Chapter 7 – The Nature of Quantitative Research
The process of quantitative research:
1. Theory
2. Hypothesis
3. Research Design
4. Devise measures of concepts
5. Select research site(s)
6. Select research subjects/respondents
7. Administer research instruments/collect data
8. Process data
9. Analyze data
10. Findings/conclusions
11. Write up findings/conclusions
1. Theory:
- Deductive approach to the relationship
o Deductive = Concept driven (look for theories/concepts)
- However, not a specification of hypothesis
- Instead, theory acts loosely as a set of concerns in relation to which the social researcher
collects data
2. Hypothesis:
- Likely to be found in experimental research
3. Research Design:
- Sometimes entail testing the hypothesis
- External validity of findings
- Researchers’ ability to impute causality to findings
4. Devise measures of concepts:
- Operationalization: Making concepts measurable, the operations by which a concept is
measured
5. Select research sites
- Appropriate setting for research
6. Select research subjects/respondents:
- Experimental research: subjects
- Survey research: respondents
7. Administration of the research instruments:
- Experimental research:
o Pre-testing subjects
o Manipulating independent variable
o Post-testing respondents
- Cross-sectional research (survey)
o Interviewing sample
o Questionnaire
, 8. Process Data:
- Once information has been collected it must be transformed into data
- Must be quantified (transforming it into numbers)
9. Analyze Data:
- Reduce the amount of data collected
- Test for relationships between variables
- Develop ways of presenting results
10. Findings
- Interpret the results of the analysis
- Connection between findings out of step 8
- Hypothesis: is it supported?
11. Writing
- Readers must be convinced that the research conclusions are important
- Findings are robust
- Significance and validity
Concepts and their Measurement
Concepts:
- Building block of theory
- Represent the point around which the social research is conducted
- Social class, job search method, health, lifestyle, emotional satisfaction etc.
Each concept represents a label that we give to elements of the social world
Concepts in quantitative research:
- Measurable concept
- Once measured: independent (manipulated) or dependent (observed) variable
Reasons to measure:
1. Delineate fine differences between people (characteristics)
- Finer distinctions are more difficult to recognize and detect
2. It gives us a consistent device for those distinctions
- Consistent instrument
o Ability to be consistent over time
o Ability to be consistent with other researchers
- Measure should be something that is not influenced by time or researcher
- Measurements are bound to be influenced by social change, not time
- Consistent results: reliability
3. More precise estimates (schatting) of the degree of relationship between concepts
- How closely are concepts related to each other?
Indicators:
- Operational definition:
o Procedures/operations used to measure/manipulate
, o Make sure concept is measured consistently
o Findings are reliable/replicable
Indicator: Allow the concept to be measured
- Already exists
- Indirect measure of a concept
Direct and Indirect Indicators:
- Their relationship to the concept for which they stand
- Data from survey = direct indicator of income
- Social class = indirect indicator of income
How indicators can be devised:
- Through a question
- Through a recording of individuals’ behavior
- Through official statistics
- Through an examination of mass media content (content analysis)
Likert-Scale:
- Multiple-indicator or multiple-item measure
- Set of attitudes relating to a particular area
- Measure intensity of feelings
- Series of statements (‘items’)
- Indicated level of agreement with statement (not questions)
- Five-point scale (strongly agree – strongly disagree)
- Items all relate to same object and should be interrelated
Potential problems with a single indicator
- Might incorrectly classify many individuals
- One indicator may capture only a portion of the concept/be too general
- Multiple-indicator makes finer distinctions
Dimension of concepts:
- Concept might comprises (omvatten) different dimensions
- Different aspects/components of the concept should be considered
- People scoring high on one dimension may not score high on other
You need a multidimensional profile for each respondent
Reliability and Validity
Reliability:
- Consistency of measures
- Does the measurement/test produce the same results when repeated under consistent
conditions?
3 different types of reliability:
1. Stability:
- Test-retest method
- Correlation: The strength of the relationship between 2 variables
- Whether a measure is stable over time
Problems with this approach:
The process of quantitative research:
1. Theory
2. Hypothesis
3. Research Design
4. Devise measures of concepts
5. Select research site(s)
6. Select research subjects/respondents
7. Administer research instruments/collect data
8. Process data
9. Analyze data
10. Findings/conclusions
11. Write up findings/conclusions
1. Theory:
- Deductive approach to the relationship
o Deductive = Concept driven (look for theories/concepts)
- However, not a specification of hypothesis
- Instead, theory acts loosely as a set of concerns in relation to which the social researcher
collects data
2. Hypothesis:
- Likely to be found in experimental research
3. Research Design:
- Sometimes entail testing the hypothesis
- External validity of findings
- Researchers’ ability to impute causality to findings
4. Devise measures of concepts:
- Operationalization: Making concepts measurable, the operations by which a concept is
measured
5. Select research sites
- Appropriate setting for research
6. Select research subjects/respondents:
- Experimental research: subjects
- Survey research: respondents
7. Administration of the research instruments:
- Experimental research:
o Pre-testing subjects
o Manipulating independent variable
o Post-testing respondents
- Cross-sectional research (survey)
o Interviewing sample
o Questionnaire
, 8. Process Data:
- Once information has been collected it must be transformed into data
- Must be quantified (transforming it into numbers)
9. Analyze Data:
- Reduce the amount of data collected
- Test for relationships between variables
- Develop ways of presenting results
10. Findings
- Interpret the results of the analysis
- Connection between findings out of step 8
- Hypothesis: is it supported?
11. Writing
- Readers must be convinced that the research conclusions are important
- Findings are robust
- Significance and validity
Concepts and their Measurement
Concepts:
- Building block of theory
- Represent the point around which the social research is conducted
- Social class, job search method, health, lifestyle, emotional satisfaction etc.
Each concept represents a label that we give to elements of the social world
Concepts in quantitative research:
- Measurable concept
- Once measured: independent (manipulated) or dependent (observed) variable
Reasons to measure:
1. Delineate fine differences between people (characteristics)
- Finer distinctions are more difficult to recognize and detect
2. It gives us a consistent device for those distinctions
- Consistent instrument
o Ability to be consistent over time
o Ability to be consistent with other researchers
- Measure should be something that is not influenced by time or researcher
- Measurements are bound to be influenced by social change, not time
- Consistent results: reliability
3. More precise estimates (schatting) of the degree of relationship between concepts
- How closely are concepts related to each other?
Indicators:
- Operational definition:
o Procedures/operations used to measure/manipulate
, o Make sure concept is measured consistently
o Findings are reliable/replicable
Indicator: Allow the concept to be measured
- Already exists
- Indirect measure of a concept
Direct and Indirect Indicators:
- Their relationship to the concept for which they stand
- Data from survey = direct indicator of income
- Social class = indirect indicator of income
How indicators can be devised:
- Through a question
- Through a recording of individuals’ behavior
- Through official statistics
- Through an examination of mass media content (content analysis)
Likert-Scale:
- Multiple-indicator or multiple-item measure
- Set of attitudes relating to a particular area
- Measure intensity of feelings
- Series of statements (‘items’)
- Indicated level of agreement with statement (not questions)
- Five-point scale (strongly agree – strongly disagree)
- Items all relate to same object and should be interrelated
Potential problems with a single indicator
- Might incorrectly classify many individuals
- One indicator may capture only a portion of the concept/be too general
- Multiple-indicator makes finer distinctions
Dimension of concepts:
- Concept might comprises (omvatten) different dimensions
- Different aspects/components of the concept should be considered
- People scoring high on one dimension may not score high on other
You need a multidimensional profile for each respondent
Reliability and Validity
Reliability:
- Consistency of measures
- Does the measurement/test produce the same results when repeated under consistent
conditions?
3 different types of reliability:
1. Stability:
- Test-retest method
- Correlation: The strength of the relationship between 2 variables
- Whether a measure is stable over time
Problems with this approach:
