Chapter+One+-+Introduction+to+Research

Educational Reseach is a formal and systematic application of scientific methods to the study of educational problems which is a process and follows numerous steps. The goal is to explain, predict, and control educational issues. Observation is a useful tool in educational research. Observers can record various behaviors. Research includes the selection and definition of problems, procedures that reflect all activities involved in collecting data, application that involves one or more statistical techniques, and the results are based on data analysis. Performing educational research will illuminate how decisions are made about topics. Understanding good research techniques helps in dealing with and making everyday decisions. When investigating, collect data to support your research. When improvements are shown, you will get feedback from others in your field who may also benefit from your analysis and results.

=The Purpose of Research=

I. Introduction to Research
Research leads to new ways of thinking and enhances the ability to understand learning. The more research that is conducted the easier it is to answer questions and address problems. We know about the world around us through research, by scientifically studying problems and selecting an appropriate format. Research is the most reliable method for educators to receive information which will help to better assist students, meet their needs, and yet challenges them to apply their critical thinking skills. By investigating students' cultural, physical, and emotional needs, teachers are able to formulate effective teaching styles. There is such an importance to continue researching, especially that on education and adolescences. Without research there would be no connection between educators. People would continue to make the same mistakes due to not being aware of what others are doing or have already done.

lI. Approaches to Research
__Quantitative vs. Qualitative:__ Quantitative: focuses on specific hypotheses which can be analyzed using statistical measures. The data is represented in charts and tables. The goal is to find relations or examine differences on specific variables. It uses a large sample size in order to make generalizations. The researcher tends to be very objective during the study, attempting to remove the possibility for biases.

Qualitative: focuses on observations of behaviors and relationships allowing a hypothesis to emerge during the study. The research is represented in narrative form. The findings are only to describe the specific population that was being observed. The researcher tends to be more subjective since the results are from personal interpretations.

It is important to mention that all research is not just qualitative or quantitative in nature. An increasing number of researchers are using both methods to ensure that their research is valid and is less likely to be criticized for a lack of information (Suter, 2006).

In response to the class notes: Qualitative research can support quantitative research as well as the other way around. There are pros and cons of both, but as mentioned in the above comment many people are using both, because it adds more support to the research. It is important to explain the conditions surrounding the statistics and data gathered. There are two quotes that come to mind every time the discussion of these two approaches. the first one is "figures don't lie, but lairs figure," which means that statistics and data can be constructed in a way that is favorable to someones agenda or bias. The other quote that is very important is, "correlation does not mean causation." This quotes means if you find a relationship between statistics, condition, and a certain behavior does not mean you have found what causes what. Keeping all the things in mind both approaches can be useful as well as the combination of the two.

__Descriptive vs. Inferential:__ A study is //descriptive// when the results of that study are not going to be used beyond the original group being tested. When results are going to be used beyond the original group being tested, this is generalization. //Inferential// studies make these larger generalizations (Suter, 2006). Using an example one of our classmated brought up during the synchronous chat, if teachers at a particular school (Hope Academy Cathedral) were given a survey about staff morale for the sole purpose of finding out staff opinions in that school, this would be descriptive. But, on the other hand, if the sample was generalized to the larger population to indicate how morale is at a particular type of school (i.e. charter), then this would be inferential.

To generalize a sample to the larger population (inferential), the sample needs to be representative of the larger population. Inferential studies will talk about whether the results are "statistically significant," or if results are different enough to matter or indicate a real difference. The book uses the example of pollsters. Pollsters take a small, representative sample of the population and generalize their results to the entire population. They also figure out if there is a statistically significant difference in the number of votes for one candidate v. another (Suter, 2006).

__True experimental vs. Quasi-experimental__

What is True experimental? True experimental looks at cause and effect relationships. Two factors make up this type of research: mainpulation and random assignment. Groups of students are created and the type of intervention they receive is manipulated. They are then randomly assigned a group. An example of this type of research would be discovering if students who memeorize mulitplication tables have quicker recall of facts then those who use manipulatives/hands-on activities to learn the facts. The multiplication facts both groups learned would be the same, differing only in the method taught. Students in the grade would then be randomly assigned one of the groups. This type of research is beneficial because the groups are more comparable when looking at variables. This type of research is prefered by the No Child Left Behind Act because it leads to identifying educational practices supported by rigorous scientific evidence (Suter, 2006).

What is Quasi-experimental? If “quasi” means //as if// or //almost//, then a quasi-experiment means an //almost// true experiment (O’Sullivan, 2003). An example of a quasi-experiment would be designing a study using a comparison group that is //similar// but not technically an exact control group. A study that examines the effects of school violence and compares two rural schools with similar characteristics when one school has implemented an anti-bullying policy and one has not implemented an anti-bullying policy would be a quasi-experiment. Additionally, studies that include a series of time are quasi-experimental. Any kind of longitudinal study (i.e., one that occurs over a series of time) would be quasi-experimental. An additional characteristic of a quasi-experimental study to be considered is the concreteness of the topic of the study. For example, a study focusing on student morale following the implementation of anti-bullying policies contains abstract opinions, and is therefore, quasi-experimental. O’Sullivan, E., Rassel, G. & Berner, M. (2003). //Research methods for public educators//. New York : Pearson Education Inc. __Causal Comparative vs. Correlational__

Casual-Comparative research attempts to ascertain a cause and effect relationship between two or more groups. The groups must be able to be compared in some way. The data for the research is collected using mostly surveys, tests, questionnaires, but interviews and observations may also be utilized. Any data accumulated must be able to be analyzed as numbers.

Correlational research explores the relations between two or more variables. This type of research is used to establish the character of a relationship. It is valuable in that it offers the capability to observe relationships between a large number of variables. It is important to note that correlational research is focused on the similarities in variables as opposed to casual-comparative which is about finding differences. As with casual-comparative research, data may be collected using surveys, tests, questionnaires, interviews, and observations.

__Teacher vs. Traditional__

Teacher research refers to teachers who study their own practice; it involves self-reflection using field notes, work samples, or journals rather than standardized test scores. Teachers may used shared teacher stories and metaphors as a way of understanding student behavior and performance in a classroom. Teacher research is a method to improve learning in classrooms and this type of research is classified as //action research// (Suter, 2006).

Traditional research refers to formal scientific research. It is considered "colder" than teacher research because data gathering is detached and standardized to lessen the impact of research bias. A criticism of traditional research is that results are often "hidden" in journals; the academic authors of these articles benefit, but classroom teachers are frustrated by the academic language (Suter, 2006).

III. Selecting a Research Topic
When stating a research question one important consideration is whether or not your topic is researchable. It will be difficult to conduct research if your question does not have measurable outcomes or is not, in essence, an "answerable question." Ask yourself, "Is my question S.A.M.?" Simple. Attainable. Measurable. These three factors are essential to stating a research question.
 * A. Stating the Research Question**

Another important consideration in forming your question is to avoid using value judgements. The aim of a research question should not be to create or respond to a debate. Research questions should aspire to produce objective findings.

Selecting a topic can be very challenging but if you rely on your personal experience than it may be easier than you think. It is important to concentrate on a particular topic that affects you personally. Try to choose a topic that could improve your classroom enviroment. For example, a topic that relates to specific teaching methods or training programs. You may also entertain a topic that relates to discipline techniques and the outcomes for specific students. Another example would be researching a topic that relates to your students specific culture, religion or even race. The text provides great examples of how to phrase a research question. For example, rather than asking "Should all children wear unforms in public school?", the research question should be posed as "Do children in schools requiring uniforms achieve higher that children in schools not requiring uniforms?". (Suter, 2006)

B. When to state hypotheses
A hypothesis derives from the Greek, hyotithenai which means “to put under” or “to suppose.” It is a question that consists of a suggested explanation for a phenomenon. However, the topic statement must be researchable. Hypothesis statements should contain two or more variables that can be measured. Researchers must not know the outcome when stating the hypothesis; they are making predictions, instead. Scientists must do at least one experiment to test the hypothesis. They conduct experiments to test their prediction until they prove that their hypothesis is true or false. The hypothesis must be stated after stating the problem, looking for possible solutions and observing related incidents.

C. Ethical Considerations
Every researcher needs to take careful consideration to make sure their research is ethical. The American Psychological Association (2002) has created 5 clusters which ethics center around. These 5 clusters include: 1. Beneficence and nonmaleficence 2. Fidelity and responsibility 3. Integrity 4. Justice 5. Respect for people's rights and dignity.

When conducting research, the researcher should always ensure there is written consent signed by the participant, or in the classroom setting, by a parent or legal guardian. The researcher also needs to inform the participant/legal guardian of the reason for the research. The findings of the research needs to be made available to all parties involved. Another aspect of ethics that should be addressed is the ability for the participants to withdraw from the research at any time for any reason. Researchers don't want their participants to feel uncomfortable or pressured into any situation.

The text outlines six (6) important ethical guidelines that educational researchers must follow. They are:

1. Clarify that participation in a research study is voluntary; no coercion of any kind should be tolerated; participants may freely withdraw at any time. 2. Obtain informed consent from participants; all aspects which might affect the decision to participate must be explained, including all risks. 3. Avoid deception, cause no harm and avoid invasion of privacy. 4. Maintain confidentiality and integrity of data; maintain anonymity of participants. 5. Debrief participants and provide information or explanation about the research. 6. Benefits shoudl outweigh risks after careful considersation.

(Suter, 2006).

These aspects of ethics, as well as others, need to be addressed before any research project begins.

IV. Research Validity
Validity is "the meaningfulness of scores, which is often assessed by the accuracy of inferences made on the basis of test scores. It also refers to the extent to which a test measures what it is supposed to measure." It is helpful to remember that validity refers to the use of the instrument for the interpretations made and not to the instument itself. This is why it is important to have a valid instrument so that you can make a solid inference. (Beebe) There are four things that should be considered regarding the inferences. 1. Is it appropriate? "Does it relate to the study?" 2. Is the information accurate? "Does it allow for correct interpretation?" 3. Is it consequential? "Is there some meaning derived from the data to make conclusions?" 4. Is it useful? "Does it relate to the outcome that you are trying to understand?"

Validity differs from reliability in that reliability simply explains how well an instrument is measuring, while validity measures the gathered informations usefulness and accuracy as it pertains to the research at hand. Validity can be broken down into three subsets: content validity, predicative validity, and construct validity. Content validity is defined as "the extent to which a test reflects the domain of content that it presumably samples." Contant Validity basically addresses the fact "does it look like it measures what it says it will measure." For example, if it looks like a duck, moves like a duck, and has feathers like a duck, it must be a duck. (Beebe) Although there is no statistical evidence to support Contant Validity, evidence is supported by the information the researcher provides. (Beebe) Predictive validity is "the extent to which test scores accurately predict an outcome (a criterion)." Construct validity is "the extent to which test scores accurately reflect the trait or ability that the test is presumed to measure." (Suter 2006) Validity is of the utmost importance to researchers. Without validity research is meaningless and not useful. Some factors that threaten validity are bias, inconstancy, and clarity, to name a few. (Beebe)

V. Research Language
__Construct__: The construct is the foundation of any research study. It is a working hypothesis or concept. A construct is an abstract idea or complex concept that represents a set of simpler related ideas. Constructs can’t be measured directly but can be represented by variables. Each complex concept (construct) has simple related ideas (variables)

__Extraneous Variable__: Extraneous variables are undesirable variables that affect the outcome of the research study. For this reason, extraneous variables can be called nuisance variables. Extraneous variables influence the dependent variable, causing mixed up results when testing for the independent variable. In order to reduce the amount of error in an experiment, an effort must be made to control the extraneous variables that are influencing the results. There may be hundreds or thousands of extraneous variables in one research study. Examples of extraneous variables that may affect an educational research study are the children’s prior knowledge, learning styles, motivation level, and their amount of alertness. Also, the teaching style or delivery style, the time of day, or the noise level in the school or classroom may also examples of extraneous variables.

__Variables__: A variable is an attribute or property belonging to a physical or abstract object, system, or process. It may change its value during the time that it is being observed or measured. Variables are measurable. Variables are either introduced to the study by the person doing the analysis and are used to answer questions about the hypothesis (for example - whether a group gets an certain type of drug treatment or a placebo) or then are random, outside factors which are more difficult to measure (for example - whether the individuals being evaluated got enough sleep the night before, ate a healthy breakfast, are experiencing personal problems, etc).

__Constants__: A constant is a characteristic that does not change. Researchers use constants as a way of eliminating unwanted influences that may affect an experiment. (Suter, 2006). It is used in an experiment as a control. Variables and constants are used together. The variable in usually the element in the experiment that the researcher manipulates. In this way a researcher can look at a variable and its relationship to the constant as a means of determining if that variable is creating a change of some sort. An example of a constant would be a survey given to all students of the same age or same sex. In the process of developing my own research around behavioral accoutability, it will be necessary to put in place a constant in order to elminate possible variances relating to the develpmental differences of children of different age groups.

Hypotheses: There are three different classes of hypotheses that educational researchers are concerned with following. 1. Research hypotheses, 2. alternative hypotheses and 3. Null hypotheses. (Suter, 2006)

__Research Hypothesis__:

A research hypothesis is often thought of as the "main hypothesis". This hypothesis is the researcher's guess about what the outcome will be, of a specific research study. This hypothesis can be stated in three different formats, all of which are based on a theory. An If A, then B format pinpoints A as the independent variable and B is referred to as the dependent variable. For example, "If a student is bullied, then he/she will have lower self-esteem". A statement or claim is another format used in stating a research hypothesis. This statement is more direct; "Students who are bullied will have lower self -esteem". Finally the research hypothesis can also appear as a question, for example, "Will young children who are bullied have lower self-esteem than those who are not?" Regardless of the format a research hypothesis must focus on a predicted outcome and the variables that are being studied.

__Alternative Hypothesis:__

An alternative hypothesis is the likelihood that an observed outcome is valid. A good educational researcher will have Alternative Hypothesis that could be thought up by other educational research critics already explained in their research. For example my research hypothesis is to prove that students taught with new math are better students of math. I should research the rival or the alternative hypothesis to answer as many questions that may come up by another researcher. An alternative hypothesis to the example above: One suggestion might be good students are better new math learners. A critical reviewer of research can come up with many reasons why students are better in math besides being taught with new math. Alternative hypothesis are couther-explanations to the research hypothesis. Its good for an educational researcher to know their stuff-product

__Null Hypothesis__: A null hypothesis is the opposite of an alternative hypothesis in that an observed outcome has resulted from chance. It basically states that there is no connection with what is being studied and the predicted outcome. For example, "In a population of 1st graders, there was no difference between physical activity levels and academic performance". A researcher strives to prove that the null hypothesis is not true, and that there is a connection between the variables.