Research Proposal 4
Research Proposal 4
Research Proposal 4
• Experimental designs
Experimental design is concerned with causal- and- effect
relationships
Causal- and effect relationship occurs when one thing or event
makes some other thing or event to happen
Experimental designs involves manipulation or control of the
independent variable (cause) and measurement of the dependent
variables (effect)
Types of research designs
• Non-experimental designs
Non-experimental designs do not involve manipulation of
various variables
Examples are cross-sectional descriptive studies,
analytical studies, and longitudinal studies.
Types of research designs
Qualitative Research Designs
• Phenomenological design: focuses on the lived experiences. It
usually describe the meaning that experiences hold for each
participant.
• Ethnographic design: involve studies of analysis of data about
the life ways or particular pattern of a culture (or sub-culture)
• Grounded theory: design that focuses on theory
construction/theory development.
• Historical study designs: Studies geared at analysing the data
from the past. Seeks to discover the events of the past and relate
them to what is happening to the present and to the future.
• Case study: Designs focusing into looking at in-depth examination
of people or groups of people, institutions, or organizations.
Selecting a Research Design
• Research questions or objectives: quantitative research
questions or objectives are better addressed by quantitative
research designs ; whereas qualitative research questions are
better addressed by qualitative research designs ( e.g. case
study)
• Available resources: when resources are limited, descriptive
studies or explanatory research are suitable research designs
• Skills of the researcher: researchers with quantitative skills
tend to prefer quantitative research designs to qualitative
designs.
Research setting (place of study)
69
Definitions of Terms Related Sampling
Population
Is a particular group of people, organizations, events
behaviours or other elements that are the focus of a
research
Target population
Is the entire set of individuals or elements who meet the
sampling criteria
Accessible population
is the portion of the target population to which the
researchers have reasonable access
Definitions of Terms Related Sampling
Sampling frame
Usually denote a complete list of all the elements in a population who
meet the sampling criteria.
Sample
Denotes a segment of population or a selected group of people or
elements for inclusion in a study
Probability sample: a sample that has been selected using random
selection so that each unit in the population has a known chance of
being selected. It is generally assumed to be a representative sample
for the population
Non-probability sample: a sample that has not been selected using a
random selection method. Essentially this implies that some units in the
population are more likely to be selected than others
Definitions of Terms Related Sampling
Sampling criteria
Sampling criteria (eligibility criteria) include a list of characteristics
essential for membership or eligibility in the target population
The sampling criteria determine the target population, and the
sample is selected from the accessible population within the target
population
A research may have both inclusion and exclusion sampling criteria:
o Inclusion sampling criteria: are characteristics that a subject or
element must possess to be part of the target population. Example
inclusion sampling criteria may be the ability to read, to write
responses
o Exclusion sampling criteria: are characteristics that can cause a
person or element to be excluded from the target population.
Example exclusion criteria inability to read or write.
Definitions of Terms Related Sampling
Sampling plan
Is a plan that describes the process and sampling techniques
(methods) for selecting a sample from sampling frame.
Sampling process is important to researcher and research process
for the following two major reasons:
It help to obtain sample from large population when the entire
population members cannot be included in the research
It helps to use the limited research resources effectively to obtain
adequate information by using sample rather than entire
population.
SAMPLING PROCEDURES
• Sampling techniques refers to the specific method, design,
approach, or strategy that a researcher decides to use to select a
sample from the larger population.
There are two mainly types of sampling procedure:
• Probability Sampling In which every unit in the population has
equal or at least a known chance of being selected in the sample.
• Non Probability Sampling which is used in some situations,
where the population may not be well defined or in those situations
in which there may not be great interest in drawing inferences from
the sample to the population.
75
Probability sampling techniques
78
Non-Probability Sampling techniques
• Purposeful sampling
• Extreme case sampling
• Maximum variation sampling
• Homogeneous sampling
• Critical case sampling
• Snowball or chain sampling
79
Sample Size Determination
N= Z2 x P (1-P)
E2
Sample Size Determination
Where;
86
Data collection techniques
Written Questionnaire
• A questionnaire is a form containing a set of questions
submitted to people to gain statically information.
• A written questionnaire is a data collection method in
which written questions are presented that are to be
answered by the respondents in written form.
• The questions can be either open-ended or closed
• Self-administered questionnaire, can be administered by:
Mail
Gathering respondents in one place
Hand-delivering questionnaires to respondents and
collecting them later. 87
Data collection techniques
Written Questionnaire
• Advantages:
Low cost when study is large
Free from interviewer bias
Respondents have adequate time to give out their
answers
Useful for Respondents not easily approachable
Useful in large studies
Permits anonymity and may result in more honest
responses.
88
Data collection techniques
Written Questionnaire
• Limitations:
Low rate of return of the duly filled in questionnaires
Bias due to non-response is undetermined
Used only when respondents are educated and cooperating
Control over questionnaire may be lost
Not flexible once the questionnaire has been dispatched out
Ambiguous replies or omissions creating difficult in
interpretation
Slowest method of all when mailing is used.
89
Data collection techniques
Interview
• This is a data-collection technique that involves oral questioning of
respondents, either individually or as a group.
• Interview is of three types:
Structured: fixed number of questions and limited choice of
answers.
Semi-structured: few questions as guidelines, other questions
added as necessary
Unstructured: No fixed questions at all, the researcher
determines the questions after immersing in the field for some
time.
• Can be administered by:
Face-to-face 90
Data collection techniques
Interview
• Advantages:
Detailed information can be obtained.
Interviewer could overcome resistance of the respondent, if
any.
Provides flexibility to the interviewer to restructure, clarify, or
add probe questions.
Observation can also be applied during interview.
Is suitable for use with both literates and illiterates.
Has higher response rate than written questionnaires.
The interviewer can collect supplementary information about
the respondent’s personal characteristics and environment. 91
Data collection techniques
Interview
• Limitations:
It is very expensive and time consuming especially when the
sample is large.
Chances of Interviewer as well as interviewee’s bias are high.
People with certain high level positions like officials or
executives may not be easily approachable under this method
and to that extent the data may prove inadequate.
Creating effective rapport with the interviewee may be a difficult.
Some subjects may demand incentives during data collection
process.
There may be a language barrier between an interviewer and
interviewee.
92
Data collection techniques
Observation
• This technique involves systematically selecting, watching and
recording behavior and characteristics of living beings, objects
or phenomena.
• It can be undertaken in different ways:
Participant observation: The observer takes part in the situation s/he
observes.
Non-participant observation: The observer watches the situation, openly or
concealed, but does not participate.
• It becomes a scientific tool and method of data collection for
the researcher when:
It serves a formulated research purpose.
Is systematically planned and recorded.
93
Is subjected to checks and controls on validity and reliability.
Data collection techniques
Observation
• Advantages:
Subjective bias is eliminated, if observation is accurately done.
The information relates to what is currently happening.
Not complicated by either past behavior or future intentions or
attitudes.
Independent of respondent’s willingness to respond and hence
less demanding on active cooperation on the part of the
respondents.
Suitable in studies which deal with subjects who are not capable
of giving verbal reports of their feelings
94
Data collection techniques
Observation
• Limitations:
It is an expensive method
Information provided by this method is very limited
Sometimes unforeseen events may interfere with the
observational task
Some subjects are rarely accessible to direct
observation
If subjects know that they are being observed, they
may change their behavior
95
Data collection techniques
Reviewing documents
• This techniques involves reviewing documents obtained from
sources such as:
Health facilities
HIMS
Census
Unpublished reports
Published reports
96
Data collection techniques
Reviewing documents
• Advantages:
It is inexpensive, because data is already there.
It permits examination of trends over the past.
• Disadvantages:
Data are not always easily accessible.
Ethical issues concerning confidentiality may arise.
Information may be inaccurate or incomplete.
97
Data collection techniques
Focused Group Discussion
• In this technique a group of 6 - 12 informants freely and
spontaneously discuss a certain subject with the guidance of a
facilitator.
• It is a qualitative method and aims to be more than a question-
answer interaction.
• It is useful to:
Focus research and develop relevant research hypotheses by exploring
in greater depth the problem to be investigated and its possible causes.
Formulate appropriate questions for more structured, larger scale
surveys.
Help understand and solve unexpected problems in interventions.
Develop appropriate messages for health education programmes and 98
Data collection techniques
Focused Group Discussion
• Advantages:
The researcher can interact with the participants, pose,
follow up questions or ask questions that probe more
deeply.
Results can be easier to understand than complicated
statistical data.
The researcher can get information from non-verbal
responses such as facial expressions or body language.
Information is provided more quickly than if people were
interviewed separately.
99
Data collection techniques
Focused Group Discussion
• Limitations:
The small sample size means the groups might not be
a good representation of the larger population
Group discussions can be difficult to steer and control,
so time can be lost to irrelevant topics
Respondents can feel peer pressure to give similar
answers to the moderators’ questions
The moderators skills in phrasing questions along with
setting can affect responses and skew results
100
Tools used for data collection techniques
Data Collection
Data Collection Tools
Techniques
Reviewing Checklist, data compilation forms
documents
117
Sorting Data
• This is important for facilitating subsequent processing and
analysis.
• Questionnaire for different study populations (e.g. health
workers, village and general population) should be numbered
separately.
• In comparative study sort the data right after collection into the
two or three groups that will be compared during data analysis.
118
Performing Quality Control
Checks
• First in the field to ensure proper and complete data collection
and recording.
• Then before and during data processing, to ensure completeness
and internal consistency.
• Incomplete or inconsistent questionnaires may have to be
excluded from further analysis or returned to the respondent for
correction.
119
Data Processing
This involves:
• Categorizing/classifying the data:
• Coding
• Summarizing the data in data master sheets, manual
compilation without master sheets, or data entry and
verification by computer
120
Data Processing
Categorizing/classifying the data:
• This may fully be done earlier for data obtained through
closed questions.
• Answers to open-ended questions can be pre-categorized
to a certain extent depending on knowledge of possible
answers but there should always be a category of “others,
specify ….” whose categorization is done afterwards,
• For numerical variables, categorization is often done after
data collection
• Wrongly categorized data cannot be reclassified during
data analysis
121
Data Processing
Data coding:
• This is a process of assigning letters, numbers or group of letters to responses.
• Developing a coding system is essential especially if data will be entered in a
computer for subsequent processing and analysis.
• The codes should be entered on the questionnaires (or checklists) themselves
and used during data processing.
• Common responses should have the same code in each question to minimize
mistakes by coders. For example:
Yes (or positive response) code - Y or 1
No (or negative response) code - N or 2
Don’t know code - D or 8
No response/unknown code - U or 9
• Codes for open-ended questions (in questionnaires) can be done only after
examining a sample of (say 20) questionnaires in which similar types of
responses are grouped into single categories to limit their number to at most 122
6
or 7.
Data Processing
Summarizing Data: This can be achieved through:
• Data Master Sheets
If data are processed by hand, it is often most efficient to
summarize the raw research data in a so-called data master
sheet, to facilitate data analysis.
On a data master sheet all the answers of individual
respondents are entered by hand.
• Data Compilation by Hand
Useful when the sample is small and the collected data is
limited.
If only one person is doing the compilation use manual sorting.
If a team of 2 persons work together use either manual sorting
or tally counting. 123
Data Processing
Summarizing Data:
• Computer Compilation
Not useful for small samples and in case data collection was
mainly through open questions.
It is costful and time consuming.
Computer compilation involves:
Choosing an appropriate computer program
Data entry
Verification or validation of the data
Programming (if necessary)
Computer outputs/prints.
The most widely used computer programs are Epi-info 6, LUTUS
124
1-2-3, dBase (versions III or IV) and SPSS
Analysis and Interpretation of Data
Analysis of quantitative data: involves the production of
frequencies, tables, graphs that describe the data
125
Analysis and Interpretation of Data
Frequency Counts
• From the data master sheets, simple tables can be made with
frequency counts for each variable in the questionnaire.
• A frequency count is an enumeration of how often a certain
measurement or a certain answer to a specific question occurs.
• It should be obtained for every question in the questionnaire.
• Frequency count helps to ensure:
The total number of responses in each question is correct
All codes are relevant to the question.
• For easy comparison of groups and especially if numbers are
large enough it is better to calculate the frequency distribution
126
in percentages.
Analysis and Interpretation of Data
Cross-Tabulations
• This involves combination of information on two or more variables
in order to describe the problem or to arrive at possible
explanations for it.
• Generally two major kinds of cross-tabulations may be required in
a study:
Descriptive cross-tabulations: which aims at describing the
problem under study by presenting a combination of variables.
They are also used to describe the subjects’ background
variables like age, sex, profession, education.
Analytic cross-tabulations: Groups are compared in order to
determine differences, or exploring relationships between
variables. 127
Analysis and Interpretation of Data
Interpretation of Results
• This refers to the task of drawing inferences from the
collected facts after an analytical and/or experimental study.
• Is done after collecting and analyzing the data.
• This has to be done very carefully to avoid drawing
misleading conclusions.
• Here the researcher can expose relations and processes
that bring about his findings.
• This involves combination of information on two or more
variables in order to describe the problem or to arrive at
possible explanations for it. 128
Steps in Data analysis
Seven Steps in Analyzing Data
1. Step 1: Take a sample of (say 20) questionnaires and list all answers for a
particular question.
2. Step 2: To establish the categories and coding
3. Step 3: List the answers again, grouping those with the same code
together.
4. Step 4: Interpret each category of answers and try to give it a label that
covers the content of all answers.
5. Step 5: Try a next batch of 20 questionnaires and check if the labels work.
Adjust the categories and labels, if necessary.
6. Step 6: Make a final list of labels for each category and give each label a
code (keyword, letter or number).
7. Step 7: Code all your data, including what have already coded, and enter
these codes in the master sheet or in the computer. 129
Common Significance Tests for Statistical
Analysis
• Significance tests are a set of statistical analysis aimed at
determining whether the differences and associations found in
the study are significant.
• Any observed difference between groups or association
between variables can be a result of:
True association/differences
Chance
Bias
Confounding
• While bias and confounding can be excluded by checking the
study design, statistical significance tests help to rule out
whether the observed results could have occurred by chance. 130
Common Significance Tests for Statistical
Analysis
• The most common significance tests are:
Student’s t – test
The chi-square test (χ2)
• The general agreed convention is to take 5% and below as the cut off
point
• In stating the results mention the P value i.e. Probability of chance
value by stating whether it was less than 0.05, 0.01 or 0.001.
• It is important to note that ‘significant’ does not necessarily mean that
an observed difference or association is an important one clinically:
Even a very small difference will show “statistical significance” if a big sample
is taken.
An important difference may fail to reach statistical significance if the sample
is small
131