Penyelidikan Pengenalan & Konsep Penyelidikan
The role of research is to provide a method for obtaining those answers by inquiring studying the facts, within the parameters of the scientific method. Pencarian Ilmu Pengetahuan Think systematically about thinking
Deductive – moving from the general assumption to a specific application
Inductive – moving from specific observations to generalizations
Assumption or hypothesis, tested by the collection and logical analysis of data.
The deductive-inductive method recognized as an example of a scientific approach. Science An approach to the gathering of knowledge 2 primary functions the development of theory; testing of substantive hypothesis deduced from theory Research on human nature; no two persons are alike no one person are completely consistent human are influenced by the research lack of adequate definition Role of theory
An attempt to develop a general explanation for some phenomena
Establishes a cause-and-effect relationships
Defines non-observable constructs that are inferred from observable facts and events that thought to have an effect Describes the relationships among key variables or explaining a current state or predicting future occurrences. Hypothesis
A formal affirmative statement predicting a single research outcome; a tentative explanation of the relationship between two or more variables
Operationally defined
The null hypothesis relates to a statistical method of interpreting conclusions about population characteristics that are inferred from the variables relationships observed in samples. Sampling
To draw valid inferences on the basis careful observation of small proportion of the population.
A population is any group of individuals who has one or more characteristics in common
A sample is a small proportion of a population selected for observation and analysis. Randomness
Selecting (who) are representatives of the population
Simple random sampling
Equating experimental and control groups (equal independent chance of being assigned to each of the groups). Systematic sample; Cluster sample Non-probability samples Sample size
Purposes of Research
Basic research / fundamentals – leads to knowledge for knowledge’s sake; development and testing of theories
Applied research – improving a product or process; to develop generalizations
Actions research – focus on immediate application; local settings; to improve practices. Types (classified/approach) of Educational Research
Historical research – what was? Descriptive research – what is? Experimental research – what will be?
Dimensi Penyelidikan Dimension I : Why? Discovery of new information and the resolving of old problems Powers the world The ways in which we proceed to solve problems (methodology)
Dimension II : Steps Beginning with a problem. Data relating to the problem. That was fact. Rationalization, guess: logical reasoning; a hypothesis. Another fact. Confirmed the hypothesis The problem was resolved. Procedure It originates with a problem It ends with a conclusion Process based upon observable facts [data] It is logical It is orderly It is guide by a reasonable guess. It confirms or rejects the reasonable guess Conclusions on basis of the data, only data Conclusion resolve the problem. Basic Pure Research Subjective The skills of articulation Planning; how might be accomplished Problems and sub-problems Through appropriate hypotheses Assumptions provide the foundations Data - specific and measurable By its nature, is circular. Dimension III : What is Research The systematic and objectives analysis and recording of controlled observations that may lead to the development of generalizations, principles, or theories, resulting in prediction and possibly ultimate control of events. Dimension IV : Where ? Indexes Reviews Volumes Net Journal Tools of Research The library and its resources Techniques of measurement Statistics The computer and its software Facility with language. “Masalah” : ‘Jantung Penyelidikan’
Mengenal pasti masalah The problem is the axial center around which the whole research effort turns. The statement of the must be expressed with the utmost precision. The problem is then fractionated into more manageable subproblems. So stated, we can then see clearly the goal and the direction of the entire research process. Characteristics Personal or researchable problems Rarely happen by accident Grounded and possess the ability Fruitful conclusions Specialist rather than generalists Rifle rather than shotgun Limited rather than broad Testing rather than proving Sources? Daily classrooms School Community Technical changes Curricular developments Educational innovations Academic experiences Reading assignments, textbooks, journals Consultation Keeping in Focus Acceptable research problem? What is to ‘think’ What is to do To formulate a problem that is carefully phrased and represents the single goal of the total research effort. What am I doing, for what purpose am I doing it? A Good Research Problems Qualities of significance Originality Feasibility What is not a Research Problem? Don’t use a problem in research as a ruse for achieving selfenlightment. Problem whose sole purpose is merely to compare to sets of data. Finding correlations between to sets of data merely to show relationship. Problems that result in a yes or no answer are not suitable problems for research.
The Statement of the Research Problem In a complete grammatical sentence in as few words as possible. Think, consider and estimate. Say precisely what you mean Edit your writing Least words possible Use a thesaurus Keep sentence short. Look critically at each thoughts Be alert to modification. Evaluating the Problem Can be effectively solved through the process of research? Significant? Is the problem new one? Feasible? - Am I competent? Are pertinent data accessible? Financial resources. Time to complete. Social hazards.
Using the Library
Finding Related Literature The Education Index Resources in Education Current Index to journals in Education Index to Doctoral Dissertations DAI Others – Psychological Abstracts, SSCI
The Review of the Related Literature Sorotan Kajian Kajian Literatur Definition
Involves systematically identifying, locating and analyzing documents containing info related to the research problem.
These documents include articles, abstracts, monographs, dissertations, books, other research reports and electronic media. The 8 Purpose
It can reveal investigations similar [what has already been done], and it can show how the collateral researchers handled these situations.
It can illuminate a method of dealing with a problem situation that may suggest avenues of approach similar difficulties you may be facing.
It can introduce to significant research personalities, of whose work and collateral writings.
It can help to see own study in historical and associational perspective.
Rationale for research hypothesis and justifying the significance of the
It can provide with new ideas and approaches. It can help to evaluate own research efforts; what need to be done.
It provides the understandings and insights necessary to develop a logical framework into your research topic. study. Getting Started
Familiar with the Library Make a list of keywords to guide Secondary sources Primary sources
Remarks
The results can be discussed in terms of whether and how they agree with previous findings
Often have difficulty in determining how broad; Which are ‘related enough’?
Tips
Avoid the temptation to include everything you find; bigger does not better.
Heavily researched areas usually provide enough references related to a specific problem to eliminate the need relying on less related studies.
New or little-research problem areas usually require review of any study related in some meaningful way to the problem in order to develop a logical framework and sound research hypothesis. Analyzing, Organizing & Reporting
Call for a different style of writing
Clarifying definitions and using them consistently
Technical writing requires documenting facts, and substantiating opinions, clarifying definitions and using them consistently Using an accepted style manual, Starting sections with a intro and ending with a brief summary
Guidelines
Make an outline Sort your references into appropriate topic piles
Analyze the relationships and differences between references in a given subheading
Do not present as a series of abstracts or annotations Discuss references least related to the problem first Conclude with a brief summary and its implications
How?
Get the proper psychological orientation Have a plan Emphasize relatedness Review, don’t reproduce it. Summarize what you have said
Interpretasi & Analisis Data Penyelidikan Constructs Regardless of the type of research, must collect data. The scientific approach is based on the collection, analysis and interpretation of data. Data are the pieces of information collected and use to examine topic, hypotheses or observation. The relationship between constructs, variables and instruments. Constructs Abstractions that cannot be observed directly They are invented to explain behavior e.g intelligence, personality, ability, creativity, Must be operationally defined, in terms of processes or operations that can be observed or measure When are operationally defined, they become variables. Variables A construct that can take two or more values or scores People differ in them Many different approaches and instruments to measure variable. An instrument is a tool used to collect data. Can be represented by different kind of measurements, identified as categorical or quantitative, dependent or independent Measurement Scales & Variables
4 types of measurement scales and associated variables; ordinal, nominal, ratio and interval Measurement scale consists of a group of several related statements that participants select to indicate their degree of agreement or lack of agreement; to obtain a the actual range of values or scores Different scales require different methods of analysis
3 Characteristics of Measuring Instruments Standardized instrument Self-developed instrument Record naturally or available data (observations or existing grade) Instrument Terminology Test A standardized test Assessment Measurement – the process of quantifying or scoring persons’ performance on assessments. Occurs after data collected 3 Types of Data Collection Paper-and-pencil Observations Interviews 4 ways of Interpreting Raw scores Using the bell-shaped curve, transformed into percentile ranks, stanines and std scores (Descriptive Statistics) 2. Norm-referenced 3. Criterion-referenced 4. Self-referenced Good Measuring Instruments Validity – the appropriateness of the interpretations made from test scores Reliability – dependability or trustworthiness Ease of se test Self-constructed tests should be pretested Test Administration Arrangements should be made beforehand Ideal testing conditions Familiar with the administration procedures Be prepared Pemungutan Data Kualitatif Qualitative Method
Ethnography/fieldwork/observation Interviewing, asking questions and conversations Documentary research Classroom observations Case study
Characteristics of Qualitative Research Diversity apparent Variety of setting, multidisciplinary Focus upon natural, ordinary, routine Data collected in a number of way
Formulated hypotheses Inductive way Developing grounded theories Emergent, creative and open-ended
Case Studies Concern with the rich and vivid description of events within the case Chronological narrative of events within the case Integral involvement of the researcher in the case Prescribing the case which is able to capture the richness of the situation Temporal characteristics; geog parameters; boundaries; individual in a particular context; characteristics of the group; role of function; shaped of organizational or institutional arrangements Action Research Systematic inquiry to collect and study data that can help to understand and improve own practice. Reflect own practices, identify areas that need improvement, Collect data pertinent to issue of interest, analyze data and to determine whether results do in fact improve practice or understanding Participants
Identify the number, source, and characteristics of the sample
Defining a Population
Also define the population
Differ depending on approaches Selecting a Sample
Selecting a Random Sample - Simple random sampling - Stratified sampling - Cluster sampling - Systematic sampling Selecting a Nonrandom Sample - Convenience sampling - Purposive sampling - Quota sampling
Qualitative sampling Most often deals with small, purposive sampling. The R’s insights guide the selection of participants Intensity sampling Homogeneous sampling Criterion sampling Snowball sampling Random purposive sampling
Instruments
How they will measure the variables stated in the hypotheses The appropriateness for the study and sample The measurement of properties The process of administering and scoring
To develop own … Measurement Scales
Consists of a group of several related statements that participants select to indicate their degree of agreement or lack of agreement. Variables
-
Is a construct that can take on two or more values or scores. Nominal/categorical variables Ordinal variables Interval variables Ratio variables
Dependent and Independent variables A dependent variable is the variable hypothesized to depend on or caused by another variable, the independent variable. The independent [also called the treatment, causal, or manipulated variable] is the intended cause of the dependent variable [also called the effect, outcome or criterion variable].
Cause [independent], effect [dependent] Types of Measuring Instruments Cognitive Tests - Achievement, Aptitude tests Affective Tests Interest inventories Attitude scales Semantic differential scales Rating scales Thurstone & Guttman scales Criteria for Good Measuring Instruments Validity - quality of data-gathering instrument of procedure that enables it to measure what it is suppose to measure Content validity Item validity Sampling validity Face validity Construct validity Reliability
is the degree of consistency that the instrument, whatever it is measuring, it does it consistently. Stability Equivalence Internal Consistency Reliability Revising and Improving
A pilot study Reliability Administrative Factors analysis Item analysis
Procedures Describes all the steps that will be followed in conducting the study from beginning to end, in order in which it will occur. The techniques to select participant; the administrations [when and how]; what going to occur An assumption A limitation Execute exactly Data Analysis
Statistical method Statistical techniques Interpretive
Pemungutan Data Kualitatif Data Kualitatif
Face to face and explain; earn trust
Feild notes
Sources : observation, interviews, phone calls, personal and official documents, photo, recordings, drawings, e-mail, informal conversations, Nonparticipant data collection
Threats
Observer bias Observer effect
Analyze
To systematically search, categorize, integrate, and interpret the data collected in a study
To narrow the focus Think and make hunches Describing what’s in the data Interpretation involves making sense of what the data mean The researcher has the key role
5 Processes of data analysis
Data managing Reading/memoing Describing Classifying Interpreting Not have to be sequentially
Cyclical, iterative process of reviewing data for commons topics or themes
The focus is on context, events, participants, describing their perspectives
Classifying small pieces into more general categories
Relies on the skills of the researcher
Results in pyramid
Interpretation based on the connections, common sense, and linkages among the data pieces, categories and patterns Computer programs are available
Guidelines
The credibility of the link between the topic studied and the data used to examine the topic
The description of the methods used to collect, analyze, and interpret the data
Expressing researcher and participant biases Checking data quality
Writing the Report
Differ with the type of report needed (dissertation, speech, journal, article,.)
Analysis and interpretation also go on during the writing. Testing the meaningfulness of ideas and logic.
Focus on key themes and interpretations in the data, not on every …. Language should be straight-foward First-person voice is accepted Often more like a story than a formal report.
.
Data Analysis What is Statistics? A body of mathematical techniques or processes for gathering, analyzing, and interpreting numerical [quantitative] data. Numerical data gathered; by means of numbers. A basic tool of measurement, evaluation and research. Statistical data describe group behavior or group characteristics abstracted from a number of individual observations that are combined to make generalizations possible. Statistical measurement is an abstraction that may be used in place of great mass of individual measures. Takes facts and translates into a numerical form of expression, and make those facts more meaningful.
Statistical method [techniques] serves the fundamental purposes of description and analysis, answering the following questions: What facts needed to be gathered to provide the information necessary to answer the question or to test the hypotheses? How are these data to be selected, gathered, organized, and analyzed? What assumptions underlie the statistical methodology to be employed? What conclusions can be validly drawn from the analysis of data? Data Characteristics Discrete Data – exist independently of each other; examples include individuals, bacteria, apples, nationalities Continuous data – together from a continum; examples include millibars, degrees of temperature, chronological age Role of Statistics in Making Data Meaningful Attempt to comprehend the facts of the real world with the aid of numbers: the levels of intelligence; the strength of believes; one’s academic achievement; skill in sports; Abstractions Add to these points of central tendency, extent of dispersion, degree of relationships of one factor to another; and the testing of hypotheses; Scale of Measurement of Data Nominal data – by assigning them a name; examples include pine tree, farmers Ordinal data – assigned by order of sequence; examples days of week, faculty rank in a university, percentile scale Interval data – measured in terms of difference in standard units, examples include A is 3 cm taller than B Ratio data – indicate that one item is so many times larger as, as bright as, as powerful as another; examples a percentage scale, electrical current The Realm of Statistics Parametric Statistics and Nonparametric Statistics Techniques Parametric Statistics Descriptive statistics – the measures of central tendency; variation; correlation Inferential Statistics – involving inferences, estimations, predictions, hypothesis testing Parametric and Nonparametric Data Parameter – a function, a characteristics, a quality of population that in concept is constant, but value is variable. Parametric – are measured data, and parametric statistical tests assume that the data are normally, or nearly normally, distributed. Applied to both interval and ratio scaled data.
Nonparametric – are either counted [nominal] or ranked [ordinal]. Sometimes known as distribution-free tests, do not rest on the more stringent assumption of normally distributed populations.
Parametric : Descriptive and Inferential Analysis Descriptive : where the center is; how broadly they spread; and how they related in terms of one aspect to another aspect of the same data. Inferential : to extrapolate beyond the known into the realms of the unknown. Descriptive Statistics: Points of Central Tendency Measures of Central Tendency or Averages - mean, mode, median Mean – the precise center of the amalgamated values Median – the precise center of the numerical array Mode – the value which appears most frequently *Normal Distribution; The Normal Curve Skewed Distributions Symbols 2. Measures of Spread or Dispersion; Deviation - range, variance, standard deviation Correlation – the existence of relationships of different types of data; a decimal fraction indicating relatedness factors is called a coefficient correlation; Pearson’s Product-moment [r], rank order correlation [p]. Relative position: percentile, z-scores, t-scores Inferential Data Analysis Two Principal Functions: To predict or estimate a population parameter from a random sample; and To test statistically based hypotheses Testing Statistical Significance Level of significance The null hypothesis Two-tailed of significance ANOVA t-Test Post-hoc ANCOVA and Partial Correlation Multiple Regression and Correlation Nonparametric Tests Data sometimes look more like a stairway than a normal curve – data ranked Non-normal curve The Chi Square Test – used in casual comparative studies
Spearman Rho – counterpart of Pearson product The Mann-Whitney Test – counterpart of the t-Test Etc.