Ettema
Tuesday, September 30, 2003
Writing Simple Research Hypotheses
Guideline 1.1
A simple research hypothesis should name two variables and indicate the type of relationship expected between them.
Guideline 1.2
When a relationship is expected only among a certain type of subject, the populations should be mentioned in the hypothesis
Guideline 1.3
A simple hypothisis should be as specific as possible yet expressed in a single sentence.
Guideline 1.4
If a comparison is to be made, the elements to be compared should be stated.
Guideline 1.5
Because most hypotheses deal with the behavior of groups, plural forms should usually be used.
Guideline 1.6
A hypothesis should be free of terms and phrases that do not add to its meaning.
Guideline 1.7
A hypothesis should indicate what will actually be studied – not the possible implications of the study or value judgments of the author.
Guideline 1.8
A hypothesis should usually name variables in the order in which they occur or will be measured.
Guideline 1.9
Avoid using the words “significant” or “significance” in a hypothesis.
Guideline 1.10
Avoid using the word “prove” in a hypothesis.
Guideline 1.11
Avoid using more than one term to refer to a given variable
CHAPTER 7: THE LOGIC OF SAMPLING
OUTLINE
I. A Brief History of Sampling
II. Nonprobability Sampling
"Any technique in which samples are selected in some way not suggested by probability theory"
A. Reliance on available subjects
B. Purposive or judgmental sampling - selecting on the basis of ones own judgement
C. Snowball sampling - each person interviewed is asked to suggest additional interviewees.
D. Quota sampling - units are selected into a sample based on prespecified characteristics so that the total sample will have the same characteristics as the population studied.
E. Selecting informants
III. The Theory and Logic of Probability Sampling
A. Conscious and unconscious sampling bias
B. Representativeness and probability of selection
C. Random selection - A sampling method in which each elemnt has an equal chance of selection independent of any other event in the selection process.
D. Probability theory, sampling distributions, and estimates of
sampling error
1. The sampling distribution of ten cases
2. Sampling distributions and estimates of sampling error
3. Confidence levels and confidence intervals
IV. Populations and Sampling Frames
V. Types of Sampling Designs
A. Simple random sampling (SRS) - units are assigned numbers, then a random number set is generated to select the sample. Seldom used.
B. Systematic sampling - Every k-th unit is selected for the sample
C. Stratified sampling - The grouping of units into homogenous groups (strata) before sampling. Improves the representativeness of a sample in terms of the stratification variables.
D. Implicit stratification in systematic sampling
E. An illustration: sampling university students
1. Study population and sampling frame
2. Stratification
3. Sample selection
4. Sample modification
VI. Multistage Cluster Sampling
A. Multistage designs and sampling error
B. Stratification in multistage cluster sampling
C. Probability proportionate to size (PPS) sampling
D. Disproportionate sampling and weighting
VII. Probability sampling in review
CHAPTER 9: SURVEY RESEARCH
OUTLINE
I. Topics Appropriate to Survey Research
II. Guidelines for Asking Questions
A. Choose appropriate question forms
B. Make items clear
C. Avoid double-barreled questions
D. Respondents must be competent to answer
E. Respondents must be willing to answer
F. Questions should be relevant
G. Short items are best
H. Avoid negative items
I. Avoid biased items and terms
III. Questionnaire Construction
A. General questionnaire format
B. Formats for respondents
C. Contingency questions
D. Matrix questions
E. Ordering items in a questionnaire
F. Questionnaire instructions
G. Pretesting the questionnaire
H. A composite illustration
IV. Self-administered Questionnaires
A. Mail distribution and return
B. Monitoring returns
C. Follow-up mailings
D. Acceptable response rates
E. A case study
V. Interview Surveys
A. The role of the survey interviewer
B. General guidelines for survey interviewing
1. Appearance and demeanor
2. Familiarity with questionnaire
3. Follow question wording exactly
4. Record responses exactly
5. Probing for responses
C. Coordination and control
VI. Telephone Surveys
A. Pluses and minuses
B. Computer-assisted telephone interviewing (CATI)
VII. New Technologies and Survey Research
VIII. Comparison of the Different Survey Methods
IX. Strengths and Weaknesses of Survey Research
X. Secondary Analysis
A. Shared data
B. Data archives
CHAPTER 6: Indexes, Scales, and Typologies
I. Indexes versus Scales
Both scales and indexes measure variables, both rank-order the units of analysis in terms of specific variables, and are composite measures, but there are differences:
Index: A type of composite measure that summarizes and rank-orders several observations and represents some more general dimension, constructed by accumulating scores assigned to individual attributes.
Scale: A type of composite measure composed of several items that have a logical or empirical structure among them, e.g. Bogardus social distance, constructed by assigning scores to patterns of responses.
II. Index Construction
A. Item selection
1. Face validity - items should appear on its face to be that item
2. Unidimensionality - items should represent one dimension
3. General or specific
4. Variance - items provides characteristic distinctions
B. Examination of empirical relationships
1. Bivariate relationships - relationship between two variables
2. Multivariate relationships among items - analysis of simultaneous relationships among several variables
C. Index scoring
2 decisions:
1) Decide the desirable range of index scores
2) Decide whether to give items in the index equal or different weights
D. Handling missing data
Methods to Handle Missing Data:
- If relatively few case of missing data, they may be excluded
- You may have grounds for treating missing data as one of the available responses
- Analysis of missing data may yield an interpretation of their meaing
- If the item has several possible values, you may assing the middle value to missing data
E. Index validation
1. Item analysis - examination of the extent to which the index is related to the individual items that it comprises.
2. External validation - the process of testing the validity of a measure by examining its relationship to other presumed indicators of the same variable.
3. Bad index versus bad validators
F. The status of women: an illustration of index construction
III. Scale Construction
A. Bogardus social distance scale - A measurement technique for determining the willingness to participate socially with other kinds of people.
B. Thurstone scales
C. Likert scaling
D. Semantic differential
E. Guttman scaling
IV. Typologies
CHAPTER 5: Conceptualization, Operationalization, and Measurement
OUTLINE
I. Measuring Anything That Exists
A. Conceptions, concepts and reality
B. Concepts as constructs - 3 classes of things scientists measure: 1) Direct Observables 2) Indirect Observables - a checkmark next to "male" is an indirect observation of gender 3) Constructs - e.g. IQ is constructed mathematically from observations of answers given to questions, can't be observed.
II. Conceptualization
A. Indicators and dimensions
B. The interchangeability of indicators
C. Real, nominal, and operational definitions
D. Creating conceptual order
E. An example of conceptualization--anomie
III. Definitions in Descriptive and Explanatory Studies
IV. Operationalization Choices
A. Range of variation
B. Variations between extremes
C. A note on dimensions
D. Defining variables and attributes
E. Levels of measurement
1. Nominal measures
2. Ordinal measures
3. Interval measures
4. Ratio measures
5. Implications of levels of measurement
F. Single or multiple indicators
G. Some illustrations of operationalization choices
H. Operationalization goes on and on
V. Criteria of Measurement Quality
A. Precision and accuracy
B. Reliability
1. Test-retest method
2. Split-half method
3. Using established measures
4. Reliability of research-workers
C. Validity
1. Face validity
2. Criterion-related validity
3. Construct validity
4. Content validity
D. Who decides what’s valid?
E. Tension between reliability and validity
CHAPTER 4: Research Design
I. Three Purposes of Research
A. Exploration - 3 purposes: 1) Satify a researchers curiosity or desire for understanding 2) Test the feasibility of underaking a more extensive study 3) Develop methods to be employed in subsequent studies
B. Description - e.g. the Census
C. Explanation - Answer "why?"
II. The Logic of Nomothetic Explanation
A. Criteria for Nomothetic Causality
1. Correlation - there is a relationship between the variables
2. Time Order - cause happens before the effect
3. Nonspurious - "eliminating all other possible explanations"
B. False Criteria for Nomothetic Causality
1. Complete causation - Probabalistic Causes are valid
2. Exceptional cases
3. Majority of cases - causal relationships can be true even if they don't apply to a majority of cases.
III. Necessary and Sufficient Causes
IV. Units of Analysis
A. Types
1. Individuals
2. Groups
3. Organizations
4. Social artifacts - any product of social beings or their behavior
B. Faulty Reasoning About Units of Analysis: The ecological fallacy and reductionism
V. The Time Dimension
A. Cross-sectional studies
B. Longitudinal studies
1. Trend - examines changes within a population over time
2. Cohort - examines changes in subpopulations over time
3. Panel - data collected from the same set of people at several points in time
Comparing the three types of longitudinal studies
C. Approximating longitudinal studies
VI. How to Design a Research Project
A. Getting started
B. Conceptualization - The menta process by which fuzzy and imprecise concepts are made precise.
C. Choice of research method
D. Operationalization - Specifying the exact operations involved in measuring a variable
E. Population and sampling
F. Observations
G. Data processing
H. Analysis
I. Application
J. Research design in review
VII. The Research Proposal
A. Elements of a Research Proposal
1. Problem or Objective
2. Literature Review
3. Subjects for Study
4. Measurement
5. Data-Collection Methods
6. Analysis
7. Schedule
8. Budget
CHAPTER 1: HUMAN INQUIRY AND SCIENCE
I. Introduction
II. Looking For Reality
A. Ordinary human inquiry
B. Tradition
C. Authority
D. Errors in inquiry and some solutions
1. Inaccurate observations
2. Overgeneralization - a few similar events can be mistaken for evidence of a general pattern - replication, or repeating a study, is used by researchers to guard against overgeneralization.
3. Selective observation - once a pattern has been identified, the tendency to focus on future events and situations that fit a general pattern
4. Illogical reasoning
E. What's really real?em>
1. The premodern view - what one sees is reality, no diversity
2. The modern view - philosophical "different strokes for different folks" diversity of views is accepted
3. The postmodern view - "there's nothing out there, it's all in here" view, all that is real is are the images we get through our points of view.
III. The Foundations of Social Science
A. Theory, not philosophy or belief - What is, not what should be.
B. Social regularities
1. The charge of triviality - "common wisdom", "obvious truths"
2. What about exceptions? - they don't discount social regularities
3. People could interfere - the possibility does not challenge social science
C. Aggregates, not individuals - individuals are rarely the subject of study
D. A variable language
IV. Some Dialectics of Social Research
A. Idiographic and nomothetic explanation
B. Inductive and deductive theory
C. Qualitative and quantitative data
D. Pure and applied research
V. The Ethics of Social Research
A. Voluntary participation
B. No harm to subjects
