Upload
uchulu
View
213
Download
0
Embed Size (px)
Citation preview
8/9/2019 Mixed Approach
1/40
MIXEDMIXED METHODMETHOD
APPROACHAPPROACHby
Prof. Philip AE Serumaga-ZakeUNISA SBL
March 23, 2010
8/9/2019 Mixed Approach
2/40
QUALITATIVE APPROACHQUALITATIVE APPROACH
Refers to research that:
focus on phenomena that occur in natural settings real world
involve studying those phenomena in all their complexities (Leedy and Ormrod,
2010).
Qualitative researchers normally try to portray issues in their multifacetedforms.
Often formulate general research problems or ask general questions at the
beginning and so, cannot know the right methods to use, but with time, as they
come to learn more and more about the phenomenon, they tend to ask specific
questions, formulate specific hypotheses and know better specific methods to
apply.Qualitative researchers must be well trained in observation and interviewing
strategies, with a firm grasp of previous research
With qualitative research, there is no quick and easy answers like yes, no
answers.
8/9/2019 Mixed Approach
3/40
What is Qualitative Research?
Often uses multiple forms of data
Takes a great deal of time
The researcher should record any potentially useful data thoroughly,
accurately and systematically, using field notes, audiotapes, sketches,
photographs, etc.
Often researchers use non-random purposive samples for which key
informants give most of the information being sought.
Normally small non-random samples are used
The researcher/interviewer is an integral part of the investigationIt is subjective but the truth is being sought
It is holistic and contextual
Normally in-depth interviewing and observation are used to collect data
Uses inductive rather than deductive process to derive themes or patterns
Thematic categorization of data, narrative description and content analysis
including coding are used in data analysis and reporting the study resultsQualitative researchers aim to gather an in-depth understanding ofhuman
behaviourand the reasons for such behaviour.
Qualitative research aim to describe systematically observations of social
behaviourwith no preconceived hypotheses to be tested (Rubin & Babbie,
1993).
8/9/2019 Mixed Approach
4/40
Hypotheses emerge from the observation and interpretation of human
behaviour, leading to further observations and the generation of new
hypotheses for exploration.The interviewer is an integral part of the investigation making the
research subjective.
Qualitative research uses detailed descriptions from the perspective of the research
participants themselves as a means of examining specific issues and problems under
study.
Qualitative approaches have the advantages of flexibility, in-depth analysis,and the potential to observe a variety of aspects of a social situation
(Babbie, 1986).
By developing and using questions on the spot in (a face-to-face interview),
a qualitative researcher can gain a more in-depth understanding of the
respondent's beliefs, attitudes, or situation.
During the interview, observational data can be of particular value when a
respondent's body language runs counter to the verbal response given to an
interview question.
8/9/2019 Mixed Approach
5/40
Gilgun (1990) suggested these steps of doing a qualitative research:
1. Identification of the area under investigation
2. Literature review - Drawing on the theoretical and research literature, suchquestions may be formulated and organized in advance to address a specific
research topic.
3. Selection of variables or parameters of study
4. Collection of data
5. Comparison of patterns of first case with those of second case6. Development of working hypothesis as common patterns emerge across
interviews
7. Formulation of additional questions and modification of questions, based
on analysis
8. Continuation of theoretical sampling
9. Review of relevant literature when patterns appear to stabilize10.Linking of relevant literature to the empirically grounded hypotheses
11.Testing of theoretical formulations derived from preceding step
12.Revision of theoretical formulations as needed to fit empirical patterns in
each subsequent step.
8/9/2019 Mixed Approach
6/40
y Produces more in-depth, comprehensive information.
y Uses subjective information and participant
observation to describe the context, or natural setting,of the variables under consideration, as well as the
interactions of the different variables in the context.
y It seeks a wide and deep understanding of the entire
situation.
Disadvantages of qualitative research
y The very subjectivity of the inquiry leads to difficulties in establishing the
reliability and validity of the approaches and information.
y It is very difficult to prevent or detect researcher induced bias.
y Its scope is limited due to the in-depth, comprehensive data gathering
a roaches re uired.
The process ends when the researcher reaches theoretical saturation,
the point at which no new data are emerging (see also Glaser & Strauss,
1967). Through this procedure emerging theories are grounded in data and
are linked to other theories and research (Gilgun, 1992).
Advantages of qualitative research
8/9/2019 Mixed Approach
7/40
Methods ofData Collection
Methods include interviews and focus group discussions, observation(Participant Observation, Non-participant Observation), field notes, various
texts, pictures, and other materials.
Structured interviews
Limited time and financial resources may lead some qualitative researchers
to pursue other data collection techniques, such as a structured interview
schedule with open-ended questions.
Qualitative Research Designs (Leedy & Ormrod, 2005)
Case study
Ethnography
Phenomenological study
Grounded theory studyContent analysis
8/9/2019 Mixed Approach
8/40
We dig deep to get a complete understanding of the phenomenon.
We collect different forms of data and examine them from various angles
to construct a rich and meaningful picture of a complex, multifaceted
situation - triangulation.
Note: There are several different research designs to use but they: all focus on a phenomenon in its natural setting in the real world.
involve studying the phenomenon in all its complexity in a multifaceted
and all its dimensions.
The researcher must keep his/her perceptions, impressions and biases to
him/herself. What matters is to get the truth.
In the academic social sciences the most frequently used qualitativeresearch approaches include the following:Ethnographic Research, used for investigating cultures by collecting anddescribing data that is intended to help in the development of a theory.Grounded Theory, an inductive type of research, based or grounded in
the observations or data from which it was developed; it uses a variety ofdata sources, including quantitative data, review of records, interviews,observation and surveys.Phenomenological Research, describes the subjective reality of anevent, as perceived by the study population; it is the study of aphenomenon.
8/9/2019 Mixed Approach
9/40
Data analysis
Interview questions and responses are typically tape-recorded and then
transcribed verbatim before analysis is begun.
Qualitative researchers often categorize data into patterns as theprimary basis for organizing and reporting results.
Narrative descriptions of data collected through interviews, observations,
and case records are also used in qualitative analysis.
Content analysis is often used in qualitative and quantitative researchmethods.
8/9/2019 Mixed Approach
10/40
When to use qualitative approach?
Purposes:
Description of situations, settings, processes, relationships or people
Interpretation enables researchers:
oTo gain new insights about a particular phenomenon
oDevelop new concepts or theoretical perspectives about a phenomenon, or
oDiscover the problems that occur within the phenomenon
Verification to test the validity of certain assumptions, claims, theories or
generalizations within real-world.Evaluation provides means through which a researcher can assess the
effectiveness of a policy, programme, practice or innovation
Note: Qualitative research cannot address cause-and-effect relationships or questions
like, why did such and such happen? You need quantitative research, especially
experiments.
8/9/2019 Mixed Approach
11/40
Qualitative Research DesignsQualitative Research Designs
1. A case study
- A particular individual programme or event is studied in depth for a specifiedperiod of time
e.g., a medical researcher studying the nature and treatment of a rare illness for a
particular patient or strategies a teacher uses to teach his/her school children
- can be generalized to similar situations or environments
- sometimes researchers study 2 or more cases for comparison, build a theory or
generalizations
- may be suitable for learning more about a little known or poorly understood
situations or investigating how an individual or programme changes over time
- case studies can be useful to generate or provide preliminary support for
hypotheses
Note:T
he problem is generalization is not easy
8/9/2019 Mixed Approach
12/40
8/9/2019 Mixed Approach
13/40
Method
Site-based fieldwork
P
rolonged engagement in the groups natural settingThe researcher observe and record processes
First step is to gain access to the site
The research must often go thru a gatekeeper, e.g., tribal chief, a principal
After gaining entry into the site, the researcher must establish rapport with and gain
trust of the people being studied
Researcher must be open about why he/she is there. This is ethical.Informed consent is important
A big net approach, intermingling with everyone and getting overall sense of the
social and cultural context
It requires considerable patience and tolerance
8/9/2019 Mixed Approach
14/40
Gradually, the researcher identifies key informants who can provide the necessary
information and insights relevant to the research question and can facilitate contacts
with other helpful individualsSometimes the researcher engages in participant observation, becoming immersed
in the daily life of the people.
Throughout the fieldwork, the researcher must be a careful observer, interviewer
and listener
Lengthy conversations and significant events can be recorded using audiotapes and
videotapes.The researcher may also collect artifacts (e.g., tools, ritualistic implements, artistic
creations) and records (e.g., accounting ledgers, personal journals, etc) from the
group.
8/9/2019 Mixed Approach
15/40
3. Phenomenological study
Phenomenology refers to a persons perception of the meaning of an event, as
opposed to the event as it exists external to the person.
It is a study that attempts to understand peoples perceptions, perspectives, and
understandings of a particular situation.
E.g., a researcher might study the experiences of people caring for a dying relative,
living in abusive relationship, etc
Sometimes the researcher wants to gain a betterunderstanding of the experience
related to a phenomenon. By looking at multiple perspectives on the same
situation, the researcher can make some generalizations.
Method
Depends almost exclusively on lengthy interviews (say, 1 to 2 hours) with a
carefully selected sample of participants. Say, 5 to 25 individuals, all of whom
have had direct experience with the phenomenon being studied (Creswell, 1998).
Often a very unstructured interviewsThe researcher listens closely as participants describe their everyday experiences
A typical interview looks more like informal conversation, with the participant
doing most of the talking and the researcher doing most of the listening.
8/9/2019 Mixed Approach
16/40
Grounded theory study
Least likely to begin from a particular theoretical framework.
Main purpose is to begin with the data and use them to develop a theory
The study uses a prescribed set of procedures for analysing data and
constructing a theoretical model from them
Typically, the study focuses on a process (including peoples actions and
interactions) related to a particular topic, with the ultimate goal of developing a
theory about that process (Creswell, 2008).
E.g., to study childrens eating habits, college students thoughts and feelings during
classroom discussions, and workers stress levels in public service agencies.
8/9/2019 Mixed Approach
17/40
Method
Data collection is field-based, flexible, and likely to change over the course of the
study.
Interviews typically play a major role in data collection, but observations,documents, historical records, videotapes, and anything else of potential relevance to
the research question may also be used.
Like in the case of the other designs, data analysis begins almost immediately the
researcher develops categories to classify the data.
Subsequent data collection is aimed at saturating the categories in essence,
learning as much about them as possible and at finding any disconfirming evidence
that may suggest revisions in the categories identified or interrelationships among
them.
The process of moving back and forth between data collection and data analysis,
with data analysis driving later data collection, is sometimes called the constant
comparative method.The theory that ultimately evolves is one that includes numerous concepts and
interrelationships among those concepts
8/9/2019 Mixed Approach
18/40
Content analysis
A detailed and systematic examination of the contents of a particular material
for the purpose ofidentifying patterns, themes or biases.Typically performed on forms of communication, including books, newspaper,
films, television, art, music, videotapes, transcripts of conversations, etc.
The researcher defines a specific research problem or question at the very
beginning.
E.g. Do contemporary television commercials reflect traditional gender stereotypes?
Method
Identify the specific body of material to be studied
Define the characteristics or qualities to be examined in precise concrete terms
Etc
8/9/2019 Mixed Approach
19/40
When to choose a qualitative method
Qualitative research serves the following purposes:
Descriptive reveal nature of certain situations, settings, processes,
relationships, system or people.Interpretation to gain new insights about a phenomenon, develop new
concepts and theoretical perspectives, etc and discover the problem that
exists within the phenomenon.
Verification it allows the researcher to test the validity of certain
assumptions, claims, theories or generalizations within the real world
Evaluation a means through which a researcher can assess theeffectiveness of a particular policy, programme, practice or innovation. For
e.g., understanding how and why programme outcomes are not achieved.
Programme evaluation studies involving the qualitative approach focus on
participants' perceptions and their experiences in the programme (Bogdan &
Taylor, 1990; Patton, 1990; Rubin & Babbie, 1993).
Tries to paint a picture of a given situation by addressing questions: who,
what, when, where and how. For e.g., an economic or employment situation
in SA
8/9/2019 Mixed Approach
20/40
Note:
Qualitative studies normally take long time periods, e.g. months and years
They do not normally give quick and easy answers to research questionsQualitative studies are subjective and it is difficult for other researchers to
repeat the studies, let alone to evaluate the studies in terms of the researcher
biases.
Unlike for a quantitative study, in the case of a qualitative research, the
researcher is the data collection instrument
Purposive sampling based on previous research is normally usedQualitative approach is weak at investigating relationships between
phenomena and predicting phenomena
Generalizations tend to be impossible because they normally use
nonrandom and small sample sizes
8/9/2019 Mixed Approach
21/40
QUANTITATIVE APPROACHQUANTITATIVE APPROACH
Uses mathematics and statistical methods to study phenomena.
Often used to study relationships between variables or phenomena and topredict
Uses statistical methods to test hypotheses
It is more objective than the qualitative approach
Its designs are mostly:
Experiments, and
Research surveys.
Quantitative studies normally use large sample sizes, randomly selected
subjects and representative samples
Randomization of any experimental groups is essential, and a control
group should be included, wherever possible.
A sound quantitative design should only manipulate one variable at a time,or statistical analysis becomes difficult and open to question.
Ideally, the research should be constructed in a manner that allows other
researchers to repeat the experiment or study and obtain similar results.
8/9/2019 Mixed Approach
22/40
Advantages
Using a quantitative research design is an excellent way of finalizing results
and proving or disproving a hypothesis.
The structure is standard across many scientific fields and disciplines.After a statistical analysis of the data, a comprehensive answer is reached,
and the results can be legitimately discussed and published.
Quantitative studies try to filter out external factors (if properly designed), and
so the results gained can be seen as real and unbiased.
Scientific experiments are useful for testing the data gained by a series of
qualitative experiments, leading to a final answer, and a narrowing down ofpossible directions for follow up research to take.
Disadvantages
Quantitative studies must be carefully planned to ensure that there is
complete randomization and correct designation ofcontrol groups.
Quantitative studies usually require extensive statistical analysis.
8/9/2019 Mixed Approach
23/40
Experiment
regarded as the most accurate and unequivocal standard for testing a
hypothesis.generates good statistical data.
manipulates one or more variables to generate analyzable data.
should be designed with laterstatistical tests in mind, by making sure that
the experiment has controls and a large enough sample group to provide
statistically valid results.
Every true experimental design must have a hypothesis to test as the
ultimate aim of any experiment.
The precursor to a hypothesis is a research problem, usually framed as a
question. It might ask what, or why something is happening.
Experiments are sometimes referred to as true science. They use
traditional mathematical and statistical means to measure and analyse dataconclusively.
8/9/2019 Mixed Approach
24/40
The Basics
With most true experiments, the researcher is trying to establish a causal
relationship between variables, by manipulating an independent variable to
assess its effect upon dependent variables.Random groups are the best way of ensuring that the groups are as
identical as possible.
The basis of conducting an experiment
With an experiment, the researcher is trying to learn something new about
the world, an explanation of why something happens.
This method of study is especially useful for trying to test theoreticalmodels by using them in real world situations.
Interpretation
The observations are often referred to as 'empirical evidence' and the
logic/thinking leads to the conclusions.
8/9/2019 Mixed Approach
25/40
2. Survey Research Design
A Survey
Five preliminary steps that should be taken when embarking on a research project
can be identified.T
hey are:choose a topic
review the literature
determine the research question or objectives
develop a hypothesis, and
operationalise, that is, find the suitable research methodology and use it to
implement the research plan to answer the research question or to achieve pre-determined objectives of the study.
Two additional considerations that are very crucial, namely:
designing a representative sample randomly selected, and
a questionnaire to be used to collect data.
By a representative sample, we mean an accurate proportional representation
of the population under study.In this sample, every characteristic in the population should be well or fairly
represented.
In other words, to obtain reliable results on the characteristics of interest of the
population, a sample, through the process of randomization should in all relevant
respects be a true image or reflection of the target population.
8/9/2019 Mixed Approach
26/40
8/9/2019 Mixed Approach
27/40
Longitudinal Study
A longitudinal study is observational research performed over a period
of years or even decades, and allows social scientists and economists to
study long-term effects in a human population.A cohort study is a subset of the longitudinal study because it observes
the effect of a factor or variable on a specific group of people over time.
There are two main sub-types of cohort study, the retrospective and the
prospective cohort study.
The major difference: retrospective looks at phenomena that have
already happened, whilst the prospective type starts from the present.Trend studies focus on a particular population, which is sampled and
scrutinized repeatedly.
An example of a trend study is an annual survey of the average hourly earnings
of workers in the manufacturing industry in South Africa over a long period of
time.
A trend line is then fitted to the data.
While samples are of the same population, they are typically not composed of
the same people.
Several data from several studies of the same population (not necessarily done
by the same researcher) may be combined to investigate the trend of the
characteristic (or variable) of interest.
8/9/2019 Mixed Approach
28/40
Cohort studies also focus on a particular population sampled and studied more
than once.
For example, a sample of the 2010 first-year students at the SBL could be
questioned regarding their attitude toward the library staff.Two years later, the researcher could question another sample of the same 2010
first-year students and study any changes in attitude.
Note: If after the two years, the 2012 first-year students were studied, the study
would be a trend study instead.
Retrospective Cohort StudyThe retrospective case study is historical in nature. The researcher looks
at historical data to judge the effects of the variable.
It is a lot easier than the prospective, but there is no control, and
confounding variables can be a problem, as the researcher cannot easily
assess the lifestyle of the subject.
A retrospective study is a very cheap and effective way of studyinghealth risks or the effects of exposure to pollutants and toxins.
It gives results quickly, at the cost ofvalidity, because it is impossible to
eliminate all of the potentially confounding variables from historical
records and interviews alone.
Note: Secondary data can be used here
8/9/2019 Mixed Approach
29/40
Prospective Cohort Study
In a prospective cohort study, the effects of a certain variable are plotted
over time, and the study becomes an ongoing process.
To maintain validity, all of the subjects must be initially free of the
condition tested for.
Ambidirectional Cohort Study
The ambidirectional cohort study is the ultimate method, combining
retrospective and prospective aspects.
The researcher studies and analyzes the previous history of the cohortsand then continues the research in a prospective manner.
This gives the most accurate results, but is an extremely difficult
undertaking, costing time and a great deal of money.
The ambidirectional study shares one major drawback with the
prospective study, in that it is impossible to guarantee that any data can
be followed up, as participants may decline to participate or dieprematurely.
These studies need to look at very large samples to ensure that any
attributional losses can be absorbed by the statistics
8/9/2019 Mixed Approach
30/40
Panel studies allow the researcher to find out why changes in the population are
occurring since they use the same sample of people every time.
T
hat sample is referred to as a panel.A researcher could for example, select a sample of the SBL students and asks
them questions on their future work expectations.
Every year thereafter, the researcher would contact the same people and ask
them similar questions and ask them the reasons for any changes in their
expectations.
Panel studies suffer from attrition, that is, people drop out of the study forvarious reasons, for example, moving away from the area of study, dying,
deciding not to participate in the subsequent surveys, etc.
8/9/2019 Mixed Approach
31/40
8/9/2019 Mixed Approach
32/40
This does not necessarily mean using the postal service; this includes
delivering it physically.
This is a good way of targeting a certain section of people and isexcellent if you need to ask personal or potentially embarrassing
questions.
The problems with this method are that you cannot be sure of how many
responses you will receive until a long time period has passed.
Structuring andD
esigning the QuestionnaireThe design of your questionnaire depends very much upon the type of
survey and the target audience.
If you are asking questions face to face it is easy to explain if people are
unsure of a question.
On the other hand, if your questionnaire is going to include many
personal questions then mailing methods are preferable.You must keep your questionnaire as short as possible; people will either
refuse to fill in a long questionnaire or get bored halfway through.
If you do have lots of information then it may be preferable to offer
multiple-choice or rating questions to make life easier.
8/9/2019 Mixed Approach
33/40
Statistical data Sets
These are a collection of data - maintained in an organized form.
The basis of any statistical analysis has to start with the collection of
data, which is then analyzed using statistical tools.
Therefore statistical data sets form the basis from which statistical
inferences can be drawn.
Statistical data sets may record as much information as is required by
the experiment or survey.
For example, to study the relationship between height and age, onlythe measures of these two variables might be recorded in the data set.
Creating a statistical data set is only the first step in research.
The interpretation and validity of the inferences drawn from the data is
what is most important.
The next step is to devise which statistical test you are going to use
and start to enter some numbers to judge the significance of your data.
8/9/2019 Mixed Approach
34/40
USE OF MIXED METHODSUSE OF MIXED METHODS
How to choose an approach or method
Depends on the research questions and the research skill of theresearcher.
Appropriateness ofMethod
Some methods are better suited to a study of certain research questions than
others. Some might even be totally inappropriate to your research question.
So type of research methodology depends on:
Research Questions/problem
Variables qualitative or quantitative
8/9/2019 Mixed Approach
35/40
Type of data nominal, ordinal, interval or ratio
1. Nominal-scaled data
These are mainly qualitative and assigned to a number of categories of equal importance,
for example, gender [male, female].2. Ordinal-scaled data
Ordinal-scaled data are mainly qualitative and assigned to coded categories. Ranking is
implied between categories, for example, Class [1st years, 2nd years, 3rd years, Honours].
3. Interval-scaled data
These are quantitative; they possess both order and distance but no origin, for example,
IQ scores and temperature (say, in degrees Centigrade).
4. Ratio-scaled data
Ratio-scaled data are quantitative with a zero origin, for example, age, height and weight.
They possess all the properties of the other data types. Ratio is regarded as the
strongest level of data and nominal the weakest.
8/9/2019 Mixed Approach
36/40
When the phenomenon being investigated is not well understood, a
qualitative research can be used first to generate preliminary hypotheses,
and then quantitative research is used to test specific hypotheses.
For comprehensive studies, it is better to use both approaches (i.e.
qualitative and quantitative), a situation where, the quantitative method
helps to test hypotheses and to make the study more objective and the
qualitative method helps to give a complete understanding or in-depth
information on the phenomenon being studied in all its dimensions.
Complex studies, having many different research questions may use
both approaches to answer particular questions or to address particular
research problems.
E.g., poverty is multidimensional in nature, i.e., it has social, political and
economic dimensions. So, a poverty study may need both approaches to
be studied a situation where the qualitative research methods are usedto investigate the social and political issues and help the voice of the poor
to be heard and quantitative methods are used to study the economic
issues.
8/9/2019 Mixed Approach
37/40
8/9/2019 Mixed Approach
38/40
Normally a semi-structured questionnaire is used to include both
closed and open-ended questions.
Note:
Social experiences and the realities of the world are multidimensional andso if phenomena are viewed only along a single dimension we cannot have a
complete picture of what is going on around us.
Qualitative empirical research tends to expose the complexity of real life
experience.
The use of mixed methods and a multidimensional approach allows the
researcher to frame questions which precisely focus on how differentdimensions and scales of social existence are related.
The particular strength of qualitative research lies in the knowledge provided
regarding the dynamics of social processes, change and social context, and in its
ability to answer how and why questions in those domains.
There are two core elements to the logic of qualitative explanation: one relates
to a qualitative logic of comparison, e.g., between cases, situations, contexts,
over time, etc and the second relates to the significance of context.
Understanding how social processes and phenomena are embedded within
specific contexts makes possible the development of cross-contextual
generalizations.
8/9/2019 Mixed Approach
39/40
This entails the use of contextual explanation, where emphasis is on explaining
how different dimensions of context together link up together with the
processes or questions driving the study.
Using mixed methods to triangulate or to corroborate each other suggests an
integrated framework, where each method and form of data is used to highlight
a specific part of the picture.
There might however be tensions arising from differences in approaches to
data collection because of using different sets of assumptions.
To overcome any such potential difficulty the solution is in how explanations
are done.
Explanations do not have to be internally consistent to have a meaning and
capacity to explain.
If indeed the realities of the world are multidimensional political, social,
cultural, economic, etc, then explanations arising from mixed data-collection methods would be likewise (see Cho and Trent, 2006; Dixon-
Woods, et. al. 2006, Moran-Ellis, 2006; Mason, 2006 in Kimani, 2009).
8/9/2019 Mixed Approach
40/40
Remember:
Using a quantitative research design is an excellent way of finalizing
results and proving or disproving a hypothesis, and
Scientific experiments are useful for testing the data gained by a series
of qualitative experiments, leading to a final answer, and a narrowing
down of possible directions for follow up research to take.