - (O4 - O3).

3. Factorial design
A modification of true experimental design, namely by showing the possibility of moderator variables that influence the treatment (independent variable) to yield (dependent variable).

4. Quasi Experimental Design
Is the development of true experimental design, which is difficult to implement. Have a control group, but can not fully function to control the external variables that affect the implementation of the experiment.
Two forms of this experiment are:
a. Time series design
This design can not be selected by randum. Before the treatment group was given a pretest to four times, in order to find out keistabilan and clarity in the given group before treatment. When the pretest for four times the value the different turns, means the group is labile, and consistent.
b. Nonequivalent control group design
The design is similar to the pretest-posttest control group design, this design only in the experimental group or faction of the controls are not selected at random.

Thanks