Research Methodology

Research Methodology



The difference between action research and experimental research:
Action research examines events. It is qualitative in nature; it is also teacher-generated. Action research projects are often longitudinal i.e. carried out over a period of time.
The strength of this research is that it allows teachers to diagnose, investigate and resolve classroom problems. This type of research has the specific aim of addressing specific situations / problems in the classroom.. it is based on participant observations, open-ended.
Experimental research examines outcomes, it is quantitative. It attempts to quantify outcomes resulting from treatments. It examines the behaviour of a representative cross-section of a population at one or more instances in time.
 Both approaches are complementary.
Action research is a practical tool for short term problem solving.
Experimental research is useful for achieving long term abstract aims, it is process-oriented and reliable

Ethnographic method  Action research. Inductive ‘bottom up’ method.
Non- statistical analysis. Observational data are used.
Qualitative data. Ungeneralizable.

Psychometric method  Experimental research. Deductive “top-down”
Statistical analysis.
Quantitative data. generalizable

In experimental approach, a number of important aspects have to be taken into consideration:
- Internal validity: the results obtained must be the result of the treatment applied and not nay other extraneous variables.
- External validity: refers to our ability to generalize the results of our study to other groups and other setting. Would these results hold true for the same population? And for other population?
- Internal reliability: refers to consistency in data collection, analysis and interpretation of results.
- External reliability: it must be possible to reproduce the research under identical conditions.
- Quasi Experiments: there are both experimental and control groups, but subjects are not randomly assigned to these groups
- Construct Validity: it is appropriate when your test is measuring an abstract concept.

Classroom interaction research:
In this type of research Chaudron suggests 4 traditions:
- The psychometric tradition: the measurement of outcomes on proficiency tests
- Interaction analysis: observation and analysis of social interaction
- Discourse analysis: observation and analysis of linguistic interaction.
- Ethnographic tradition: interpretation of behaviour from participant’s point of view.



Research Question:
A good research question should be evocative, relevant, clear and researchable. The RQ can be descriptive: describe what is going on or what exists; relational: designed to look at the relationships between 2 or more variables; causal: designed to determine whether 1 or more variables cause or effect 1 or more outcome variables.
The RQ should begin with a general topic that is narrowed down to a specific Q. it has to be short, sharp, specific and free from value judgments and bias.


Variables:
It is a participant’s characteristic that varies for group members.
A construct: a characteristic that represents human beings.
- Quantitative variables: are ones that exist along a continuum that runs from low to high (e.g. scores of tests).
- Qualitative variables: they do express differences in amount, they are called categorical variables e.g. gender, religion…
 A variable is an observable or quantified representation of a construct.
- Dependent Variable: the variable that is influenced by another variable. The dependent variable is the major variable that will be measured in the research.
- Independent variable: it stands alone, affects the dependent variable, and cannot be affected by other variables.
- Moderator variable: modifies the relation ship between variables.
- Control variables: are variables that the researcher choose to keep constant
- Intervening variables: make the relationship between the dependent and the independent variable happen.
- Extraneous variables: an independent variable that the researcher did not control it, it may or may not influence the result.

Formulating the hypothesis:
There are 3 phases in constructing the hypothesis: hunch (guess), test and investigate, verification (confirmed or rejected). A hypothesis can be right – wrong – partially right, it should be operationalizable, unidimensional, verifiable, rooted in the body of knowledge, conceptually clear. A hypothesis comes from the researcher’s expectations generating from his knowledge (the review of literature). A hypothesis expresses the relationship between 2 or more variables; it is a tentative statement about the outcome and results of the research.
After collecting data and testing the hypothesis, the hypothesis is either supported, rejected or partially supported.

Methods of data collection:
How the researcher obtains the empirical data to be used to answer his or her research questions. There are 6 major methods of data collection:
Tests: standardized test that includes information on reliability, validity and nouns as well as test constructed by researchers for specific purposes. They are used to measure aptitude, performance, specific knowledge and skills.

Strengths weaknesses
- Can provide measures for many characteristic - Can be expensive
of people. - Often standardized. - Reactive effects such as social
- Strong psychometric properties desirability can occur.
- Allows comparability between research - open ended questions and probing are
population. not available
- Can provide quantitative data - some tests lacks psychometric data


Questionnaires: (can be called survey) is a self report data collection instrument that is filled out by research participants. There are some principles of questionnaire construction: Q items should match with the research objectives / understand the participants (cultural background..)
Use familiar and not technical or highly specialized language / items should be clear, not long, precise, not leading or loaded / avoid combining 2 or more issues in a single Q, avoid double negation.
Piloting questionnaires: we need what is called “guinea pigs” i.e. people similar to our sample so that we try out the questionnaire on them, check the format, content, ambiguity, time…
Types of questions:
a- Open- ended: provide qualitative data in the participant’s own words for e.g. WH questions, sentence completion…
b- Closed-ended: provide exact information needed by the researcher, they provide quantitative data; dichotomous questions: yes /no. MCQ: choices and categories re provided- have to be discrete, not overlapping. Rank ordering: we rank categories in terms of preference, intensity…Rating scale: 5 categories is the best number.

 A structured question is based on closed –ended items.

Strengths weaknesses
- Good for measuring attitudes and eliciting - usually must be kept short
other content from participants - reactive effects may occur
- Inexpensive mainly mail Q - non response to selective items
- Quick turnaround - data analysis can be time consuming
- Ease of data analysis especially closed ended Q



Interviews: it is a way of collecting data orally; it can be quantitative (closed-ended Q) or qualitative (open-ended Q). A structured interview is planned, including interview protocol and listing the Qs to be asked mainly open-ended ones; whereas unstructured interview is less formal using unplanned questions.

Strengths weaknesses
- Allows probing and follow up questions. - usually in-person interviews are
- Provide in depth information expensive.
- High measurement validity for well constructed - reactive effects may occur
interviews - can be time consuming
- Useful for exploration and confirmation. - measures need validation



Focus groups: it is a situation where a focus group moderator keeps a small and homogenous group focused on the discussion of a research topic or issue.

Strengths weaknesses
- Allows probing - reactive effects may occur if
- allows quick turnaround. participants notice that they are being - Provide in depth information watched and it is difficult to
generalize results
- may be dominated by 1or 2
participants

Observation: the researcher observes participants in natural environment. There are 2 types of observation: - laboratory (done in lab set up by the researcher) or - naturalistic (done in real world settings).
There are 2 forms of observation: - quantitative (sampling procedures are often used in quantitative observation) - qualitative (is exploratory and open-ended). The qualitative observer may be: Complete participant (a member of the group but without telling the participants that they’re being observed).  Participant as observer (informing the participants that they’re being observed).  Observer as participant (spending a limited time inside the group and informing them).  Complete observer (observing from outside).

Strengths weaknesses
- Allows the researcher to see directly what pple do - reactive effects may occur if
- Provide firsthand experience respondents know that they are being - Objective measurement of behaviour watched and it is difficult to
- Can be used with subjects with weak verbal skills - data analysis can be time
consuming
- sampling of observed people may be
limited

Secondary / Existing data: data originally used for a different purpose as opposed to primary data (collected for a new research). Secondary data can be: - document (letters, diaries, newspapers, reports…) - physical data (a data that was created or left by humans that might help the researcher) – archived data (old data collected by another researcher for other purpose, often save in CDs or tapes…)

Strengths weaknesses
- Reactive effects are very unlikely to occur - may be incomplete
- Provide insight into what pple think and do - may be representative of only 1
- Useful for exploration perspective
- Ease of data analysis for archived data - may not apply to general population
- archived data may nit be available
for the population of interest to the
researcher

Sampling
It is the process of selecting units (population, organization) from a population of interest so that by studying the sample we can generalize our results.
Sampling: drawing a sample from a population
Sample: a set of elements taken from a larger population
A statistic: is a numerical characteristic of a sample
A response rate: the % of people in the sample selected for the study.
Sampling frame: a list of all the people that are in the population
Population mean: general characteristic.
Sampling methods: - simple random sampling (equal chance for each member) / - systematic sampling (selecting every 10th member) / - stratified sampling / - cluster sampling
Principles of sampling: P1- there is a difference between the sample statistic (ss) and the population mean. P2- the larger the sample is the smaller the difference between the ss and population mean would be. P3- the larger the difference in the variables the less accurate and representative the sample would be.


Ethics: is a code of conduct in which we can find different principles.
Ethical principles:
 Informed consent: what, when, here, how, why
 Coercion (obliging someone)
 Deception: the researcher is ethically obliged not to use more deception than is needed (deception is when the researcher provides misleading information about the purpose of the study)
 Freedom to withdraw: participants must be informed that they’re free to withdraw at any time they want.
 Confidentiality and anonymity the researcher agree not to reveal the identity of the participants. Anonymity means that the identity of the participant is not known by anyone including the researcher.


Deduction vs. Induction:
Deductive reasoning works from the more general to the more specific, sometimes this is called a top-down approach, we light begin with a theory about our topic of interest; we then narrow down into more specific hypothesis that we can test, we narrow down even further when we collect observations. This ultimately leads us to be able to test the hypothesis with specific data (a confirmation of our original theories).
 Deductive is quantitative: start from the theory, research, confirm the initial theory.

Inductive research works the other way moving from specific observation to broad generalization and theory. We sometimes call this a bottom-up approach. In inductive reasoning we begin with specific observations and measures, detect patterns and regularities, formulate sensitive hypothesis that we can explore and finally end up developing some general conclusions and theories.
Inductive reasoning is more open-ended and exploratory, deductive reasoning is narrower and is concerned with testing or confirming hypothesis.

- Qualitative research tends to use the inductive form of the scientific method; it is especially strong in describing and exploring phenomena and generating explanations, furthermore, qualitative research is very helpful in adding new dimensions of understanding.
 Nevertheless, we take a mix research view and believe that both qualitative and quantitative research help each other in understanding the world; the yare complementary approaches to research.

- Triangulating data: compare qualitative with quantitative data and then confirm the
hypothesis.


Key Features of Mixed Research: mixed research tries to mix the best of qualitative and quantitative into research studies. Mixed research takes an eclectic and pragmatic approach, suggesting that the researcher mix quantitative and qualitative data that works best for the given research question being studied. Mixed research uses both inductive and deductive methods.

Operationalize a research question: once we know what the question is, we have to identify ways to test it through careful manipulation. This manipulation requires the definition of at least one independent and one dependent variable.

Population Vs Sample: a sample is a set of elements taken from a population that is a small group of persons or elements selected from the population which is the full set of elements or people.

Experimental Vs control group: experimental receives the experimental treatment, a control group gives a point comparison, e.g. if one group receives a pill we also need to know what would have happened in the absence of receiving a pill.

Components of a research plan:
1-Introduction: it includes a statement of the topic, problem, prior literature, statement of the research purpose, the research questions and any hypotheses.
2-Method: it includes information about the research participants, data collection instrument, the procedures followed in carrying out the study.
3-Data analysis: includes information on how you intend on organizing and analysing the data you collected.
4- Conclusion.

Primary sources vs secondary sources:
Primary sources are designed by the researcher whereas secondary sources are left behind by others.
All primary sources of data shall match the research objectives.