mixed method a brief history - fed.cuhk.edu.hk

Post on 09-Feb-2022

6 Views

Category:

Documents

0 Downloads

Preview:

Click to see full reader

TRANSCRIPT

2008/10/10

1

Mixed methodsEDM 6402

Qualitative Method in Educational Research

Lai Man Hong

Department of Educational Administration and Policy, CUHK

HTB 302, 26096454

mhlai@cuhk.edu.hk

Research questions

• Who should organized and pay for the education and training of working adults?

• Who participate in adult education and training programs, and why do they participate?

• What are the strengths and limitations of different types of education and training that purport to improve worker competencies?

Mixed Method

• Mixed method studies are those that combine the qualitative and quantitative

approaches into the research methodology

of a single study (Tashakkori & Teddlie, 1998)

A brief history

2008/10/10

2

• Third methodological movement (Tashakkori & Teddlie, 2003)

• adolescence

• The ultimate goal of any research project is to answer the questions that were set

forth at the project’s beginning.

• Because social phenomena are so complex, different kinds of methods are needed to best understand these complexities (greene & Caracelli, 1997)

• Mixed methods are becoming increasingly popular, especially because they are often more efficient in answering research questions than either the qualitative or the quantitative approach alone (Tashakkori & Teddlie, 1998)

• Methods should be mixed in a way that has complementary strengths (Johnson & Turner, 2003)

• Mixed methods provide the opportunity for presenting a greater diversity of divergent views.

• Mixed methods research provides stronger inferences

• A major advantages of mixed methods research is that it enables the researcher to simultaneously answer confirmatory and exploratory questions, and therefore verify and generate theory in the same study.

• Whether mixed methods, as compared to other designs, best addresses the research problem?

• Are both qualitative and quantitative research questions raised?

• Are conclusions drawn from both the qualitative and quantitative data?

Example

• Steven (2002)

• Hypothesis: teachers in schools with distinguished educator (DE) would perform better on measures of teacher effectiveness than would teachers in school without the DE

• How this had occurred?

2008/10/10

3

Title

• Adolescent development and transition to motherhood: a mixed methods study

Major types of mixed methods

research• Triangulation design

QUAN QUAL

QUAN + QUAL

• Embedded design

qual

QUAN

QUAN (qual)

quan

QUALQUAL (quan)

• Explanatory design

QUAN qual QUAN qual

• Exploratory design

qaul QUAN qual QUAN

Choosing a research design

• Theoretical drive

• Orientation (dominant, less dominant)

• Sequence

• Researchers should carefully select a single design that best matches the research problem. This will make the study more manageable and simpler to implement and describe.

2008/10/10

4

Choosing a research design

• A primary consideration is that the design should

match the research problem

• Researchers should evaluate their own

expertise

• Consideration must also be given to the

available resources (time, funding)

Choosing a research design

• Timing

– The sequence of data collection

• Weighting

– The relative weight of the quantitative and

qualitative approaches

• Mixing

– The approach to mixing the two datasets

• qual → QUAN

– Sequential exploratory design

• QUAN → qual

• quan → QUAL

– Sequential explanatory design

Sampling

• Probability sample (QUANs) ------Purposive samples (QUALs)

• The sampling strategy should stem logically from the conceptual framework as well as from the research questions being addressed by the study

• The sample should be able to generate a thorough database on the type of phenomena under study

• The sample should at least allow the possibility of drawing clear inferences from the data, the sample should allow for credible explanation

• the sampling plan should be feasible

• 61 firms, 8700 employees

• Stratified random sampling– Manufacturing and service sectors

– Major industries and trades

– Firms with different types of ownership– Firms of different sizes ranging from large to medium to small

– One or two intact work groups or production lines on the floor for a vertical sample consisting of all personnel at a work site

• Purposive sampling:

– Three enterprise with different level of technological change

– Workers with different job training experiences

2008/10/10

5

• In sequential mixed methods studies,

information from the first sample is often

required so as to draw the second sample.

• A probability sample is drawn first to test

hypotheses based on the extant literature. Then,

a purposive sample is selected to assess

research questions that go beyond what is

known in the current literature.

Making inferences

• Convergence – Life course pattern of males

– Their views, subjective perceptions, personal stock taking over their occupational lives.

– The theoretical statements were first tested deductively through quantitative empirical data.

– A second empirical study using qualitative methods to provided additional evidence for the theoretical hypothesis and to validate the assumption.

Making inferences

• Complementary– The level of school education

– Sociodemographic status of father

– Investment in human capital? / home atmosphere?

– Cultural capital / class exclusion in school?

– It can be meaningfully applied in all cases where one single research method does not suffice to collect adequate and enough empirical data to support the initial theoretical assumption

– In such cases, quantitative and qualitative methods serve different purposes and help to illuminate different aspects of the sociological phenomenon under study.

– Qualitative and quantitative results are not interchangeable.

Making inferences

• Complementary– Many young people were successful in their job

– They were highly criminal during their leisure time

– The empirical data analyzed by statistical means provided only limited evidence

– It served as a starting point for a second empirical inquiry

– It was possible to show that initially assumed relation among employment status, delinquency, and severity of punishment was far more complex than expected – an insight that would not have been acquired using one single method.

2008/10/10

6

Making inferences

• Divergence– Transition from schooling to work in East Germany– Quan: official allocation authority

– Qual: the role of individual job seeking

– The official allocation authority legitimizing the individual actions of graduates

– In many cases, outcomes of qualitative and quantitative research may at first sight even contradict each other

– The qualitative data thereby revealed that the simple and straightforward picture produced by quantitative data was incorrect and misleading

– “Double-speak”

• Divergence– It may be necessary to revise and modify the

initial theoretical assumptions

– To explain the inconsistent and divergent findings, it is then necessary to seek theoretical concepts and statements that can change the divergence of results to convergence or complementary

top related