RESEARCH METHODS

QUALITATIVE & QUANTITATIVE

RESEARCH DESIGNS & METHODOLOGY

LECTURE 3
Qualitative Research Approach
 Qualitative research is a scientific inquiry that uses
methods that do not depend on comparing quantities
 Qualitative research captures unquantifiable
phenomena/data such as emotions,
viewpoints/perspectives, value judgements,
 Understanding of a situation is gained through an
holistic perspective rather than numerical analysis of
data

TEND TO USE QUALITATIVE RESEARCH IF THE RESEARCH PROBLEM

REQUIRES YOU TO:
1. Learn about the view of individuals
2. Assess a process over time
3. Generate theories based on participant perspectives
4. Obtain detailed information about a few people or research sites
Qualitative Research Designs
1. Ethnographic study (field study)
 Studies a society or some aspect of a society, culture or group in

depth
 Ethnography is a form of field research that seeks to learn the

culture of a particular setting or environment.

 It focuses on an entire cultural group

 Ethnography involves immersing the researcher in the community to

gain a deep understanding of community dynamics, culture, and

social norms
2. Case studies
1. Case studies involve in-depth analysis of a single case or a small

number of cases. Researchers use this method to understand

complex issues and how they play out in real-world situations
 Case study research is a qualitative approach in which the
investigator explores a bounded system ( a case) or multiple systems
(cases) over time
 A case study is an in-depth study of a single unit such as
 one individual,
-one group,
-One organisation
-One program and so on
 The goal is to arrive at a detailed description and understanding of

the entity
3. Phenomenological studies
 Phenomenological studies begins with the assumption that

multiple realities are rooted in subjects’ perspectives.

 That is to say an experience has different meanings for each

person.
 It describes the meaning for several individuals of their lived

experiences of a specific phenomenon.

 It seeks to describe how individuals experience a specific

phenomenon
4. Grounded Theory
While phenomenology emphasises the meaning
of an experience for a number of individuals,
- The intent of grounded theory approach is to
move beyond description and to generate or
discover a theory
 A qualitative design in which the enquirer
generates a general explanation ( a theory) of
a process, action or interaction shaped by
views of a larger number of participants
5. Historical
 Historical research analyses documents and

artifacts to gain insights into what has

happened in the past.
 Its success depends on the accuracy and

completeness of these records

Qualitative Sampling Approaches
 The sample size varies depending on the
breadth and complexity of the inquiry/ issue
 Samples are generally smaller than those

used in quantitative studies and are studied

intensively.
 Adequacy of the sample size is determined by

the principle of theoretical saturation.

Refers to the point at which no new concepts
emerge from the review of successive data from a
sample that is diverse in pertinent characteristics
and experiences
QUALITATIVE SAMPLING
TECHNIQUES
 Qualitative sampling approaches use NON-
PROBABILITY sampling techniques such as:
1. Purposive/Judgmental sampling
 Sampling which groups participants

according to preselected criteria relevant to a

particular research question
 This type of sampling usually deliberately

targets individuals within a population.

2. Quota Sampling
 It is a method in which elements are selected

to ensure that the sample represents certain

characteristics in proportion to their
prevalence in the population
 Samples are taken from the population in

proportions relative to different categories

like gender, ethnicity, age groups, socio-
economic groups and region of residence
3. Convenience/Availability Sampling
 Is based on who is available and willing to participate in the
research.
 Usually the sample is made up of individuals who are the
easiest to recruit (Kelley et al. 2003; p. 263).
 A sampling procedure in which the researcher selects
participants because they are willing and available to be
studied
4. Snowball Sampling
 A method where by participants or informants with

whom contact has been made use their social networks

to refer the researcher to other people who could
potentially participate in the study
 The researcher chooses a few respondents for instance

through accidental sampling or any other method, and

then asks them to recommend other people who meet
the criteria of the researcher who might be willing to
take part
 Snowballing is also referred to as ―chain referral.

 It is often used to find and recruit hidden ―populations

(schutt, 2004).
 It is important to carefully consider the strengths
and limitations of each sampling technique when
designing a community development research
study, and to choose a method that is appropriate
for the research question and the population being
studied.
DATA COLLECTION METHODS &
INSTRUMENTS
 Both qualitative and quantitative research methods can
be useful for advanced community development
research. Qualitative methods are useful for exploring
complex social, cultural, and economic factors that
influence communities, while quantitative methods can
help to identify trends and patterns, and provide
statistical evidence to support community development
interventions.
Here are some examples of qualitative research
methods that can be used for advanced community
development research:
DATA COLLECTION METHODS &
INSTRUMENTS
1. Unstructured/ semi-structured interview (intensive interviewing)
 One-on-one interviews with KI can be used to gather detailed information

about community members' experiences, attitudes, and perceptions related

to community development issues
 Uses interview guide
 Audio recorder

2. Focus Group Discussion

 Focus groups can be used to gather insights from multiple community

members at once, allowing for a broader range of perspectives and group

dynamics to emerge.
 Discussion guide

 Uses checklist

 Audio Recorder

3.Participant/ non- participant observation

 Observation involves watching people in their natural environment and taking

notes on their behavior. It can be useful in understanding the social dynamics

of a community and identifying patterns of behavior.
 Uses Observation schedule/Checklist/ Camera/video recording
3. Document Review/ analysis
 It refers to a written summary of all secondary sources for one‘s research

 Documents include but are not limited to

i. institutional documents (clinical, programmatic, or organizational records),

ii. personal documents (diaries, letters, artistic expressions), and
iii. public historical documents (legislative testimony, legal documents).
 Others include;

i. printed texts, film, audio tape, presentations and lectures, paintings,

handwritten diaries, archival sources, legislation, websites, artefacts,
CDs, DVDs and theses
 Document analysis involves reviewing documents such as
government reports, newspapers, and organizational records. It can
be useful in understanding the historical context of a community
and identifying key issues and concerns.
TOOLS USED IN DOCUMENT REVIEW
1.Content analysis
 Used when originally qualitative information is needed to
understand participants categories and come up with
quantifiable units or categories
2. Literature review
 Used to identify gaps in knowledge

3. Diaries
 Used to get first hand information about

phenomena
DATA ANALYSIS TECHNIQUES

1.Thematic Analysis/ Discourse Analysis

2. Narrative analysis
2. Analytic induction,
3. grounded theory,
4. Qualitative Comparative analysis
Computer Assisted Qualitative
Data Analysis (CAQDA)
1.Nvivo
2. Eviews
3. etc
QUANTITATIVE RESEARCH
APPROACH

What is Quantitative research?

 Formal, objective, rigorous, systematic process for
generating information

Features of Quantitative Research

i. The goal of Quantitative research is to:
 Describe new situations, events, or concepts
 Examines relationships among variables
 Determines the effectiveness of treatments
i. Quantitative Research is Based on the philosophy of “logical
empiricism” or “positivism” in which reality is studied
objectively
ii. Researcher adopts a distant and non-interactive posture to
prevent bias
iii. Quantitative research, is more concerned with questions
about: how much? how many? how often? to what extent?
 QUANTITATIVE RESEARCH
APPROACHES/DESIGNS
 Quantitative research design is the technique and measurement that produces
quantifiable/discrete values (Kothari, 2007)
 The approaches to quantitative research include:
1. Experimental
a. True experiment
b. Quasi-experiments
2. Non-experiments
 Non-experimental design is basically quantitative research which does not involve
experiments in the process of data collection. They include:
a. Surveys
b. Descriptive design (Causal-comparative design, case study design)
c. Correlational design
A. EXPERIMENTS
1.EXPERIMENTAL DESIGN(TRUE EXPERIMENT)
 The Experimental design involves gathering a set of individuals or
other units of analysis and then randomly dividing them into two
groups:
 those who will receive the intervention, i.e., the treatment group
and
 those who will not receive the intervention, i.e., the control group.
 True experiment where the researcher randomly assigns the
participants to treatment groups.
 Experimental or randomized designs are the gold standard of
evaluation research designs and are the most robust of evaluation
methodologies.
 Experimental designs can be expensive and time consuming in
certain situations, particularly in the collection of new data.

2.QUASI-EXPERIMENTAL DESIGN (PARTIA- EXPERIMENT)

 The Quasi-experimental or semi-experimental designs are those
in which the allocation to treatment and comparison groups is
not randomized.
 Quasi-experiments where the researcher uses control and
experimental groups but does not randomly assign participants to
groups
 Quasi-experimental designs are the best available option for
evaluation research, in cases where it is not feasible to apply
experimental designs
B. NON-EXPERIMENTAL DESIGNS
1. SURVEYS
 This type of research design provides a numeric description or

trends of a population by studying a sample of that population.

 According to McNeill and Chapman, (2005) it is a method of

obtaining large amounts of data, usually in a statistical form, from a

large number of people in a relatively short time using closed-ended
questions.
 It is concerned with the present and attempts to determine the

status of the phenomena being investigated (Singh, 2006).

 This method comes in handy when a researcher is studying several

variables using a large sample size and rigorous statistical analysis

(Sjøberg et al., 2007).
 Survey research observations can be cross-sectional or longitudinal studies.
a. Cross-sectional - one-time survey
 the observations are done at one point in time while

b. Longitudinal study - one sample taken and followed over time

 means the observations are carried out at different points in time.
 This survey is often implemented for trend analysis or studies where the primary

objective is to collect and analyze a pattern in data.

i. Cohort study = a group identified by a certain characteristic and followed overtime.
ii. Panel design/study= The data is collected from different types of cases but
followed overtime such as household, schools, organizations, companies, businesses
 Cross sectional research design is descriptive, exploratory and explanatory while

longitudinal survey focus on trend analysis cohort design and panel design
(Jongbo, 2014).
 It either uses questionnaires or structured interviews for data collection with the

intention of generalizing the sampled data to a population (Fowler, 2009).

 The principal advantage of survey studies is that they provide
information on large groups of people, with very little effort, and
in a cost-effective manner (Marczyk et al., 2005).
 The major drawbacks of survey design are:

i. controlling against sample bias which can greatly compromise

generalization of the findings to the population
ii. its dependency on the cooperation of the respondents which in
the long run determines the degree of the results’ reliability.
iii. Information that is not known by the respondents can hardly
by unearthed and the information that is personal or secretive
may easily be inaccurate (Wabwoba & Ikoha, 2011).
2. Descriptive Designs
 • Most used design
 • Examines characteristics of a single sample
 • Identifies phenomenon, variables, and conceptual and operational definitions and

describes definitions

a. Causal-comparative
 Causal-comparative research is a type of non-experimental quantitative design where

the researcher compares two groups or more e.g. cross-cultural studies . Studies a
phenomenon in different countries and draw a line of similarity and difference
 This comparison is performed with respect to a cause (which is the independent

variable) which has already occurred (Creswell, 2014).

 Jongbo, (2014) attests that it is employed when there are two groups which differ on

independent variable and the researcher wishes to investigate the difference of one or
more dependent variables or independent variables
b. Case Study Designs
 • Exploration of single unit of study (e.g., family, group, or
community)
 • Even though sample is small, number of variables studied is
large.
4. Correlational research
 Correlational research is a non-experimental

quantitative design in which the researcher applies

correlational statistics to measure and describe the
degree of association among variables or sets of
scores (Creswell., 2012).
Quantitative sampling techniques
Sampling plan
 Outlines strategies used to obtain a sample for a

study
 Probability sampling plans (quantitative)
Probability sampling scheme
 Is one in which every unit in the population has an equal
chance of being selected in the sample and this probability can
be accurately determined.
 Probabilistic sampling includes the following approaches:
Simple Random Sampling, Systematic Random Sampling,
Stratified Sampling, and Multistage Cluster Sampling.
Probability sampling techniques
1. Simple Random Sampling
 Simple random sampling is the most basic form of probability sampling.
 With simple random sampling each unit or element in the population has an

equal probability of inclusion in the sample (Schutt, 2004: p.171)

 Simple random sampling involves the following steps;

- First one has to establish the sampling frame.

- Then you assign a single number to each and every element in the list while
making sure that no any element is skipped in the process (Brymann, 2008).
- After this you create a table with random numbers where elements are selected as
samples.
- Sometimes a computer is used to select study elements randomly.
- Although random sampling is easy to conduct it becomes labourious and
impractical if you have a large sampling framework.
- In addition minority sub groups of the interest population may not be present in
sample in sufficient numbers for study.
2. Systematic Sampling
 Bryman (2008) looked at the term systematic sampling as a probability

sampling method in which units are selected from a sampling frame

according to a fixed intervals or fixed ratio (p.700).
3. Stratified Sampling
 It is a quantitative sampling procedure in which researchers stratify the

population on some specific characteristic and then sample, using

systematic random sampling, from each stratum of the population
 - The characteristic may be for example gender
 - Stratified sampling are motivated by the need of increasing

representativeness and reducing error.

 A Stratified sample is only feasible in cases where the variable can be

grouped into distinctive strata/is categorical.
3. Cluster sampling
 Population disbursed over a large area is split into geographical region
and select a number of them.
4.Multi-Stage Cluster Sampling
 This is defined as a sampling method in which natural groups are sampled initially

with members of each group being sub-samples afterwards

 - E.G. sampling districts at regional level( northern cluster, central cluster, southern

cluster, eastern region cluster).

 With multi-stage cluster, the researcher can conduct a random study over a vast

geographic setting.
 Some challenges

 The method has been deemed labour intensive

 It has been argued that though highly efficient it is less accurate.

 multi-stage cluster also risks of being affected by two types of errors thus sampling

units error and sampling cluster error

Determining the sample size
 Population
 Sample frame
 Sample size

What is a sample?
 Defines the selected group of people or elements from which
data are collected for a study.
Sample Size Calculation
 Very few subjects in sample may give biased results hence hard to
generalize results to large population
 Larger samples are preferred, with caution in clinical trials – you don’t
want to put a lot of people at risk making study unethical at times
 In general, sample size calculation depends on
• Acceptable level of significance
• Power of the study
• Expected effect size
• Underlying event rate in the population
• Standard deviation in the population.
 Further note that there are numerous ways for computing sample sizes
depending on your context. You need to do a proper review before
deciding what method to use
Sample size calculation (Some considerations)
• Acceptable level of significance – Probability that the result is observed
due to chance and NOT due to our intervention, i.e. accepting change when
no change really exists,
• Power of the study – Defined probability of the false negative we are
willing to accept
• Expected effect size - The difference between the value of the variable in
the control group and that in the test group, from previous similar studies
• Underlying event rate in the population – Prevalence rate of the
condition under study in the population. If it turns out to be low when the
study is initiated, need to adjust for sample size
• Standard deviation in the population – Measure of variability and
dispersion in what is being measured.
Sample size calculation (Infinite population basic formula)
• Normally, a simple random sample without replacement
• For estimating mean, given as: 𝑛0 = z²∗ 𝞭²
e²

 𝑒 - acceptable sampling error/confidence interval,

 𝞭 - estimate of population variance,
 Z - value of standard normal variate for desired level of confidence,
• Formula can also e given in another form (estimating proportion) like:
no = 𝑍∗(𝑝(1 − 𝑝)
𝑒²
𝑝 – Estimated percentage of population picking a choice, (1 − 𝑝) is the opposite
 Most common Z values used are

1.645 = 90 percent confidence level

1.96 = 95 percent confidence level
2.576 = 99 percent confidence level
Sample size calculation(finite population
basic formula)
 • For a finite population, we can use the finite population
correction factor (FPC) to get the sample size.
 • Here, the sample size calculated from the infinite

formula (𝑛0) and the new sample size can be given as:
𝑛 = no∗𝑁
n+ (N-1)
• Further note that the FPC can be applied on the sample if
the calculated 𝑛! is at least 5% of the population N.
Sample Size Calculation (Yamane Method , 1967)
n = Where;
n- Sample size
N- Population under study
e- Sampling error/ Marginal error
QUANTITATIVE DATA COLLECTION
METHODS AND INSTRUMENTS