RESEARCH ON PURPOSE
Research (research) is the use of scientific method is formal and systematic to study a problem. In education we know it is "Education Research," the use of scientific method is formal and systematic study of education issues.
The basic difference between education and social research with other scientific research lies in the nature of the event or phenomenon being studied. Compared with other scientific research, education and social research is more complex because the difficulty of explaining, make an estimate, and control the circumstances of human concerns.
Many variables are known, or unknown, effect on education and social environment that is very difficult to generalize. Strict control was done and that can be maintained at a chemical laboratory may not be implemented in the field of education and social issues. For example: an observer can be subjective in making a "record" of behavior, and the person being observed may behave unusually as it is in the field of education and social pengamatan.Dalam also difficult to create conditions such that the desired study investigators. Moreover, the condition of the object of research can not be repeated. Should be repeated, the result will not be the same as the first condition.
Other than it difficult to carry out careful measurements in educational research. Most of the measurements are indirect measurements, there is no comparable instrument with the tools to measure heat or electric current measuring instrument for measuring for example: intelligence, achievement, or attitudes.
Although there are differences in the nature of the phenomenon being studied, the steps in conducting education research / social are the same as the steps in conducting another study, namely: the selection and limitation issues, the implementation procedures of the study (data collection), making data analysis , and the creation and inferences.
In the classification of research can be divided into two parts, namely "Research Pure and Applied Research". It's hard to say that both types of research are separate because they are actually in a union activity. In the original form, pure research conducted solely for the purpose of developing and refining / improving the theory. While applied research as seen from its name, is done for purposes pengetrapan or testing of theory and assessment of their role in solving problems.
Pure research related to the creation of the general concepts of understanding (learning), whereas applied research relating to its use in education. For example, a pure research done on animals to study the problem of reward (reward / punishment) and their effects on comprehension. The study continues with applied research is to test these concepts to determine their effectiveness in improving the understanding and behavior.

A. Research Evaluation (Evaluation Research)
The purpose of the evaluation study is to facilitate decision making with respect to the merits or relative superiority of two or more choices of action.

2. Research and Development (Research and Development)
The main objective of the research and development (Research and Development) is not to formulate or to test theory but to develop effective outcomes to be used in schools or other institutions.
3. Research Action (Action Research)
The purpose of Action Research is to solve local problems (eg class) using the scientific method. This study, just pay attention to local issues; it does not matter whether the results obtained can also be applied to other places or not.

C. RESEARCH BASED METHOD
In general, research and strategy to take similar steps. The measures consist of the manufacture of a problem statement, data collection, data analysis, and inferences.

A. Research History (Historical Research)
Historical research related to the investigation, understanding, and explanation of past events. The purpose of historical research is to reach a conclusion with respect to the causes, effects, or trends of past events that may help explain the events of today and help anticipate events to come. The study of history generally does not collect data with management instruments (eg, testing with a test) to the individual, but look for data that already exists. In the study of history known as the primary and secondary data sources. Primary data source is a first-hand knowledge (eg, eyewitness reports, and original documentation); secondary data source is a second-hand information. Evaluation of the data consists of historical research and criticism in the outside criticism. In this connection, beyond criticism assess / evaluate the authenticity of the data; critical in assessing the usefulness of the data.

2. Descriptive studies (Deseriptive Research)
Descriptive research related to data collection to provide an overview or affirmation of a concept or phenomenon, also answer questions in connection with a subject of current research, such as attitudes or opinions of the individuals, organizations, and so forth. Descriptive data is generally collected through a questionnaire surve, interviews, or observation. Because research in general makes the questions for specific purposes that the instruments should be created for each investigation, according to the hypothesis. A common obstacle in descriptive studies is the lack of response - the subject's reluctance to return the questionnaire or absence of a subject during an interview scheduled.

3. Correlation studies (correlational Research)
Correlation study aims to determine the relationship, and how far there is a relationship between two variables (which can be measured) or more. For example, the relationship between intelligence with creativity, spirit with achievement, height for age, english niali the statistical value, and so forth. The purpose of an investigation of the correlation is to define or express a relationship or use the relationships to make predictions (forecasts)
Despite the fact there is a close relationship between two variables, one can not conclude that one variable is the cause of another variable. This is because there may be a third factor that affects the first variable, or the second variable, or may affect both. By ignoring the presence or absence of a causal relationship, the close relationship would allow us to make forecasts. Level of relationship between two variables is usually expressed by a correlation coefficient between 0.00 to 1.00.

4. Causal-Comparative Research and Experiment.
Causal-comparative research and experimentation are both trying to create a causal relationship, both involving comparison groups, the difference is in the experimental study, the cause is "suspected of" playing (manipulated), the comparative study kausual. In the experimental study, the activity or the nature of trust makes a difference is a cause or a service or action (treatments), and more generally as the independent variables (independent variables). In the experimental study, research plays (manipulate) at least one independent variable and observe its effect on the dependent variable (dependent variable) or more.
Topics for research experiments, such as "reward effect on student achievement. "Here the independent variable, or cause, is a type of reward (reinforcement). Here may be a prize reward, or punishment, or with no reward. Or the dependent variable is the effect of student achievement (Achievement studying two groups of students who earn different rewards). Two groups of students were obtained at random before ekspimen. Then the groups get the same experience, except the type of reward received. After some time, their learning achievement compared.
Random selection procedure group (random) owned by the research experiment, not on other research methods. More than that, because of the direct manipulation and control, research ekspimen can really determine cause-effect relationships. In the causal-comparative research, the independent variable, or the cause, not manipulated; uni thing has happened. In these studies, differences between groups occurred in the independent variable is determined by the researcher. Because less manipulation and control, cause-effect relationships obtained from the causal-comparative research is very weak and transient.
Tofik for causal-comparative research, for example, "the effect on achievement belajarsiswa kindergarten, at the end of class I general tests". Here, the independent variable or cause of kindergarten education (primary school students entered the school through kindergarten first primary school or direct entry). Dependent variable or effect, is a common learning achievement at the end of grade repetition I. These two groups of students (who entered the primary school through kindergarten and groups who are not). Sitting in class I get the same treatments and experiences. later learning achievement at the end of class I general tests of each comparison.