FACE TO FACE 2
FEM 3002 RESEARCH METHODOLOGY
rozumah baharudin, pem 3001 fem 3002, first semester 0405 2
VARIABLE AND MEASUREMENT
rozumah baharudin, pem 3001 fem 3002, first semester 0405 3
VARIABLES
Measurable characteristics or properties of people or things that can take on different values.
Vary + Able
rozumah baharudin, pem 3001 fem 3002, first semester 0405 4
Variable
• Variables are what is studied by researchers.
• It have several types: 2 IMPORTANT TYPES
• Dependent• Independent
rozumah baharudin, pem 3001 fem 3002, first semester 0405 5
Dependent Variables
Indicates whether the treatment or manipulation of the independent variable had an effect – Synonym
• Outcome variable• Result variable• Effect • Criterion variable
rozumah baharudin, pem 3001 fem 3002, first semester 0405 6
Independent Variables
A variable that is manipulated to examine it’s impact on a dependent variable or outcome variableTreatmentSynonym
•Factor •Predictor
• The dependent variable is placed on the y-axis
• The independent variable is placed on the x-axis.
Antecedent variable
• An antecedent variable is a variable that occurs before the independent variable and the dependent variable.
Academic achievement
ParentingBehavior
Parental Characteristics• Age• Education • Self-efficacy
Family Contexts• # of children• Family income• Parental Marital Q
Quality
Child Characteristics• Age• Sex• Aspiration
Conceptual Framework for a study on “Predictors of Parenting Behavior and Child Academic Achievement
AV
IV DV
Types of variables based on Adjectives
• Quantitative Variables– Discrete Variables– Continuous Variables
• Qualitative or Categorical Variables
• A variable that can be measured numerically is called a quantitative variable. The data collected on a quantitative variable are called quantitative data.
• A variable whose values are countable is called a discrete variable. In other words, a discrete variable can assume only certain values with no intermediate values.
• Example: A household could have:– three children or six children, but not 4.53
children. – two or three cars, but not 2.5 cars.
• A variable that can assume any numerical value over a certain interval or intervals is called a continuous variable.
• Example: A person can be:– 5.7 inches tall, & 80.1 kg in weight
• A variable that cannot assume a numerical value but can be classified into two or more nonnumeric categories is called a qualitative or categorical variable. The data collected on such a variable are called qualitative data.
Variable
Quantitative Qualitative orcategorical (e.g.,
make of a computer,hair color, gender)
Continuous(e.g., length,age, height,weight, time)
Discrete (e.g.,number of
houses, cars,accidents)
rozumah baharudin, pem 3001 fem 3002, first semester 0405 16
Definition of Variable:
• Conceptual• Operational
rozumah baharudin, pem 3001 fem 3002, first semester 0405 17
Conceptual definition
• Definition that explain the idea/concept the variable is trying to capture.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 18
Operational definition
• Definition of how the variable will be measured in practice.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 19
e.g.,
Variable = academic achievement Conceptual = performance of student in all the
courses taken since enrolled. Operational = The accumulative great point
average of the student.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 20
MEASUREMENT
• Procedure for assigning symbols, letters, or numbers to empirical properties of variables according to rules.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 21
• Difficulty in measuring concepts directly (e.g., academic achievement)
• Usually measure indicators of concepts (e.g., CGPA)
rozumah baharudin, pem 3001 fem 3002, first semester 0405 22
• Level of measurement determines the type of statistical analysis.
• The level of measurement you use depends on how you want to measure an outcome.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 23
Levels of Measurement
• Nominal• Ordinal• Interval • Ratio
rozumah baharudin, pem 3001 fem 3002, first semester 0405 24
Nominal• Latin word nomin (name)• Variable categorical in nature• Differ in quality not quantity (numbers have no meaning only
label)
• Characterizes observation into one (and only one) category mutually exclusive
• Solely qualitative• No obsolute zero (0)• Matematical operation not possible.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 25
Ordinal• Describes variables that can be ordered along some
type of continuum.• Not only categories, order as well.• Ranking according to various outcomes, e.g., big &
little.• No obsolute ‘0’, only relative position; e.g., Zul is
taller than Sheereen and Sheereen is taller than Rozumah (no information on how much taller).
• Matematical operation not possible.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 26
Interval• Latin word intervalum (spaces between walls).• Describes variables that have equal intervals btw them.• Allow us to determine the difference btw points along the
same type of continuum (e.g., the difference btw 300 and 400 is the same as the difference btw 700 and 800; i.e., 100).
• 0 is arbitrary (subjective, temporary).• Simple matematical operation.• More precise & convey > info than nominal & ordinal; but
must be cautious in interpreting.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 27
Ratio
• Latin word ratio (calculation).• Describes variables that have equal intervals
btw them & have absolute 0.• Most precise.• Complex matematical operation.• Highest level of measurement.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 28
Summary
• Nominal level variables are categorical in nature (lowest level)
• Ordinal -- are ranked.• Interval -- have equidistant points along some
underlying continuum.• Ratio -- have a true zero (highest level).
rozumah baharudin, pem 3001 fem 3002, first semester 0405 29
What is the level of these measurements of height?
• Precise height measured on a scale with true zero.• Tall and Short (have some meaning, but nature of
difference is not known).• A is 5 inches taller than B (know precise
difference).• Categorize people into A and B (people different
in height).
rozumah baharudin, pem 3001 fem 3002, first semester 0405 30
RELIABILITY & VALIDITY OF
MEASUREMENT
rozumah baharudin, pem 3001 fem 3002, first semester 0405 31
Reliability and validity are the hallmarks of good measurement.
Assessments tools must be reliable and valid, otherwise the research hypothesis may be incorrectly rejected.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 32
• Reliability is a practical measure of how consistent and stable a measurement instrument or a test might be.
• Reliability is often measured using the correlation coefficient.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 33
Synonyms for Reliability
• Dependency• Consistency• Stablility• Trustworthiness• Predictability• Faithfulness
rozumah baharudin, pem 3001 fem 3002, first semester 0405 34
Types of Reliability
1. Test-retest2. Parallel forms3. Inter-rater4. Internal consistency
rozumah baharudin, pem 3001 fem 3002, first semester 0405 35
1. Test-retest:
• A measure of stability.• Examines consistency over time.• Administer the same test/measure at two
different times to the same group of participants.
• Coefficient: rtest1.test2
rozumah baharudin, pem 3001 fem 3002, first semester 0405 36
2. Parallel Forms
• A measure of equivalence.• Examines consistency between forms.• Administer different forms of the same test to
the same group of participants.• Coefficient: rform1.form2
rozumah baharudin, pem 3001 fem 3002, first semester 0405 37
3. Inter-rater
• A measure of agreement.• Examines consistency across raters.• Have two raters, rate behaviors and determine
the amount of agreement between them.• Coefficient: % of agreement.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 38
4. Internal consistency
• A measure of consistently each item measures the same underlying construct.
• Examines reliability within a particular set of item.• Correlate performance on each item with overall
performance across participants.• Coefficient: Chronbach’s alpha
rozumah baharudin, pem 3001 fem 3002, first semester 0405 39
Validity
• Is the quality of a test doing what it is designed to do.
• The test or instrument you are using actually measures what you need to have measured.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 40
Synonyms for Validity
• Truthfulness,• Accuracy• Authenticity• Genuineness• soundness
rozumah baharudin, pem 3001 fem 3002, first semester 0405 41
Types of Validity
1. Content2. Criterion
i. Concurrentii. Predictive
3. Construct
rozumah baharudin, pem 3001 fem 3002, first semester 0405 42
Content Validity
• A measure of how well the items represent the entire universe of items
• Established by asking expert if the items assess what you want them to.
• History test = test items ask questions on history not Science.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 43
Criterion Validity
i. Concurrent validity A measure of how well a test estimates a
criterion.
Established by selecting a criterion and correlate scores on the test with scores on ther criterion in the present.
Good student = test result + reports by lecturers.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 44
ii. Predictive Validity A measure of how well a test predicts a criterion.
Select a criterion and correlates scores on the test with scores on the criterion in the future.
High merit on STPM/Diploma = Score high CGPA.
Pass driving test = Good driver.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 45
Construct Validity A measure of how well a test assesses some underlying
construct.
Assess the underlying construct on which the test is based and correlate these scores with the scores.
Theoretically and practically sound.
Intelligence test actually measures intelligence.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 46
Relationship between Reliability and Validity
• A test can be reliable without being valid but the reverse is not true.A test can be reliable, but not valid, but a test cannot be
valid without first being reliable.Reliablity is a necessary, but not sufficient, condition of validity.You are answering questions on simple addition, but we called it
spelling test! Obviously it is not a test on spelling lack of validity, does not affect reliability.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 47
POPULATION AND SAMPLE
rozumah baharudin, pem 3001 fem 3002, first semester 0405 48
POPULATION
• Definition
A group of potential participants to whom you want to generalize the results of a study.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 49
Generalize : The key to a successful study; because it is only the results that can be generalized from a sample to a population; that research results have meaning beyond the limited setting.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 50
Not generalize : The sample selected is not an accurate representation of the population.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 51
Important Research Terms:
rozumah baharudin, pem 3001 fem 3002, first semester 0405 52
Population vs. Census
• Population the a group of people or things you are interested in.
• Census is a measurement of all the units in the population
rozumah baharudin, pem 3001 fem 3002, first semester 0405 53
Population Parameter vs. Statistic
• PP = number that results from measuring all the units in the population.
• Statistic = number that results from measuring all the units in the sample; statistics from samples are used to estimate PP.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 54
Sampling Frame vs Unit of Analysis
• SF = specific data from which sample is drawn, e.g., a phone book.
• UA = type of object of interest, e.g., arsons, fire departments, firefighters.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 55
Sampling Frame
• Is a list or quasi list of the members of a population. • Resource used in the selection of a sample.• A sample’s representativeness depends directly on
the extent to which a sampling frame contains all the members of the total population that the sample is intented to represent.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 56
e.g., Sampling Frame
• The data for this research were obtained from a random sample of parents of children in the third grade in government primary schools in Selangor.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 57
SAMPLES
Definition :Sample is a subset of the population.
– Good sampling : include maximizing the degree to which this selected group represent the population.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 58
POPULATION
Sample
Sample
rozumah baharudin, pem 3001 fem 3002, first semester 0405 59
WHY SAMPLE?
rozumah baharudin, pem 3001 fem 3002, first semester 0405 60
Types of sampling
1. Probability sampling2. Non probability sampling
rozumah baharudin, pem 3001 fem 3002, first semester 0405 61
Probability sampling• Allows use of statistics, tests hypotheses.• Can estimate population parameter.• Eliminates bias.• Must have random selections of units.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 62
Non-probability sampling
• Exploratory research, generates hypotheses.• Population parameters not of interests.• Adequacy of sample unknown.• Cheaper, easier, quicker to carry out.• Cant generalized findings.• Non-representative.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 63
PROBABILITY SAMPLING
• A type of sampling where the likelihood of any one member of the population being selected is known.
• Commonly used because the selection of participants is determined by chance.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 64
–e.g., if there are 4,500 students in the Faculty of Human Ecology, and if there are 1,000 seniors, the odds of selecting one senior as part of the sample is 1000:4,500 or 0.22.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 65
NON-PROBABILITY
• Where the likelihood of selecting any one member from the population or where the probability of selecting a single individual is not known.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 66
–e.g., if you do not know how many seniors in the Faculty of Human Ecology, the likelihood of anyone being selected cannot be computed.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 67
TYPES OF PROBABILITY SAMPLING
1. Simple Random Sampling2. Systematic Sampling3. Stratified Random Sampling4. Cluster Sampling
rozumah baharudin, pem 3001 fem 3002, first semester 0405 68
1. Simple Random Sampling When the population’s members are similar to one another.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 69
Adv: • Ensures a high degree of
representativenessDisadv:
• Time consuming and tedious
rozumah baharudin, pem 3001 fem 3002, first semester 0405 70
Using Table of Random Numbers
• See Handout
rozumah baharudin, pem 3001 fem 3002, first semester 0405 71
2. Systematic Sampling
When the population’s members are similar to one another.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 72
Adv : • Ensures a high degree of
representativeness; no need to use a table of random numbers.
Disadv : • Less truly random than simple
random sampling
rozumah baharudin, pem 3001 fem 3002, first semester 0405 73
3. Stratified Random Sampling
When the population is heterogeneous in nature and contains several different groups.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 74
Adv : • Ensures a high degree of
representativeness of all the strata in the population.
Disadv : • Time consuming and tedious
rozumah baharudin, pem 3001 fem 3002, first semester 0405 75
Two Types of Stratified Random Sampling (SRM)
• Proportionate SRM• Non-Proportionate SRM
rozumah baharudin, pem 3001 fem 3002, first semester 0405 76
Proportionate SRM
• Sampel selected is in proportion to the size of each stratum in the population
rozumah baharudin, pem 3001 fem 3002, first semester 0405 77
example: PSRM
• Population = 100• Layer 1 = 40 males• Layer 2 = 60 females• For a sample size of 10, you will take 4 males +
6 females.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 78
Non-proportionate SRM
• Selection of sample is not according to size of stratum in the population
rozumah baharudin, pem 3001 fem 3002, first semester 0405 79
e.g., NPSRM
• Population = 100• Layer 1 = 40 males• Layer 2 = 60 females• For a sample size of 10, you will take 5
males + 5 females.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 80
4. Cluster SamplingWhen the population consist of units rather than individuals.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 81
Adv : • Easy and convenient
Disadv : • Possibility that members of units are
different from one another, decreasing the sampling’s effectiveness
rozumah baharudin, pem 3001 fem 3002, first semester 0405 82
TYPES OF NON-PROBABILITY SAMPLING
1. Convenience Sampling2. Quota sampling3. Purposive Sampling4. Snowball sampling
rozumah baharudin, pem 3001 fem 3002, first semester 0405 83
TYPES OF NON-PROBABILITY SAMPLING
1. Convenience SamplingWhen the sample is captive.– Adv :
• convenient and inexpensive– Disadv :
• results in questionable representativeness.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 84
2. Quota sampling
When strata are present, and stratified, sampling is not possible– Adv :
• Ensures some degree of representativeness of all the strata in the population
– Disadv : • Results in questionable representativeness
TYPES OF NON-PROBABILITY SAMPLING
rozumah baharudin, pem 3001 fem 3002, first semester 0405 85
3. Purposive Sampling
• Researcher uses own judgement in the selection of sample members
• Sometimes called a judgmental sample.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 86
4. Snowball sampling
A technique often used in rare populations; each subject interviewed is asked to identify others.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 87
SAMPLING ERROR
• Lack of fit between the sample and the population.
• The difference between the characteristics of the sample and the characteristics of the population from which the sample was selected.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 88
• Reducing sampling error is the major goal of any selection technique.
• Larger sample, lower sampling error.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 89
SAMPLE SIZE
• How big?• Depends on type of research design.• Desired confidence level of results.• Amount of accuracy wanted.• Characteristics of population of interest.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 90
SAMPLE SIZE
• Big enough to answer research question.• But not so big that the process of sampling
becomes uneconomical.
• Heterogeneous sample = bigger size• Homogeneous sample = smaller size
rozumah baharudin, pem 3001 fem 3002, first semester 0405 91
• General Rule of Thumb
30 participants/ respondents in each group.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 92
General guide for sample size:
1. Larger sample, smaller sampling error, better representativeness.
2. If using several subgroups, starts with large enough subjects to account for the eventual breaking down of subject groups.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 93
3. If mailing out surveys or questionnaires, increase sample size by 40-50% to account for lost mails or uncooperative subjects.
4. Big is good, but appropriate is better.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 94
DATA COLLECTION
• Gathering information about a situation, problem or phenomenon.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 95
Methods in Information Gathering:
1. Secondary Data Information required is already available &
need only be extracted.2. Primary Data
Information must be collected.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 96
Secondary sources
• DocumentsGovernment publicationsEarlier researchCensusPersonal records
rozumah baharudin, pem 3001 fem 3002, first semester 0405 97
Primary Sources
1. Observation Participant Non-participant
2. Interviewing Structured Unstructured
3. Questionnaire Mailed questionnaire Collective questionnaire
rozumah baharudin, pem 3001 fem 3002, first semester 0405 98
OBSERVATION
• Is a purposeful, systematic, and selective way of watching and listening to an interaction or phenomenon as it takes place.
• Appropriate in situations where full and/or accurate information cannot be elicited by questioning.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 99
Types of observation
1. Participant observation2. Non-participant observation
rozumah baharudin, pem 3001 fem 3002, first semester 0405 100
Participant observation
• Researcher participates in the activities of the group being observed in the same manner as its members, with or without knowing that they are being observed.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 101
Non-participant observation
• Researcher does not get involved in the activities of the group but remains a passive observer, watching, & listening to its activities and drawing conclusions from this.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 102
Problems with using observation:
• Respondent may be aware & change behavior.• Observer bias.• Interpretation btw observer inconsistent.• Possibility of incomplete observation and/or
recording.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 103
Situations for observation
1. Natural Does not intervene.
2. Controlled Introduce stimulus to observe reactions.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 104
Methods of recording observations:
• Narrative• Scales • Categorical recording• Recording on mechanical devices
rozumah baharudin, pem 3001 fem 3002, first semester 0405 105
Narrative
• Take brief notes first• Soon after makes detailed notes• Adv: provides deep insight into the
interaction.• Disadv: observer bias & incomplete recording.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 106
Scales
• Develop scale to rate interactions or phenomenon.
• Adv: quick, easy to record.• Disadv: does not provide in-depth information
about interaction.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 107
Categorical Recording
• Depend on classification develop by researcher; e.g. passive/active, etc.
• Adv: quick, easy to record.• Disadv: does not provide in-depth information
about interaction.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 108
Recording on Mechanical Devices:
• Observation recorded on a video tape and then analyzed.
• Adv: can watched it many times b4 making conclusion; can invite expert to view to make right conclusion.
• Disadv: respondent uncomfortable, or behave differently.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 109
INTERVIEW
• Person-to-person interaction with specific purpose.
• Most common method.• 2 types:
1. STRUCTURED 2. UNSTRUCTURED
rozumah baharudin, pem 3001 fem 3002, first semester 0405 110
Unstructured Interview
• Known as in-depth interview.• Use interview guide/framework; no specific
set questions.• + spontaneous questions.• Can be conducted in …….
1. One-to-one2. Group interview (focused group)
rozumah baharudin, pem 3001 fem 3002, first semester 0405 111
• Use for in-depth information.• Or when lack of information.• Flexibility on what to ask of a respondent;
elicit rich information.• Thus, sometimes used to contruct structured
instrument.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 112
• Disadv.: – No specific set question, comparability difficult.– Questions may keep changing; info at beginning
may be different from later.– Freedom may lead to interviewer bias.– More skill needed to use interview guide than
structured interview.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 113
Structured interview• Pre-determined set questions in interview schedule:
– Same wording– Same order of questions
• Interview schedule/research instrument: – Written list of questions– Open-ended/ closed– For use by interviewer– In person-to-person interaction (face-to-face, by
telephone, or by other electronic means)
rozumah baharudin, pem 3001 fem 3002, first semester 0405 114
• Adv: provides uniform info, which ensures comparability of data.
• Required fewer interviewing skills than unstructured interviewing.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 115
QUESTIONNAIRE
• Is a written list of questions; answer recorded by respondents.
• Respondent read the questions, interpret & write down answers him/herself.
• Different from interview, where interviewer asks qn & write respondents replies on interview schedule.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 116
Rules for questionnaire:• Questions must be clear & easy to understand.• Layout is easy to read, pleasant to the eye,
sequence of qn easy to follow.• Interactive style – as if someone talking to
respondent.• Sensitive qn – prefaced with statement of
explanation (use different font for preface to distinguish them from acual question).
rozumah baharudin, pem 3001 fem 3002, first semester 0405 117
Choose Interview Schedule or Questionnaire?
Depends on:• Nature of investigation
– Sensitive questions, questionnaire better.• Geographical distribution of study population
– Respondents scattered, use questionnaire – cheaper.• Type of study
– Illiterate, very young or very old, or handicapped – use interview schedule.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 118
Ways of Administering a Questionnaire
1. Mailed questionnaire• Send out to prospective rspdnt• Must have addresses• Prepaid self-address envelope• With covering letter (brief explanation of
study, indicate confidentiality & participation is voluntary, + other impt qn).
• A Major problem --- low response rate.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 119
2. Collective questionnaire• Captive audience (e.g., students in lecture
hall)• High response rate coz few will refuse.• Can explain purpose & importance of study
face-to-face + can clarify qn.• Quickest was of collecting data• Save money
rozumah baharudin, pem 3001 fem 3002, first semester 0405 120
3. Administration in public place•Approach & request participation of potential rspdnt
•More time consuming•Adv same as collective qnn.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 121
Questionnaire or Interview?
• Adv & Disadv of Questionnaire• Adv & Disadv of Interview
rozumah baharudin, pem 3001 fem 3002, first semester 0405 122
Adv & Disadv of Questionnaire• Adv:
Less expensiveGreater anonymity
• Disadv: Limited application (only for those who can read & write) Low response rate if mailed. Self-selecting bias (only those with good attitudes or
motivations will response; may not be representative of study population).
rozumah baharudin, pem 3001 fem 3002, first semester 0405 123
Spontaneous response not allowed for.Response to a question may be influenced by
response to other questions.Possible to consult others.A response cannot be supplemented with other
information.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 124
Adv & Disadv of Interview
Adv:• More appropriate for complex situations.• Useful for collecting in-depth information.• Information can be supplemented (from
observations of non-verbal reactions).• Questions can be explained.• Interviewing has a wider application.
– Any type of population – children, illiterate, young & old.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 125
Disadv:• Time-consuming & expensive.• Quality of interaction can influence quality of
data.• Quality of interviewer can influence quality of
data.• Quality of data vary when many interviewers are
used.• Researcher may introduce his/her bias (e.g., in
framing the question).• Interviewer may be biased (e.g., in the way of
questioning).
rozumah baharudin, pem 3001 fem 3002, first semester 0405 126
Forms of Questions
• Form & wording of questions may affect type & quality of information obtained.
• Types of question:
Open-endedClose-ender
rozumah baharudin, pem 3001 fem 3002, first semester 0405 127
Open-ended Question
• Possible responses are not given.• Respondent writes the answer (for
questionnaire)• Interviewer record the respondents’ answers
(verbatim or summary)• Useful for seeking opinions, attitudes or
perceptions.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 128
Closed-ended Question
• Possible answers given.• Respondent or interviewer tick the answer.• Useful for eliciting factual information
rozumah baharudin, pem 3001 fem 3002, first semester 0405 129
Adv & Disadv of Open-ended Question
Adv:• Provide in-depth & wealth of info.• Provide opportunity for respondent to express
their opinion, resulting in more variety of info.• Allow respondents to express themselves
freely; eliminate the possibility of investigator bias.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 130
Disadv:• Analysis more difficult (must do content
analysis in order to classify the data).• Some respondents may not be able to
express themselves, so information may be lost.
• Greater chance of interviewer bias.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 131
Adv & Disadv of Closed-ended Question
Adv:• Ready-made categories; help ensure info
needed is obtained.• Easy to analyse.Disadv:• Info lacks depth & variety.• Investigator bias – may list answer he/she is
interested in.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 132
• Given response could condition thinking of respondents
• May create tendency among respondents and interviewers to tick a category/ries without thinking through the issue.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 133
Considerations in formulating questions:
• Always use simple & everyday language.• Do not use ambiguous questions.• Do not ask double-barrelled questions.• Do not ask leading questions.• Do not ask questions that are based on
presumptions.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 134
FORMULATING QUESTIONS
• Is anyone in your family having ‘H1N1”?• Is difficult for you to be a father and a
husband?• Are you happy with your cafeteria?• How often and how much time do you spend
visiting your lecturer?• In your opinion, eating lemang with rendang
or peanut sauce is nice?
rozumah baharudin, pem 3001 fem 3002, first semester 0405 135
• Smoking is bad, isn’t it?• ‘Ponteng kuliah’ is bad, isn’t it?• How many cigarettes do you smoke in a day?• What contraceptives do you use?
rozumah baharudin, pem 3001 fem 3002, first semester 0405 136
COLLECTING DATA USING SECONDARY SOURCES
Sources of Data:• Government or semi-government publications• Earlier research• Personal records• Mass-media
rozumah baharudin, pem 3001 fem 3002, first semester 0405 137
Problems using secondary data
• Validity & reliability• Personal bias• Availability of data• Format
rozumah baharudin, pem 3001 fem 3002, first semester 0405 138
DATA ANALYSIS
rozumah baharudin, pem 3001 fem 3002, first semester 0405 139
DATA ANALYSIS
• Ways to use/organize/manipulate data in order to reach research conclusions.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 140
DATA ANALYSIS
1. UNIVARIATE ANALYSIS2. BIVARIATE ANALYSIS3. MULTIVARIATE ANALYSIS
rozumah baharudin, pem 3001 fem 3002, first semester 0405 141
UNIVARIATE ANALYSIS
• Is the examination of the distribution of cases on only one variable at a time.– Distributions– Central tendency– Dispersion
rozumah baharudin, pem 3001 fem 3002, first semester 0405 142
• The full original data usually difficult to interpret.
• Data reduction is the process of summarizing the original data to make them more manageable; while maintaning the original data as much as possible.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 143
PROCESSING DATA
1. EDITING DATA2. CODING DATA3. DEVELOPING A FRAME OF ANALYSIS4. ANALYSING DATA
rozumah baharudin, pem 3001 fem 3002, first semester 0405 144
Editing Data
• Data Cleaning• Checking the completed instruments; to
identify and minimise errors incompleteness inconsistencies misclassification etc. (illegible writing)
rozumah baharudin, pem 3001 fem 3002, first semester 0405 145
Coding Data
2 Considerations for Coding:– Measurement of a variable (scale?, structure –
open/closed ended?).
– Communication of findings about a variable (measurement scale?, type of statisitical procedures?) (e.g., Ratio scale – mean, mode, median)
rozumah baharudin, pem 3001 fem 3002, first semester 0405 146
Process for coding:
• For analysis using computer, data must be coded in numerical values.
• The coding of raw data involves 4 steps:– Developing a code book (master-code book)– Pre-testing the book– Coding the data; and – Verifying the coded data.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 147
Developing A Frame of Analysis
• Develop from beginning of research and evolve continuously to end.
• Frame of analysis:– Identify variable to analyse– Determine method to analyse– Determine cross-tabulations needed – Determine which variable to combine for constructing
major concepts or develop indices – Identify which variable for which statistical procedures
rozumah baharudin, pem 3001 fem 3002, first semester 0405 148
Analysing Data
1. UNIVARIATE ANALYSIS2. BIVARIATE ANALYSIS3. MULTIVARIATE ANALYSIS
rozumah baharudin, pem 3001 fem 3002, first semester 0405 149
UNIVARIATE ANALYSIS
• Is the examination of the distribution of cases on only one variable at a time.
• Distributions• Central tendency• Dispersion
• Can be generated thro’ Descriptive statistics in the SPSS.
• Purpose of univariate analysis is purely descriptive.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 150
Distributions
• Attribute of each each case under study in terms of the variable in question.
• Reporting marginals• E.g., how many respondents, what % of them
fall under a certain variable.500 of 1000 FEM students have CGPA = 3.5 &
above.50% of 1000 FEM students.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 151
Frequency Distribution
• Shows the number of cases having each of the attributes of a given variable.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 152
Central Tendency
• Reporting summary• In term of averages
– Mode (most frequent attribute)– Mean (arithmetic mean)– Median (middle attribute)
rozumah baharudin, pem 3001 fem 3002, first semester 0405 153
Which measure of Central Tendency to use?
Measure Level of Measurement
Examples
Mode Nominal Eye color, party affiliation
Median Ordinal Rank in class, birth order
Mean Interval & ratio Speed of response, age in years
rozumah baharudin, pem 3001 fem 3002, first semester 0405 154
Dispersion
• Spread of raw data/info of a variable.• Detailed information of distribution of a
variable.Range (simplest measure)PercentileStandard deviation (more sophisticated)
rozumah baharudin, pem 3001 fem 3002, first semester 0405 155
• Range: distance separating the highest from the lowest value.
(e.g., the respondents mean age is 22.75 with a range from 20 to 26).
rozumah baharudin, pem 3001 fem 3002, first semester 0405 156
Percentile
• A number or score indicating rank by telling what percentage of those being measured fell below that particular score.
• e.g., scored 75th percentile, means 75% of the other people scored below your score and 25% scored at or above your score.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 157
Standard Deviation
• Is a measure of the average amount the scores in a distribution deviate from average (mean) of the distribution.
• Observation near mean, small SD. Observation far from mean, large SD.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 158
BIVARIATE ANALYSIS
• Focuses on the relationships/association between two variables.
• Among the many measures of bivariate association are eta, gamma, lambda, Pearson’s r, Kendall’s tau, and Spearman’s rho.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 159
MULTIVARIATE ANALYSIS
• Is a method of analyzing the simultaneous relationships among several variables and may be used to understand the relationship between two variables more fully.
• e.g., multiple regression, factor analysis, path analysis, discriminant analysis.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 160
STATISTICAL ANALYSIS
1) Descriptive Statistics2) Inferential Statistics
rozumah baharudin, pem 3001 fem 3002, first semester 0405 161
What is Descriptive Statistics?
• A medium in describing data in manageable forms (dealing with collection, tabulation, and summarization of data so as to present meaningful information).
• Quantitative descriptions• Describe single variables• Describe the associations that connect one
variable with another
rozumah baharudin, pem 3001 fem 3002, first semester 0405 162
Descriptive Statistics
1. Data Reduction2. Measures of Association3. Regression Analysis
rozumah baharudin, pem 3001 fem 3002, first semester 0405 163
Descriptive Stat:Data Reduction
• Reduction of data from unmanageable details to manageable summaries.
• e.g., for 100 respondents you may get data on 100 different ages; these data can be summarize to manageable form by coding it into 3-4 categories.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 164
Descriptive Stat:Measures of Association
• Provides information on the nature and extent of the relationship between any two variables.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 165
• Measures of association for two nominal variables = Lambda,
• For two ordinal variables = Gamma,
• For two interval or ratio variables = Pearson’s product-moment correlation (r).
rozumah baharudin, pem 3001 fem 3002, first semester 0405 166
The Value of r
• 0.0 = no linear relationship btw the 2 variables• + 1.0 = Strong positive linear relationship; as X
increases in value, Y also increases and vice versa.• - 1.0 = Strong inverse linear relationship; as X
increases in value, Y decreases in value; as X decreases in value, Y increases in value.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 167
Positive or negative r?
rozumah baharudin, pem 3001 fem 3002, first semester 0405 168
Descriptive Stat: Regression Analysis
• Represents the relationships between variables in the form of equations, which can be used to predict the values of a dependent variable on the basis of values of one or more independent variables.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 169
• The basic regression equation – for a simple linear regression:
Y = a + bx + e
• Y = value estimated of the dependent variable• a = constant variable / alpha or intercept• b = slope, numerical value (multiplied by X, the
value of the independent variable)(beta coefficient).
• e = error
rozumah baharudin, pem 3001 fem 3002, first semester 0405 170
• Simple linear regression model does not sufficiently represent the complexity of social life.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 171
Multiple Regression
• A social phenomenon (DV) is normally affected simultaneously by several IVs.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 172
• Multiple regression equation:
Y = a + b1x1 + b2x2 + bi xi + e
• Y = value estimated of the DV• a = constant variable• X1 to Xi = predictors• b = slope (beta coefficient) for X• e = residual (error)
rozumah baharudin, pem 3001 fem 3002, first semester 0405 173
What is Inferential Statistics?
• Typically it involves drawing conclusions about a population from the study of a sample drawn from it.
• i.e., Generalizing your findings to a broader population group.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 174
POPULATION
Sample
Infer from sample (statistic) to population (parameter)
rozumah baharudin, pem 3001 fem 3002, first semester 0405 175
• Techniques that allow us to determine if hypothesis is supported, while considering sampling error hypothesis testing.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 176
• Inferential statistics can help us estimate or predict population parameter from sample statistics.
• Population value = parameter• Sample value = statistics
rozumah baharudin, pem 3001 fem 3002, first semester 0405 177
Inferential Statistics
• Inferential statistics are based on the assumption that population distributions of variables from which samples are selected are normal in shape (Normal Curve/Distribution).
rozumah baharudin, pem 3001 fem 3002, first semester 0405 178
Normal Curve
• Represents how variables are distributed.• Characteristics: Bell-shaped; unimodal,
symmmetric and asymptotic.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 179
Characteristics of Normal Curve:
• Unimodal = mean, median & mode same value.• Symmetrical = left & right halves of curve are
mirror images.• Asymptotic = tails of curve get closer to X axis,
but never touch it.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 180
• See diagram on normal curve.
• The area under the curve is very important in inferential statistics.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 181
• Accuracy of inference depends on representativeness of sample from population.
• Random selection– Equal chance for anyone to be selected makes
sample more representative
rozumah baharudin, pem 3001 fem 3002, first semester 0405 182
• Inferential statistics help researchers test hypotheses and answer research questions, and derive meaning from the results.
A result found to be statistically significant by testing the sample is assumed to also hold for the population from which the sample was drawn.
The ability to make such an inference is based on the principle of probability.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 183
– Researchers set the significance level of each statistical test they conduct.
– By using probability theory as a basis for their tests, researchers can assess how likely it is that the difference they find is real and not due to chance
rozumah baharudin, pem 3001 fem 3002, first semester 0405 184
Test of Significance
• What inferential statistics does best is allow decisions to be made about populations based on the information about samples.
• One of the most useful tools for doing this is a test of statistical significance
rozumah baharudin, pem 3001 fem 3002, first semester 0405 185
• Inferential statistics test the likelihood that the alternative (research) hypothesis (H1) is true and the null hypothesis (H0) is not.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 186
– In testing differences, the H1 would predict that differences would be found, while the H0
would predict no differences.
– By setting the significance level (generally at .05), the researcher has a criterion for making the following decision:
rozumah baharudin, pem 3001 fem 3002, first semester 0405 187
• If the .05 level is achieved (p is equal to or less than .05), then a researcher rejects the H0 and accepts the H1.
• If the .05 significance level is not achieved, then the H0 is retained.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 188
Convention for significance levels (alpha levels)
.05
.01.001
rozumah baharudin, pem 3001 fem 3002, first semester 0405 189
Alpha levels are often written as the “p-value”.
e.g., p =.05; p < .05; (p less than .05)p < .05 (p equal to or less than) (the chance of making 5 in 100 or 1 in 20 of
making an error)
rozumah baharudin, pem 3001 fem 3002, first semester 0405 190
Variance of statistical test = Degrees of Freedom
• Df are the way in which the scientific tradition accounts for variation due to error.– It specifies how many values vary within a
statistical test.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 191
– It specifies how many values vary within a statistical test
• Scientists recognizes that collecting data can never be error-free
• Each piece of data collected can vary, or carry error that we cannot account for
• By including df in statistical computations, scientists help to account for this error
rozumah baharudin, pem 3001 fem 3002, first semester 0405 192
Testing Hypothesis
• If reject H0 and conclude groups are really different, it doesn’t mean they’re different for the reason you hypothesized
• May be other reason
rozumah baharudin, pem 3001 fem 3002, first semester 0405 193
• Since H0 is based on sample means, not population means, there is a possibility of making an error or wrong decision in rejecting or failing to reject H0
• Type I error• Type II error
rozumah baharudin, pem 3001 fem 3002, first semester 0405 194
• Type I error – rejecting H0 when it was true (it sound have been accepted)
– If alpha = .05, then there’s a 5% chance of Type 1 error.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 195
• Type II error – accepting H0 when it should have been rejected
– If increase alpha, you will decrease the chance of Type II error
rozumah baharudin, pem 3001 fem 3002, first semester 0405 196
Choosing Appropriate Statistical Test of Difference
• One variable One-way chi-square• Two variables
( 1 IV with 2 levels; 1 DV) t-test• Two variables
( 1 IV with 2+ levels; 1 DV) ANOVA
rozumah baharudin, pem 3001 fem 3002, first semester 0405 197
• Three or more variables ANOVA
• See handouts for more other examples of inferential statistics
rozumah baharudin, pem 3001 fem 3002, first semester 0405 198
WRITING QUANTITATIVE REPORTS
Using the APA Style
rozumah baharudin, pem 3001 fem 3002, first semester 0405 199
9 Major Components
1. Title Page2. Abstract3. Introduction (Chapter 1)4. Review of the Literature (Chapter 2)5. Method (Chapter 3)6. Results (Chapter 4)7. Discussion (or Summary, Conclusion, &
Implications) (Chapter 5)8. Bibliography9. Apendices (letters, instruments)
rozumah baharudin, pem 3001 fem 3002, first semester 0405 200
1. Title Page
• Summarize the main topic• About 10 -12 words
PREDICTORS OF MATERNAL BEHAVIOR AND THEIR EFFECTS ON THE ACHIEVEMENT OF CHLDREN
Write in Top Heavy style
PREDICTORS OF MATERNAL BEHAVIOR AND THEIR EFFECTS ON THE ACHIEVEMENT OF CHLDREN
Bottom Heavy
rozumah baharudin, pem 3001 fem 3002, first semester 0405 201
2. Abstract
• Comprehensive summary• About 120 words• For manuscript submitted for review, typed on
separate page.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 202
3. Introduction
• Begin with current scenario, country data / statistics, what are the symptoms in the society that make you want to study the problem. Place the problem in the context of other research literature
• Statement of the Problem• Purpose of the Study (May incorporate under
Statement of Problem, check with your supervisor)• Research Objectives
rozumah baharudin, pem 3001 fem 3002, first semester 0405 203
• Theoretical Framework• Conceptual Model• Conceptual and Operational Definitions• Rationale for the Present Study (May include
under Statement of Problem, check with your supervisor)
rozumah baharudin, pem 3001 fem 3002, first semester 0405 204
4. Review of Literature
i. Inform reader about previous research conducted on the topic being research.
ii. Also reflect how knowledgeable writer is on the topic.iii. Review studies which have focused on the DV.iv. Indicates the theory (if any) on which the study is based;
critique and weigh studies as theory is built.v. Identify knowledge gap.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 205
i. Present review in logical and comprehensive manner. Organize with reference to the objectives of the study.
ii. Write a summary paragraph which identifies all the major variables found to influence or related to the DV. Add a statement to show how your research topic flows from or adds to the research reviewed.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 206
3. Method
• Location of Study• Sampel (number, selection, characteristics)• Measures (Instrumentation)• Procedure / Data Collection
rozumah baharudin, pem 3001 fem 3002, first semester 0405 207
4. Results
• Results of data analysis and statistical significance testing
• Include tables and figures.
rozumah baharudin, pem 3001 fem 3002, first semester 0405 208
5. Discussion
• Interpret and evaluate your results• State whether hypotheses were supported.• Answer basic questions
– what your study contribute?– how study helped to solve study problem?– what conclusion and theoretical implications can
be drawn from your study?)
Top Related