KMF2024 Research Methodology
1_Introduction to RESEARCH
Purpose : secondary research
helping others understand the research results
using the results in appropriate contexts
raising new topics/question to study
Purpose : primary research
inquiry to gain understanding about some problems
Understanding things
Inductive (specific to general) - qualitative
developing generlisations from limited number of related observations or experiences
Deductive (general to specific) - quantitative
developing specific predictions from general principle, observtions, or experiences
Fundamental differences
Quantitative
state hypotheses before data are collected
controlled context
rarely interact
large number of participants
Qualitative
Not to state hypotheses before data are collected
uncontrolled context
interact extensively
small number of participants
Limitation: Scientific and Disciplined Inquiry Approach
inability to answer 'should' questions
inability to capture the full richness of the research site and participants' complexity
limitions of measuring instruments
the need to address participants' ethical needs and responsibilities
2_Ethics in RESEARCH
Basic Principles of ethical research
Informed consent
Coercion - not be forced into participation
Confidentiality
Debriefing - provide any necessary info about the nature of the research
Deception
Withdrawal from the investigation - participants rigth to withdraw at any time
Protection from harm (both physically and psychologically)
3_Selecting and Defining a Research Topic
Identify a topic or question to research
solving problems rather than finding a problem or topic to study
realities of research andf topic indentification
True research
identified topic - plan to carry out the research - collect pertinent data - analyse - make research outcome available
Non linear
Sources of research topics
Theories
Personal questions
observe surrounding environments, society, news articles, and ask questions
Replication
not identical but altered
Library search
most useful after a topic has been narrowed
Characteristics of GOOD research topic
Interesting
Researchable (can be investigated thru collection and analysis of data)
Significant (contributes to the improvement)
Manageable
Ethical
Stating the research topic
Quantitative (specific)
Qualitative (more general initially)
Review of related literature (start with backgorund of studies)
Identify, locate and analyse documents
f(x) - determine what has already been done / what needs to be done / rationale for research hypthesis / facilitates interpretation of study results.
when to stop ??? when search and benda sama je yang kua
evaluating the sources
relevancy (keyword - at least 3)
Quality (e.g. date of publication, refereed or not journals - must have peer review , evidence to support interpretations)
Astracting
involves locating, reviewing, summarising, and classifying references
Analysing, organising, and reporting the literature
A good hypotheses
based on sound reasoning
provides a reasonable explanation for the predicted outcome
clearly states the expected relationship between defined variables
testable
4_Selecting a Sample
Definition
SAMPLING - process of selecting a number of participants for a study in such a way that they represent the larger group from which they were selected (population).
POPULATION - the group of interest to the research to which the result of the study will ideally generalise (first step in sampling)
Target population (generalise)
Accessible population (realistically select from)
Basic random sampling techniques
Simple random sampling
a type of probability sampling in which the researcher randomly selects a subset of participants from a population.
Stratified sampling
researchers divide subjects into subgroups called strata based on characteristics that they share (e.g., race, gender, educational attainment).
Cluster sampling (most useful & feasibls method) (less time and expense)
a probability sampling method in which you divide a population into clusters, such as districts or schools, and then randomly select some of these clusters as your sample
Systematic sampling
a probability sampling method where researchers select members of the population at a regular interval
Determining sample size
10-20% of population (descriptive research)
Selecting a nonrandom sample
convenience sampling
units are selected for inclusion in the sample because they are the easiest for the researcher to access.
purposive sampling
units are selected because they have characteristics that you need in your sample
quota sampling
data gatherers are given exact characteristics and quotad of persons to be interviewed
obtained from easily accessible individuals
5_Experimental Research
Experiment?
a research situation in which at least one independent variable, called the experiment variable, is deliberately manipulated or varied by the researcher.
Experiment Variables
an independent variable but not all independent variables are experimental varibales
Criteria (Well-Designed Exp)
Adequate experimental control
Lack of artificiality
Basis for comparison
Adequate information from the data
Uncontaminated data
No confounding of relevant variables
Representativeness
Parsimony
all characteristics being equal, a simpler design is preferred to a more complex one
Experimental Validity
Internal validity
basic minimum of control, measurement, analysis, and procedures necessary to make the results of the experiment interpretable
External validity
deals wirh the extent of generalisability of the results
Threats
History
Maturation
Instrumentation
Statistical regression
Differential selection of subjects
Experimental Designs
Posttest-Only Control Group Design (control for most of the same threats to internal and external validity as the classic controlled experimental design.)
Pretest-Posttest Control Group Design (a true experimental research design that tests two groups before and after both groups receive the independent variable)
Solomon Four-Group Design
Factorial Designs (an experiment that has multiple factors or independent variables)
Repeated Measures Designs (an experimental design where the same participants participate in each independent variable condition)
Quasi-Experimental Research (studies that aim to evaluate interventions but that do not use randomization)
Posttest-Only, Nonequivalent Control Group Design (participants in one group are exposed to a treatment, a nonequivalent group is not exposed to the treatment, and then the two groups are compared)
Pretest-Posttest, Nonequivalent Control Group Design ( a treatment group that is given a pretest, receives a treatment, and then is given a posttest)
6_Correlation Research
Definition
determine whether, and to what degree, a relationship exists between two or more quantifiable variables
degree of relationship
Purpose
determine relationships between variables OR use these relationships to make predictions
Process
Problem selection
Participant and Instrument Selection (30 min participants)
Design and Procedure
Data analysis and Interpretation (indicates the size and direction of a relationship)
near +1.00 (high size and +ve direction)
near 0.00 (variables not related)
near -1.00 (high size and -ve direction)
reliability
high .90s
moderate .80s
low .70s
Data collection
identify variables - indentify population(select a sample) - administer instruments - collect data (short period of time)
Data analysis
computing a correlation coefficient
linear relationship
curvilinear relationship
11_Action Research
an approach designed to develop and improve practice
help to make informed decisions-decisions that are data driven
Characteristics
Five C's
commitment
collaboration
concern
consideration
change
Conducting
Identify a topic or issue to study
conducting a literature review
developing research questions
Collect data
concept of triangulation (e.g. interview, observation, survey, etc)
spiraling nature (looking;observes the setting, thinking;interprets, acting;identifies a solution)
Analyse and interpret the collected data
Carry out planning and share the findings (verbally or written)
Taking action
help understand or try out new or needed methods
leads to : new question, new forms of understanding and deeper insights in practice
10_Qualitative Research
Nature
understanding of a social setting or activity as viewed from the perspective of the research participants
General steps
Indentify research topics
Review of research
Selecting participants
Collecting data
Analysing data
Reporting, evaluating, and interpreting research
Approaches
Case study
Ethanography (seeks to describe and analyse all or part of the culture of a community by identifying and describing)
Ethology (aims to generate a theory that explains, at a conceptual level, a process, and action, or a concept) - observation and interviews
Ethnomethadology
Grounded theory
Phenomenology
Symbolic interaction
Action research ( a study that can solving problem)
step 1 : selecting a research topic
step 2: reviewing the literature
step 3: selecting research participants
step 4: collect data
interview
experiences and feelings
types
structured (a specified set of questions)
unstructured (question being prompted by the flow of the interview)
semistructured (combination)
observation
types
participant observer
external or nonparticipant oberver (one who wathces but does not participate)
Field notes (records of what the observer has specifucally seen or heard - emid data)
Protocol (process / checklist) - step by step
memo writing (a form of thinking on paper)
Threat quality
observer bias
Halo effect (initial impressions concerning an observe affect subsequent observations)
observer effect
Triangulation (combination of qualitative, quantitative and documentation) - used for corrobrate data
Historical research
Data analysis
multistage process of organising, categorising, synthesising, interpreting, and writing about the data
Qualitative research
Data managing
Reading/memoing
Describing the context and participants
Classifying
Interpreting
Cyclical process
9_Inferential Statistics
inferences about about populations based on the results of samples
Standard Error
not affect realability
chances of any sample being exactly indentical to its population are virtually nil.
sampling error - expected variation among the means
sampling error (a statistical error that occurs when an analyst does not select a sample that represents the entire population of data)
normally distributed
mean of the means
standard error of the mean (SD > bahaya , so kena cut)
Test of significance
Two-tailed test (two direction) A=B
One-tailed test (one direction)
t Test (two means)
pretest-posttest
t-test for pretest, if the difference is not significant, then t-test on the posttest
t-test for pretest, if the difference of significant, then ANCOVA (analysis of covariance)
ANOVA one way (two or more means)
post hoc test (determine which means are significantly different from other means)
Chi Square (nominal data)
8_Descriptive Statistics
Preparing data for analysis
scoring procedures
standardised instrument (follow test manual)
self-developed instruments
tabulation and coding procedures
tabulation : organising data, 2 x 2 factorial
code : ID for each participant (e.g. 1 for M and 2 for F )
major types
measures of central tendency
typical or average score of a group score
measures of variability
how spread out a group of scores are
measures of relative position
comparison
measures of relatioship
the degree to which two sets of scores are related
Pearson, r
Spearmen, Rho
7_Survey Research
assessing - attitudes, opinions, preferences, demographics, practices, and procedures
classifying
collection of survey data
self report
questionnaire
crtiticisms (misuse of d data)
paper-and-pencil Q (less time, less expensive, collection from much larger sample)
stating problem (no trivial issues)
selecting participants
constructing the Q (attractive, brief, easy to respond to)
followup activities (initial followup strategy)
Interview
telephone
observation (quantitative)
survey frequency
cross-sectional survey (advantage - convenience / disadvantage - selecting comparable samples)
single time period
stand-alone study
entire population - census survey
longitudinal survey (advantage - no comparability problem / disadvantage - sample tend to shrink over time, extended commitment)
collect 2 or more times
study the dynamics of an issue over time