Research

The process of systematically gathering information for the purpose of answering one or more questions.

This is the key. You can ask around your school and get a number of different opinions about whether a new reading method works. One or two complaining parents might make you think that your teaching is no good. But you will get a better sense of what is going on if you systematically gather information about it. In the case of the reading program, collect assessment scores from all the students and formally interview representatives from everyone affected by the reading method—teachers, parents, children. In the case of your own teaching, do a survey of all parents to find out what they like and what they don’t like.

Participants/subjects: the people or animals studied in a research project.

Types of Research

Descriptive Correlational Experimental Microgenetic Longitudinal Action/Teacher as researcher

There are several different ways to carry out a research project. Each research structure has different advantages and disadvantages and each one focuses on developing a different type of knowledge.

More on research

A research report is more or less like a recipe.

Recipe analogyThe structure of our cooking depends on our goals. If we are hungry and just want to fill our stomachs, we might microwave a frozen dinner (or, a bag of popcorn). If we are trying to impress a potential significant other, we might cook a four course gourmet dinner.

The structure of research also differs based on the goals of the researcher. Just as it is difficult to impress a would-be girl/boyfriend with a nuked frozen dinner, it is impossible to answer some research questions with certain structures.

Recipe analogy continued

The goal of a recipe writer is to write something that you can replicate successfully in your own kitchen

Researchers want you to know exactly how they did their work in case you want to try out the same research procedure to see if you get the same results.

Recipes

When you want to cook something, you might look up several recipes and compare the ingredients. You might pick the one that is best for you based on what ingredients you have in your pantry and what ingredients you actually like.

When you read research, you need to read it as critically as you read a recipe. Do the procedures make sense? How does this situation compare with the situations you are familiar with? Are the authors making reasonable claims?

Still more on research

No one piece of research defines a whole truth. For example, for every reading method ever devised under the sun, there is at least one piece of research that says it is effective. Because of how research is carried out, how questions get asked, how people understand any given idea at a given time, results are not necessarily consistent. When several well-constructed pieces of research yield the same set of results, then you can basically trust those findings.

Types of research: Descriptive research Uses interviews, observations, and surveys to describe

opinions, attitudes, or events. May be ethnographic. May also use participant observation. May be a case study of a single person, such as a single child

working on some aspect of reading.

Ethnography: a descriptive approach to research that focuses on life within a group and tries to understand the meaning of these events to the people involved.Participant observation: a method for conducting descriptive research in which the researcher becomes a participant in the situation in order to better understand life in that group.

Education has borrowed research methods from anthropology. Anthropologists often live in the tribes or communities they are studying in order to understand the people.

Case study: intensive study of one person or one situation.

Descriptive Research Examples

•What do parents learn from our report card format?

•How do students respond to new teaching strategy?

•Which programs in our school do parents value the most?

•Interview four parents from each class (systematic data collection).

•Observe classroom several times as strategy is implemented. Use video tape to record, make notes during observation.

•Create a survey that asks parents about each school program on a scale of 1-5. Then ask parents to rank programs according to importance.

•Qualitative data analysis: categorize each statement a parent makes. Count the number of similar statements made about each aspect of the report card.

•Qualitative data analysis: categorize each aspect of the observation. Count similar events/statements and report them.

•Tally up responses and rankings.

Evaluating Descriptive Research Validity: if several research reports have

similar findings, then the information is probably valid (true).

Applicability: if the research is carried out on college students, how applicable is that to a third grade classroom?

Predictability: descriptive studies do not predict the future. Just because something was found in one setting does not necessarily mean that it will be found in another setting.

Correlational research

A correlation is a relationship, either positive or negative, between two or more variables.

Correlational research is the process of looking for relationships between two or more variables.

Go slowly through the next few slides—this concept has some challenging moments.

Correlational research

Why do colleges use admissions tests such as ACT or SAT (or, in the case of graduate school, the GRE)? The answer is because of correlational research. These tests have been correlated with college grades. The theory is, if you do well on the admissions tests, correlational research predicts that you will do well in college. (In actuality, the correlation between SAT and college grades is only .42, which means it really isn’t a good predictor of college grades).

A perfect correlation would be 1, where a high SAT is always followed by high grades.

Correlational researchThe relationship between any two numerical

variables can be explored.

Variable 1 (Pick one)•Number of questions asked in class•Height of students•Time spent studying•Time spent on athletic pursuits•Amount of calories consumed daily•Time spent playing video games•Length of second toe of right foot•Time spent in music classroom•Number of words written in daily notes•Body mass index of students

AND

Variable 2•Student achievement as measured on final exam

Think about each possible pairing and what kind of correlation might exist.

Correlational researchSome correlations will be positive, such as the relationship between amount of questions asked and student achievement, time spent studying and student achievement, and number of words written in notes and student achievement (this is not an exclusive list; it’s likely that music students and athletes would also do well in achievement).

Achievem

ent

Number of hours spent studying

Student who spent a lot of time studying & did well on test

Student who spent a lot of time studying and didn’t do so well.

Student who didn’t study but did well anyway

Line that shows the overall relationship between the two variables

This type of graph is called a “scattergram.”

Positive correlation: a relationship between two variables in which the two increase or decrease together.

Correlational research

Achievem

ent

Number of hours spent playing video games

Some relationships are negative, such as the relationship between time spent playing video games and achievement.

High achieving student who didn’t spend a lot of time playing video games

Student who played a lot of video games and still did well on the test

Student who didn’t play video games but still didn’t do well on the test.

Line that shows the overall relationship between the two variables

Student who played a lot of video games and did poorly on the test.

Negative correlation: a relationship between two variables in which a high value on one is associated with a low value on the other.

Correlational research

Achievem

ent

Length of second toe on right foot

Sometimes there is absolutely no relationship whatsoever.

High achievement, short toe

Low achievement, short toe

High achievement, long toe

Low achievement, long toe

There’s no red line because there is no mathematical relationship.

Correlational research

A correlation DOES NOT MEAN CAUSE. A high SAT does not CAUSE good grades in

college.

BREAD IS DANGEROUS !Research on bread indicates that:

1. More than 98 percent of convicted felons are bread users [there is a high positive correlation between using bread and committing a crime].

2. Fully HALF of all children who grow up in bread-consuming households score below average on standardized tests.[…and the other HALF score above average on standardized tests]

3. In the 18th century, when virtually all bread was baked in the home, the average life expectancy was less than 50 years; infant mortality rates were unacceptably high; many women died in childbirth; and diseases such as typhoid, yellow fever, and influenza ravaged whole nations. [there was a high positive correlation between cooking bread at home and serious disease during the 18th century]

4. More than 90 percent of violent crimes are committed within 24 hours of eating bread.

5. Bread is made from a substance called "dough." It has been proven that as little as one pound of dough can be used to suffocate a mouse. The average American eats more bread than that in one month!

6. Primitive tribal societies that have no bread exhibit a low incidence of cancer, Alzheimer's, Parkinson's disease, and osteoporosis.

The danger of assigning causality to a correlational study…

More about bread!!

7. Bread has been proven to be addictive. Subjects deprived of bread and given only water to eat begged for bread after as little as two days.

8. Bread is often a "gateway" food item, leading the user to "harder“ items such as butter, jelly, peanut butter, and even cold cuts.

9. Bread has been proven to absorb water. Since the human body is more than 90 percent water, it follows that eating bread could lead to your body being taken over by this absorptive food product, turning you into a soggy, gooey bread-pudding person.

10. Newborn babies can choke on bread.11. Bread is baked at temperatures as high as 400 degrees

Fahrenheit! That kind of heat can kill an adult in less than one minute.

12. Most American bread eaters are utterly unable to distinguish between significant scientific fact and meaningless statistical babbling.

Conclusion about bread

In light of these frightening statistics, it has been proposed that the following bread restrictions be made:

1. No sale of bread to minors.2. A nationwide "Just Say No To Toast" campaign, complete

celebrity TV spots and bumper stickers.3. A 300 percent federal tax on all bread to pay for all the societal

ills we might associate with bread.4. No animal or human images, nor any primary colors (which may

appeal to children) may be used to promote bread usage.5. The establishment of "Bread-free" zones around schools

Moral of the story: Don’t assign CAUSATION to a correlational study!!!!!!! Also beware of how statistics get interpreted in the media.

Experimental Research

Experimental research systematically manipulates variables in attempts to determine cause and effect.

Scientific methodSuppose you want to find out what bacteria grows where? You take several identical petrie plates (independent variable) with identical broth (independent variable) and you expose them in different places (dependent variable). You are careful to avoid exposing one plate (control). You place all the plates in the same incubator (independent variable) for the same amount of time (independent variable) and at the same amount of heat (independent variable). Because there is only one thing that is different (where the plates were exposed) you know that there is a CAUSAL relationship about what you get on the plates and where they were exposed. If the control plate remains clean during the incubation process then you know that any growth on the other plates was due to their exposure and not due to a mistake on your part (accidentally exposing the plates during incubation instead of deliberately exposing them during the experiment).

Scientific method

This is all part of the process of developing a hypothesis and then testing it systematically.

The idea behind the manipulation of variables is that you can only be sure of your results if you have controlled all variables (independent) except for one (dependent variable).

Independent variables: the ones you controlDependent variable: the results depend on the dependent variable… the independent variables CAUSE the status of the dependent variable.

Experimental research in education Frequently, students are assigned to two different groups: a

control group and an experimental group. All conditions of the groups are supposed to be identical, except that the experimental group gets the experimental treatment (such as a new type of instruction).

Students are randomly assigned to the groups. Random assignment means that an individual has an equal likelihood of being assigned to either group.

The idea is that if the groups are basically identical and the treatments they get only differ in terms of one factor (the experimental treatment), then any differences that result are CAUSED by the treatment. (Remember, correlational research does not reveal cause. Experimental research is designed to give cause).

Random: without any definite pattern; following no rule.

An exampleQuestion: does the use of manipulatives increase student achievement in mathematics?

Correlational research: compare time spent using manipulatives with achievement on a chapter test. Results give you an association, but NOT a cause. You still don’t know if the manipulatives CAUSED the results.

Experimental research: choose a topic. Divide children randomly into two groups (random assignment ensures both groups are identical). Both groups will study the same topic (independent variable). Both will have the same test (independent variable). Instruction (independent variable) will be identical in both groups but the experimental group will have manipulatives (dependent variable) to work with while the control group will not. Any differences in scores will be known to be CAUSED by the experimental condition (use of manipulatives).

Evaluating experimental research Comparability of experimental and control groups: were subjects

assigned randomly? Most research studies discuss the makeup of both groups and show how they are comparable.

Maximum control of extraneous variables: for instance, if the same teacher taught both groups (control and experimental), then that is better control than having two different teachers, which introduces the possibility that one teacher is simply better than the other.

Sample size: remember in both educational research and medical research that bigger is better. A study of several thousand people in relation to the effectiveness of a drug is much better than a study of 50 people. In a small study, one unusual person can alter the statistics. In a larger study, a small group of people with unusual reactions will be less likely to change the overall results.

Clearly described manipulation of the variables. What exactly was done? How did the scientists ensure that conditions were exactly alike except for the dependent variable?

Evaluating Experimental Research Statistical significance: if you read them

carefully, the statistics in a research report should tell you how much of a chance that the results could have happened by chance. That would be something like this: p<.05, which means less than 5 in 100 chance that the results occurred randomly. The smaller p is, the less likely results occurred by chance.

Statistically significant: not likely to be a chance occurrence.

Scientific method and human beings

Of course, people are much more complicated than the bacteria described above. Further, no one cares if you manipulate the lives of microscopic beings but it is unethical to manipulate the lives of people beyond a certain point.

For instance, we might find out a lot about raising children if we set up certain types of parenting and randomly placed children in these families, but this would be completely unethical. (The Nazis did unethical scientific research on people).

Therefore, while experimental research potentially yields the most dependable results, it is impossible to use it to explore many aspects of human psychology.

Scientific method and human beings, continued Further, people are enormously complicated and the manipulation

of variables necessarily has to be simplistic—perhaps oversimplified. For example, the naturalistic setting of a classroom is enormously complex. Suppose two groups (an experimental and a control) are taught in the same physical classroom by the same teacher, but during one session, something exciting begins to happen outside on the playground. How comparable are the groups in that instance?

To get around that sort of problem, scientists may do experiments in a more highly controlled laboratory setting. Yet how comparable are people’s reactions in the artificial setting of the laboratory compared to the complexities of real life?

You need to keep all of this in mind when you evaluate the usefulness of research for your teaching practice.

Three other research models

Single-subject Microgenetic Longitudinal

Single-subject

This design tests the effectiveness of something on a single person or group. For example, if you want to know how praise affects a student’s performance, do a baseline with no praise (count how many times the student does the behavior with no response from you), a time of praising a specific behavior, another baseline of no praise, and then another time of praise.

Single-subject experimental studies: Systematic interventions to study effects with one person, often by applying and then withdrawing a treatment.

Single subject critique

I actually watched this specific research project being carried out in a classroom. The kids hated it because they perceived the teacher as being “mean” sometimes and “nice” sometimes (depending on whether they were on baseline or in one of the experimental phases of the research). They would whisper to each other during class changes about whether the teacher was nice today or not. If you are going to do this sort of research project with student, think through the effect it could have on them.

Microgenetic studies

This research involves the close examination of a change process that happens in an individual or a small group of individuals. It involves setting a task to learn and intensively observing the learning process, including videotaping, interviewing, and transcribing (writing down) everything the subjects say.

Microgenetic studies: detailed observation and analysis of changes in a cognitive process as the process unfolds over a several day or week period of time.

Longitudinal research

This is research that takes place over time, such as when researchers assess pre-schoolers and then follow their progress through high school.

This is expensive and difficult research to do but it yields valuable results.

Most often people substitute “cross-sectional” research: studying pre-schoolers, grade schoolers, high schoolers all at once in attempting to get at the same type of information that might be yielded by a longitudinal study of a single population.

Action research

Action research is a form of applied research designed to answer a specific school- or classroom-related question.

Action research is often undertaken by teachers in order to improve their own teaching practice. Many journals publish action research done by teachers (something nice to add to a resume).

Conducting Action Research

Identify and diagnose a problem that is important to you

Systematically plan and conduct a research study

Implement the findings to solve or improve a local problem

Use the results of the study to generate additional research

Action research may be descriptive, correlational, or experimental in design.

Examples of action research

The effect of a new classroom management plan on student behavior

How a new teaching strategy affects student achievement

What students like best about their classroom and what they wish were different

How a new program affects attendance and achievement of at-risk students

The effect of a new procedure on the amount of instructional time available in the classroom

The effect of a behavioral contract on the outbursts of a particular student

Can you imagine how you might carry out one or more of these research projects?

Research and the Development of Theory A principle summarizes results consistently

supported by large number of research studies A theory is a set of related principles derived from

observations that, in turn, are used to explain additional observations.

Theories allow us to explain and predict

Principle: established relationship between factors.Theory: integrated statement of principles that attempts to explain a phenomenon and make predictions.

An astronomical example

Using a telescope, Galileo observed the movements of stars and planets.With repeated observations he discovered some basic principles of their movements—this star moves this way, that planet moves that way. Using these principles, he developed a theory: the earth moves around the sun (and not the other way around, as was generally believed at the time). His theory not only explained the movements of the stars that he observed, but it also predicted (successfully) their future movements.

A psychological example

Principles (observed multiple times)•People don’t immediately remember everything they learn•People can be in an environment yet apparently not remember significant aspects of that environment•When people practice something they tend to remember it a long time later•People seem to become overwhelmed and unable to remember when a lot of information is presented in a short period of time.

Theory:Information processing model. It explains all these principles and it predicts future behaviors in relation to learning information.

Research and Teacher Decision Making Critical decision making: The role of classroom

context. Teachers need to use professional judgment and critical thinking in order to determine how applicable a research study is to their own classrooms and teaching practices.

Practical decision making: how feasible is it to apply a set of research results? Is it going to work in a classroom of 25 students? Is it going to be do-able?

Artistic decision making: You can be creative about how you make teaching decisions and how you apply research. Artistic decision making is one of the most fun aspects of teaching.