Dr. Kanagaraj EaswaranAssociate Professor
Department of Social WorkMizoram University
Aizawl - 796004
Aspect Quantitative QualitativeData Numbers WordsPoint of View Researcher ParticipantsRelationship: Researcher andResearched
Distance Close
Role of Theory Testing EmergingDesign Static Process-
Emergent
Aspect Quantitative Qualitative
Questions Structure Unstructured
Outcome Generalisation Contextual
Understanding
Nature of Data Hard,
Reliable Data
Rich, deep data
Level Macro Micro
Focus of Research Behaviour Meaning
Setting Artificial Natural
Hardy and Bryman (2004)
Data Reduction:
Answering research questions:
Concerned with relating Results to
literature:
Concerned with variation:
Frequency as a springboard for analysis
Hardy and Bryman (2004)
Deliberate distortion not to occur: Importance of Transparency:Address the question of error:Research methods appropriate to research questions:
Employing Combination of quantitative and
qualitative methods
Perceived legitimacy of both methods
Multi method, convergence, integrated, and
combined
Data derived from mixed methods - mutually
illuminating not use of them in tandem
Campbell and Fisk(1959) used multi methods - validity of psychological traits
Encouraged others to employ multi methodStarted mix methods: qualitative data (field
methods viz., observation, interview combined with quantitative data(traditional surveys)(Sieber 1973)
Recognition: All methods have limitations – biases inherent in single method could neutralise or cancel the biases of other methods.
Triangulating data sources – a means for seeking convergence across qual quan methods (Jick 1979).
By early 1990s movement from convergence to integrating and connecting(Takhakkori and Teddlie 1998).
Research methodology evolves and
develops continuously
Step forward utilizing strengths of both
Inadequacy of application of quantitative
or qualitative methods alone to complexity
of social or human phenomena
Interdisciplinary nature of research –
research teams with diverse
methodological orientations
Expanded understanding greater insight
from combination
Embedded Method Argument : Research methods rooted in epistemological and ontological
commitments Methods not mere data gathering procedures but commitment Mixed methods not feasible or even desirable. Ignores underlying assumptions and transforms of qualitative
inquiry into a procedural variation of quantitative inquiry(Smith and Heshusius(1986).
Research Methods do not carry with them fixed epistemological and ontological implications – capable of used in a variety of ways.
Quantitative vs Qualitative Paradigms:
◦Two incompatible world views –
incommensurable
◦Integration superficial level and within
a single paradigm
Contentions about interconnections between methods and paradigms cannot be demonstrated.◦No means clear that quantitative and
qualitative approaches are paradigms. ◦There are overlaps and commonality
between them
Epistemological Version: Incompatible Mixed Methods is not possible
Technical Version: Strengths of both and capable of mixing Distinctive epistemological ontological assumptions
recognised Connection between assumptions and methods not viewed
as fixed and ineluctable. Methods are autonomous and hence compatible,
feasible and desirable
Alan Bryman (2006) Purposes or ways
1. Triangulation or Grater Validity: Mutual corroboration
2. Offset: Research methods have own strengths and
weaknesses – combining offsets – draw strengths of both.
3. Completeness: more comprehensive account
4. Process: Quan: Structure – Qual : Process Dynamics
5. Different Research Questions : Different methods
6. Explanation: one methods results are interpreted with the help of the other
7. Unexpected Results: Surprising results of one are explained by the other
8. Instrument Development: Development of Questionnaire or Scale – identify close ended answers.
9. Sampling: one approach is used to facilitate sampling of respondents or cases.
10. Credibility: both together enhance integrity of findings
11. Context: qualitative research providing contextual understanding coupled with externally valid findings of survey
12. Illustration: qualitative data to illustrate the findings of quantitative research: putting meat on the bones of dry quantitative findings.
13. Utility or Improving the Usefulness of Findings: combination is more useful to practitioners
14. Confirm or Discover: qualitative data to generate hypotheses and quantitative to test that
15. Diversity of Views: Combining researcher’s and participants views – uncovering relationships and revealing meanings among participants.
16. Enhancement or Building upon Quantitative/Qualitative Findings: Making more of or augmenting either quantitative or qualitative findings by gathering data
Aspect Quantitative Mixed Qualitative Methods/Design
Predetermined
Both Emerging
Questions Closed Both Open endedType of Data Performance,
Attitude , Observational
Multiple forms
from all possibiliti
es
Interview Observation Documents, Audio Visual
Analysis Statistical Both Text and Image
Interpretation
Statistical Across databases interpretation
Themes, Patterns Interpretation
Timing Weighting Mixing TheorizingConcurrentNo Sequence Equal Integrating ExplicitSequential :Qualitative First Qualitative Connecting
ImplicitSequential: Quantitative First
Quantitative Embedding
Quan Qual data are merged on one end of the continuum, kept separate on the other end Combined some way between these two extremes Connecting:
Quan Qual methods are connected: First phase data collection, analysis second phase data collection. Integrating
Data collected concurrently - the two data bases are merged Embedding:
One form of data main source – other supportive
Whether a larger, theoretical perspective guides the entire design.
Social Science theory or broad theoretical lens(gender, race, class etc.)
Explicitly mentioned or implicitly not mentioned
Explicit lens shapes questions and guides research process and drawing implications
Creswell et al.
1. Sequential Designs: One after another
2.Concurrent Designs: Together Simultaneously
1. Sequential Explanatory Design
2. Sequential Exploratory Design
3. Sequential Transformative Design
1. Qual – quan
2. Quan - qual
1. Concurrent Triangulation Design1. Data Results are compared2. Equal Weight to both
2. Concurrent Embedded Design1. One takes primary and another
supportive3. Concurrent Transformative Design
1. Emancipatory theory as lens2. QUAN + QUAL or 3. qual + QUAN
Questions Please
Thanks