2. Research: Scientific Approach
“... the activity of solving problems which leads to new
knowledge using method of inquiry which are
currently accepted as adequate by scholars in the
field” (Helmstadter, 1970:5)
“... a systematic approach to searching for answers to
questions” (Hatch & Lazaraton,1991:9)
“... is the formal, systematic application of the scientific
method to the study of problems; educational
research is the formal, systematic application of the
scientific method to the study of educational
problems.”
3. TYPES OF RESEARCH
• Penelitian Eksploratif
– Memperdalam pengetahuan/pemahaman tentang
gejala/fenomena tertentu;
– Menurunkan hipotesis
• Penelitian Deskriptif
– Menggambarkan sifat, keadaan, individu, gejala atau
kelompok dalam bentuk hasil analisis statistik dasar
(deskriptif): tabel frekuensi, tabel silang, grafik, nilai rata-
rata, median. Modus dan varians.
• Penelitian Eksplanatoris (Menerangkan)
– Menguji hipotesis tentang ada-tidaknya hubungan sebab
akibat antar-variabel
4. TYPES OF RESEARCH
Luas Penerapan
Murni
Terapan
Layanan
Tindakan
Deskriptif Perpustakaan
Prediktif Lapangan
Diagnostik Laboratorium
Tingkatan Hasil Tingkat Kendali
5. Jenis-Jenis Penelitian
KUALITATIF KUANTITATIF
Introspeksi Obsevasi Praeksperimen Eksperimen
non-partisipan
Observasi Deskriptif Eksperimen
partisipan Semu
SINTETIK ANALITIK
HEURISTIK DEDUKTIF
Didorong oleh data Dirorong oleh hipotesis
Tanpa prakonsepsi Membuat prediksi
Menghasilkan hipotesis Menguji hipotesis
Produk: deskripsi/hipotesis Produk: teori
6. TYPES OF RESEARCH
BY PURPOSE
Basic vs applied
research
Evaluation research
Research &
Development
Action research
BY METHOD
Historical research
Descriptive research
Correlational research
Causal-comparative &
experimental research
7. Types of research …
Basic Research
Semata-mata untuk pengembangan dan penyempurnaan
teori;
Applied Research
Menerapkan, menguji teori, dan mengevaluasi
kemanfaatannya dalam pemecahan masalah;
Evaluation Research
Menfasilitasi pengambilan keputusan tentang manfaat
dan nilai dari dua atau lebih program tindakan
alternatif;
8. Types of research …
Research & Development
Mengembangkan produk, model yang efektif dan
inovatif;
Penelitian Tindakan
Memecahkan masalah yang dihadapi kini, di sini,
yang menerapkan pendekatan ilmiah;
“…proses yang diupayakan oleh praktisi untuk
mengkaji secara ilmiah masalah sendiri untuk
membimbing, memperbaiki, dan mengevaluasi
keputusan dan tindakan mereka.”
9. Guidelines for Classification
Cause-effect Relationship?
No
Yes
Relationship?
Prediction?
Independent Variable
Manipulated?
Yes No
Yes
No
Experimental
Causal
Comparative
Descriptive
Correlational
10. Groupings of Research Methods Found in Five Leading Texts
Ary et al. Best & Kahn Borg & Gall Gay Travers
11. RESEARCH STEPS
Selection and definition of problem
Execution of research procedures
Analysis of data
Drawing and stating conclusion
12. Define the Problem
-from education assessment
-from a review of whether initial
education/instructional objectives
were met
-from exploratory research to
clarify problem areas or as a
precursor to a full-scale survey
Specify Research Process
Formulate the research objectives or
hypotheses
Formulate the Research Proposal
-devise the research plan
-estimate time and costs
Data Search
Specify information requirements.
Explore available resources from individuals &
organisations.
Search for information from secondary sources
(published & on-line) and primary sources.
The Research Design
Create a research design: descriptive, diagnostic, predictive.
Choose an appropriate data collection method – survey,
observation, experimentation.
Sampling: decide on sampling technique
-probability or non-probability.
Data collection & processing
Data analysis – interpretation of findings
Research conclusion: evaluating & presenting results.
Problem Solving
Will the research
outcomes help to
solve the problem?
Cost/benefit
analysis
Justify costs of
the research
undertaking &
establish the
benefits to the
client.
13. Research Problem
• Is researchable
• Has theoretical and practical significance
• Appropriate:
our current level of research skills
available resources
time and other restriction
14. Problem Statement
• Indicates all variables of interest, and
specific relationship between those
variables
• Defines all variables (directly or
operationally)
* Background: justification for the study in
terms of significance of the problem
15. Hypothesis
• A tentative explanation for certain behavior, phenomenon,
or events that have occurred or will occur
• Not to “prove” but to support or not to support
Characteristics of Good Hypothesis:
• Consistent with previous research
• Tentative, reasonable explanation for the occurrence of
certain behavior, phenomenon, or events
• Clear and concise expected relationship (or difference)
between two variables
• Testable
16. Hypothesis
• Inductive: generalization based on observation;
• Deductive: derived from theory, supports, expands,
contradicts a given theory;
• Research hyp.: declarative
• Statistical hyp.: stated in null form;
• Nondirectional hyp.: relationship or difference exists;
• Directional hyp.: indicates the nature of the relationship or
difference;
Ss who get X do better on Z than
Ss who do not get X (or get some other X)
17. Research Plan: Components
Introduction
- Statement of the problem
- A review of related literature
- A Statement of the hypothesis
Method
- Subjects (population: characteristics,
size from which sample will be
selected)
- Instruments
- Materials/apparatus
- Design: (basic structure of study,
variables involved, the number of
groups, wehther randomly formed,
pretest, if any)
Procedure
- Steps to follow from beginning to end;
- Technique used in selecting sample
- Administration of pretest and posttest, if any.
Data Analysis
- Statistical technique to be used to analyzed
study data depending on:
* how groups are formed (random
assignment, matching, using existing groups);
* How many different treatment groups
involved;
* How many independent variables involved;
* The kinds of data collected (nominal,
ordinal, interval, ratio)
Time Schedule
Budget
18. Sampling
• Population
• The group to which
research results to be
generalizable
* Target/theoretical population
* Accessible / available
population
• Sample
a number of individuals
selected to represent the
population from which they are
selected
Representative:
homogeneous
heterogenous
21. Measurement Instruments
Validity
Content Validity
Construct validity
Concurrent validity
Predictive validity
Reliability
Test-retest reliability
Equivalent-forms rel
Split-half reliability
Interscorer Reliability
Intrascorer reliability
• Types of Measurements
Achievement test
Non-projective instr.
Personality inventory
Attitude scales
Tests of creativity
Interest inventory
Learning style inv.
Aptitude test
- Standardized instrument
- Self-developed instrument
- Naturally available data (e.g. GPA)
22. VALIDITY
Validity
The degree to which a test
measures what it is supposed to
measure;
* for a particular group
* for a particular purpose
Content Validity
• Measures intended content area
• Item validity: test items
represent measurement in the
intended content area;
• Sampling validity: how well the
test samples the total content
area;
• By expert judgement
Construct Validity
• The degree to which a test
measures an intended
hypothetical construct;
• Nonobservable trait;
• Involving testing hypothesis
deduced from theory
concerning the construct;
Concurrent Validity
The degree to which the scores on a
test are related to the scores on a
another;
Determining relationship between
scores on the test and the scores on
some established test or criterion;
23. VALIDITY …
• Predictive Validity
The degree to which a test can predict how well an
individual will do in the future situation;
Establishing the relationship between scores on
the test and some measure of success in the
situation of interest;
Test predictor
Behavior predicted criterion
24. RELIABILITY
The degree to which a test
consistently measures
whatever it is supposed to
measure;
Expressed numerically, a
coefficient;
High coefficient
High reliability
Test-retest Reliability
The degree to which scores are
consistent over time;
The same test administered to
the same group at different
occasions;
Correlation of two sets of
scores coefficient of
stability;
Problem:Time interval
too short: remember responses;
too long: increase ability,
maturation;
25. RELIABILITY …
Equivalent-form Reliability
Two tests identical in every
way except for the actual items
included;
Establishing relationship
between scores resulting from
administering two different
forms of the same test to the
same group at the same time;
Correlation coefficient of
equivalence
Split-half Reliability
Establishing relationship
between the scores on two
equivalent halves of a test
administered to a group at one
time;
Called: internal consistency
reliability;
Apply Spearman-Brown
prophecy formula:
2r split half
rtotal test = -----------------
1 + r split half
26. RELIABILITY …
Rational Equivalence Reliability
Establishing internal consistency by determining how all items on a
test relate to all other items and to the total test;
Determined by applying Kuder-Richardson formula (KR-20 or KR-
21);
Require items scored dichotomously: correct or incorrect (1 or 0);
KR-21 formula:
(K)(SD2) – X (K-X)
r total test = -------------------------
(SD2) (K-1)
Where: K = the number of items in the test
SD = the standard deviation of the scores
X = the mean of the scores
27. Standard Error of Measurement
An estimate of how often we can expect errors of a given
size;
Small SE-m indicates high reliability;
Large SE-m indicates low reliability;
SE-m allows ua to estimate how much difference there
probably is between a person obtained score and true
score; the size of this difference being a function of the
reliability of the rest;
SE-m = SD 1 – r
Where SEm = standard error of measurement
SD = standard deviation of the test scores
r = the reliability coefficient
28. Descriptive Method
1. Self-report research
- Survey research
- Developmental studies
- Follow-up studies
- Sociometric studies
2. Observational research
- Nonparticipant observation (naturalistic obs., simulation obs., the case
study, content analysis);
- Participant observation
- Ethnography
Statement of the problem
Selection of subjects
Construction of the Questionnaire
Validation of the questionnaire
(interview guide);
Pretesting the questionnaire
Analysis of results
Defining observational variables
Recording of observation
Assessing observer reliability
Training Observers
Monitoring observers
29. Correlational Method
Collecting data to determine whether, and to what degree, a relationship
exists between two or more variables
Degree of relationship: correlation coefficient (between +1.00 and –1.00)
Corr. Coefficient: negative or positive
- Problem selection
- Sample and instrument selection
- Design and procedure
- Data analysis and interpretation
Relationship studies
Prediction studies
30. Causal-comparative Method
CC or ex-post facto, research: attempt to determine the cause, or reason, for
existing differences in behavior or status of groups of individuals; to
identify cause-effect relationship (tentative);
Starting with an EFFECT, then seeking POSSIBLE CAUSES;
(Variation: CAUSE first then investigating EFFECT)
Independent variable (alleged cause) is not manipulated, already occurred
- Design and procedure
- Control procedures (lack of randomization, manipulation, control)
- Matching, comparing homogeneous groups/subgroups, ancova
- Data analysis and interpretation
Descriptive and inferential statistics
t-test (two groups), analysis of variance (more than two groups),
31. Experimental Method
Manipulates at one independent variable, controls other variables, and
observes the effect on one or more dependent variables;
IV = experimental variable, treatment, cause (activity/characteristics
believed to make a difference)
DV = criterion variable, effect, posttest
- Selection and definition of Problem
- Selection of subjects and measuring instrument
- Selection of design
- Execution of procedures
- Data analysis and formulation of conclusions
32. Threats to Experimental Validity
Uncontrolled extraneous variables
Internal Validity
condition that observed differences on the DV are a direct result of
manipulation of the IV, not some other variables;
(History, maturation, testing, instrumentation, statistical regression,
differential selection of subjects, mortality)
External Validity
condition that results are generalizable, or applicable to groups and
environments outside of the experimental setting;
(pretest-posttest interaction, selection-treatment interaction, specificity of
variables, reactive arrangements, multiple treatment interference,
contamination and experimenter bias)
Control of extraneous variables
Randomization
Matching
Comparing homogeneous groups/subgroups
Using subjects as their own control
Analysis of covariance
33. Group Designs
Pre-experimental designs
The one-shot case study X O
The one group pretest-posttest design: O X O
The static group comparison: X O X1 O
--------- ----------
O X2 O
True experimental designs
The pretest-posttest control group design R O X1 O
R O X2 O
The posttest only control group design R X1 O
R X2 O
The Solomon four group design R O X O
R O O
R X O
R O
Quasi-experimental designs
Non-equivalent control group design O X O
--------
O O
Time series design O O O O X O O O O
Counterbalanced design X1 O X2 O X3 O
X3 O X1 O X2 O
X2 O X3 O X1 O
35. A model for stating hypotheses for an experimental or
causal-comparative study.
• If X is the independent variable, Y is the
dependent variable, and S is the Subject, we can
state our research hypothesis as Ss who get X do
better on Y than subjects who do not get X (or
get some other X).
• Here is an example: In our study we will
hypothesize that first grade girls will show better
reading comprehension than first grade boys.
Now lets state that as a research hypothesis. It
would be a causal-comparative research study.
36. • Research Hypothesis: Girls will achieve higher reading
comprehension test scores than boys at the end of the
first grade.
• Operational Variables: Reading comprehension will be
measured by the Iowa Tests of Educational
Development, Reading Comprehension, administered at
the end of the year.
• Statistical Hypotheses: When we get ready to analyze
our data we might also wish to state statistical
hypotheses for our problem. The statistical hypotheses
consist of the null hypothesis (H0) and the alternative
hypothesis (H1). If we let stand for the mean of the girls
and stand for the mean of the boys, our null and
alternative hypotheses would be:
37. • In other words the null hypothesis states that there is no
difference between the two means on the reading
comprehension test scores, while the alternative
hypothesis states that the girls mean score on the reading
comprehension test will significantly exceed that of the
boys. Generally we use a statistical test (e.g. the t-test) to
decide if the girls score significantly higher than the
boys. We will discuss statistical hypotheses and the use
of inferential statistics in lessons 9 through 13 when we
look at the design, proceedures, and data analysis for
each of the types of quantitative research types.