Research is the process of trying to answer a question empirically. Review the main ideas Research is the process of trying to answer a question empirically.

Lines of Research What lines of Research help Research You? Lines What does the Research Say? What do you Want to do?

Reading a Research Report In reading a research report, the first thing we need to know is what the research question was. In most reports the researcher also explains why this question was asked – why the answer to this question is worth knowing.

My Research ??? Suppose I tell you that I conducted research to answer the question, do most children like school? I tell you that the answer I got is yes. Do you now believe that most children like school? Of course not. Whether you believe my answer depend on whether you think I did a good job of conducting my research: Whether you believe that my research process was a valid one for answering this question.

Validity??? In fact, this term validity is a critical one in research. Scientists talk about the validity of research at length and in detail. There are even named kinds of validity, which you will learn about later in this course. Validity refers to the research process in this way: Was the process a valid one for answering the question that the research posed?

My Research To return to the example of whether most children like school What kinds of things will you want to evaluate my research process and, ultimately, my answer? You will want to know almost everything there is to know about what my process was. Suppose I tell you that I asked children. Now do you believe my answer? Not necessarily. How I asked them is important – such things as who did the asking, How the question was phrased, and where the children were at the time.

My Research (cont.) Now do you believe that my answer is valid? Suppose I tell you that I had teachers ask the children in their classes to raise their hands if they like school. Now do you believe that my answer is valid? Probably not.

My Research (cont.) You realize that children may not give honest answers to such question in such a situation; they may slant their answers to what they think the teacher wants to hear or what they think will be acceptable to other children.

Evaluating My Research Process In evaluating my research process, you will want to know who these children were. Were they elementary-school children, high-school children, “normal” children, gifted children, learning-disable children? Were they dominant-culture or culturally diverse children? Were they in public or private schools? And so forth.

What did the researcher do to try to answer the questions? Researchers describe what they did to try to answer the research question. Readers need to know a great deal about the research process in order to decide whether they believe the research results.

What did the researcher do to try to answer the questions? (Cont.) In response to my study to whether children like school, you may decide to believe that these kinds of children, when asked in this kind of situation, say they like school, but not to believe that most children really like school?

What answer did the researcher get? Research questions are often complex. The answer is usually not as simple as yes or no. Also, our research processes are never perfect; limitations in the process have to be discussed in terms of how limit the answer.

What answer did the researcher get? (Cont.) In this section researchers discuss: The conclusions they think are defensible from the data they collected. Reservations they have about the process and the answers it yielded. Under what conditions they think the answers apply. What they think the answers mean: such things as how results fit into the bigger picture,

What answer did the researcher get? (Cont.) What new questions were raised, and what directions future research might take to deal with the problems encountered in this research or to check or expand on the findings of this study

Research Questions Research is often conceptualized as either applied or basic. Applied research has potential for meeting an immediate, perceived need of some groups of practitioners (e.g., teachers, principals, or school counselors). Practicing educators have immediate needs: How can gifted children be identified? How can administration be improved? Practicing educators look to educational research for answers to some of their questions.

Research Questions Basic research has the potential for contributing to basic knowledge in some areas of scientific interest (learning, social interaction, or creativity). Although basic knowledge can and often does ultimately lead to applied finding, this is not its immediate goal.

The Importance of Theory One of the higher goals of science is the understanding of cause and effect relationships. One of the best tools in pursuing such understanding is theory

Comprehensive Causal Explanation. A theory is a fairly comprehensive, causal explanation. Theory differs from lower-level, noncomprehensive explanations. To say that my child learned not to touch the stove by getting burned is a causal explanation; however, it is not a theory because it is not comprehensive.

Comprehensive Causal Explanation. Reinforcement learning theory is a comprehensive causal explanation. It explains the causes of learning in terms of the consequences of behavior:

The Importance of Theory (cont.) These are comprehensive explanations because each applies to many kinds of learning, in many kinds of contexts, by many kinds of organisms.

The Importance of Theory (cont.) The word theory is usually reserved for comprehensive causal explanations that go beyond what is already widely accepted to be true. Theories are thus, in one sense, merely proposed explanations that must be tested in order to be accepted or rejected.

Theory Building In actual practice, important theories are seldom definitely accepted or rejected. Instead, tests of a theory provide knowledge that is used to revise the theory. The revised theory then undergoes further testing and revision.. Scientists often use the term theory building to describe their work, reflecting this ongoing process of testing, revision, and retesting

Pursuing Causal Understanding Theory is a useful tool for pursuing causal understanding because it has these two properties: 1)It is comprehensive, and 2) it has the potential for extending knowledge beyond what is already known.

Empirical Observation One way to define research (some scientists argue that it is the only way) is to say that research is the empirical part of theory building. Theory comes, in part, from research: Scientists pose theories based on empirical observation. And research is used to test theory: Scientists observe empirical events to see whether they confirm to theory.

Theory Building Theory building typically begins with observation and pattern recognition. In observing events, scientists detect patterns in the events: When X happens, Y also tends to happen.. For example, when an organism is hurt, it tends to avoid the source of discomfort, and when an organism is rewarded, it tends to seek the source of reward.

Induce Causal Explanations: From such observation and pattern recognition, scientists induce causal explanations: for example, learning is caused by the consequences of behavior The word induce in the previous sentence is an important one.

Inductive Reasoning The word induce in the previous sentence is an important one. Philosophers use the term induction to describe one kind of reasoning: from the particular to the general. Causal explanation that is based on observation and pattern recognition is inductive.

induction When induction results in a theory, the theory is considered to have been proposed, not proved. Scientists must then test theory in further observation that is independent of the observations that generated the theory. This is because induction is not foolproof. The observed patterns may or may not reflect the operation of the general principle. Some apparent patterns are simply the result of coincidence.

Induction Philosophers use the term induction to describe one kind of reasoning: from the particular to the general. Causal explanation that is based on observation and pattern recognition is inductive reasoning

Induction is reasoning When research results in the generation of new theory, the relationship between the research and theory is inductive; � Inductive, theories resulting from induction have to be tested on independent bodies of data.

Induction Induction is reasoning from the particular to the general :

Deductive, theory-testing research � Deductive, theory-testing research is usually part of an ongoing process of modification and re-testing of existing theories.

Deduction.. When research is done to test existing theory, the relationship between the theories and research is deductive. Deduction is reasoning from the general to the particular. With deduction, we begin with a general principle and deduce particular examples from it.

Deduction.. If I begin with the general principle that learning is caused by the consequences of behavior, then I can deduce that my puppy should learn to come when I call her if I give her a treat.

Deductive Reasoning Using deductive reasoning to test theory involves an if-then kind of reasoning: If this general principle is true, then this particular case should be true. If this learning is caused by the consequences of behavior, then giving my puppy a treat should teach her to come when I call her.

Theory building Theory building thus involves two kinds of research: Inductive research, from which scientist hope to generate theory; and deductive research, from which scientists test theory.

Deductive Tests Typically, deductive tests do not definitively confirm or disconfirm theory; rather they suggest ways to modify theory. The modified theory is then subjected to further testing, further revision, and so forth. Over time, some theories come to be generally accepted. Reinforcement learning theory is an example.

LITERATURE REVIEWS Most research reports begin with two related parts: 1) A description of the question the researcher was trying to answer when he did the research; and an explanation of why this question was chosen — 2) Why it is important and worthwhile to pursue. This explanation involves what is usually called a review of the literature.

LITERATURE REVIEWS (cont.) Students often consider literature reviews to be unimportant. In fact, a good literature review can be the most important part of a research report. It places the research question in the context of scientific thought: What is already known about this issues and what avenues of further investigation look promising to the scientific community.

LITERATURE REVIEWS (cont.) If the research question is an applied one, with no theoretical claims, the researcher will usually review literature that documents a need for the kind of answer the researcher has tried to find. More often, however, researchers do make theoretical claims; they attempt to link the research being reported on with theory building.

LITERATURE REVIEWS (cont.) Occasionally such links seem tenuous, and it is tempting to conclude that the researcher has “invented” a theoretical rationale after the fact, in order to get his research funded or published

Chapter 4 Selecting a Sample Educational Research Chapter 4 Selecting a Sample

Key Ideas Obtaining permissions for data collection Selecting participants for data collection Identifying data options Recording and administering data collection

Procedures for Collecting Quantitative Data Obtain permissions identify the unit of analysis secure permissions obtain informed consent from participants

Procedures for Collecting Quantitative Data Select participants specify a population and sample use probability and non-probability sampling choose a sample size

Obtaining Permission: Obtaining Permissions Institutional or organizational (e.g. school district) Site-specific (e.g. secondary school) Individual participants or parents Campus approval (e.g. university or college)

Sampling

Select Participants: Specify a Population and Sample A population is a group of individuals that comprise the same characteristics A sample is a sub-group of the target population that the researcher plans to study for the purpose about making generalizations about the target population. Samples are only estimates The difference between the sample estimate and the true population is the “sampling error.”

Populations and Samples Target Population Sample Sample Sample Population -All teachers in high schools in one city -College students in all community colleges -Adult educators n all schools of education -All high school biology teachers -Students in one community college -Adult educators in 5 schools of education in the Midwest

Types of Quantitative Sampling Quantitative Sampling Strategies Probability Sampling Non-Probability Sampling Simple Stratified Multi-Stage Random Sampling Cluster Sampling Sampling Convenience Snowball Sampling Sampling

Select Participants: Use Probability and Non-Probability Sampling Probability sampling is the selection of individuals from the population so that they are representative of the population Non-probability sampling is the selection of participants because they are available, convent, or represent some characteristic the investigator wants to study.

Types of Probability Samples Simple Random: selecting a sample from the population so all in the population have an equal chance of being selected Systematic: choosing every “nth” individual or site in the population until the desired sample size is achieved

Random Sampling Methods

Part of a Table of Random Numbers 011723 223456 222167 032762 062281 565451 912334 379156 233989 109238 934128 987678 086401 016265 411148 251287 602345 659080 059397 022334 080675 454555 011563 237873 666278 106590 879809 899030 909876 198905 051965 004571 036900 037700 500098 046660 063045 786326 098000 510379 024358 145678 560132 345678 356789 033460 050521 342021 727009 344870 889567 324588 400567 989657 000037 121191 258700 088909 015460 223350 667899 234345 076567 090076 345121 121348 042397 045645 030032 657112 675897 079326 987650 568799 070070 143188 198789 097451 091126 021557 102322 209312 909036 342045

Nonrandom Sampling Method

Representative vs. Non-representative Samples

Types of Probability Samples Stratified sampling: stratifying the population on a characteristic (e.g. gender) than sampling from each stratum. Multi-Stage Cluster Sampling: a sample chosen in one or two stages because the population is not easily identified or is large

Types of Non-Probability Samples Convenience Sampling: participants are selected because they are willing and available to be studied Snowball Sampling: the researcher asks participants to identify other participants to become members of the sample.

Proportional Stratification Sampling Approach Population (N=9000) Boys N=6000 .66 of pop. 200 Girls N=3000 .33 of pop 100 Sample = 300

6 Sampling

Selecting a Stratified Sample

Population as Opposed to Ecological Generalizing

Ecological Generalizability Results of a study can be extend to other settings or conditions The setting under which a study takes place Research results from urban school environments may not apply to suburban or rural school environments

Select Participants: Choose a Sample Size Select a sample size as large as possible from the individuals available Select a sufficient number of participants for the statistical tests you will use (e.g. 15 per group for experiments) Calculate the sample size using a sample size formula

Collecting Qualitative Data

Key Ideas Gaining site permission Purposive sampling Types of qualitative data Protocols and Issues regarding administering and recording qualitative

Gaining Permission Gain permission from Institutional Review Board Gain permission from “gatekeepers” at the research site Gatekeeper: individual at the site who provides site access, helps researcher locate people and identifies places to study The gatekeeper may require written permission about the project

Information for the gatekeeper Why their site was chosen What time and resources are required What will be accomplished at the site What potential there is for your presence to be disruptive What individuals at the site will gain from the study How will use use and report the results

Qualitative Sampling Unique characteristics of qualitative research In-depth inquiry Immersion in the setting Importance of context Appreciation of participant’s perspectives Description of a single setting The need for alternative sampling strategies

Qualitative Sampling Purposive techniques – relying on the experience and insight of the researcher to select participants Intensity – compare differences of two or more levels of the topics Students with extremely positive and extremely negative attitudes Effective and ineffective teachers

Qualitative Sampling Purposive techniques (continued) Homogeneous – small groups of participants who fit a narrow homogeneous topic Criterion – all participants who meet a defined criteria Snowball – initial participants lead to other participants

Qualitative Sampling Purposive techniques (continued) Random purposive – given a pool of participants, random selection of a small sample Combinations of techniques Inherent concerns related to generalizability and representation

Differences Between Random and Purposeful Sampling Random “Quantitative” Sampling Select Representative individuals To generalize from sample to population To make claims about the population To build/test “theories” that explain the population Purposeful “Qualitative” Sampling Select people/sites who can best help us understand our phenomenon To develop detailed understanding That might be “useful: information That might help people “learn” about the phenomenon That might give voice to “silenced” people

Types of Purposeful Sampling When Does Sampling Occur? Before Data Collection After Data Collection has started What is the intent? To develop many perspectives What is the intent? To describe what is “typical” to those unfamiliar with the case To describe some sub-group in depth To take advantage of whatever case unfolds To explore confirming or disconfirming cases Maximal Variation Sampling Homogenous Sampling Typical Sampling To describe a case that illustrates “dramatically” the situation To describe particularly troublesome or enlightening cases Opportunistic Sampling Confirming/ Disconfirming Sampling To generate a theory or concept To locate people or sites to study Extreme Case Sampling Theory or Concept Sampling Theory or Concept Sampling Snowball Sampling

Convenience Sampling