Professional Documents
Culture Documents
6) Participants/units of study
https://sitn.hms.harvard.edu/flash/2019/not-so-basic-research-the-unrecognized-importance-
of-fundamental-scientific-discoveries/
1) The purpose or utility (Research Types I:
Basic or Applied)
Basic Applied
Theoretical? Quick Answers?
More Invasive? Less Invasive?
Laboratory Based? Field Based?
Internal Validity?
Tightly Controlled? Loosely Controlled?
Lacks External Validity? Externally Valid?
Focus on Mechanism Focus on Effect
More Reductionist Less Reductionist.
1) The purpose or utility (Research Types I:
Basic or Applied)
You conduct in-depth interviews with 15 students and ask them open-ended
questions such as: “How satisfied are you with your studies?”, “What is the
most positive aspect of your study program?” and “What can be done to
improve the study program?”
Based on the answers you get you can ask follow-up questions to clarify things.
You transcribe all interviews using transcription software and try to find
commonalities and patterns.
You survey 300 students at your university and ask them questions such as: “on
a scale from 1-5, how satisfied are your with your professors?”
Differences among Qualitative and Quantitative
Research Methods
Qualitative Research Quantitative research
Focuses on exploring ideas and Focuses on testing theories and
formulating a theory or hypothesis hypotheses
Analyzed by summarizing, categorizing Analyzed through math and statistical
and interpreting analysis
Mainly expressed in words Mainly expressed in numbers, graphs
and tables
Requires few respondents Requires many respondents
Open-ended questions Closed (multiple choice) questions
Key terms: understanding, context, Key terms: testing, measurement,
complexity, subjectivity objectivity, replicability
Types of research design can significantly vary according to your subject
and department.. There can be unique methods in medicine, psychology,
or education…
One-to-one interview
Ethnography
Focus groups
Case Study
Qualitative Research Methods
One-to-one interview:
Conducting in-depth interviews is one of the most common qualitative research methods.
It is a personal interview that is carried out with one respondent at a time. This is purely a
conversational method and invites opportunities to get details in depth from the
respondent.
One of the advantages of this method provides a great opportunity to gather precise data
about what people believe and what their motivations are. If the researcher is well
experienced asking the right questions can help him/her collect meaningful data. If they
should need more information the researchers should ask such follow up questions that
will help them collect more information.
These interviews can be performed face-to-face or on phone and usually can last between
half an hour to two hours or even more. When the in-depth interview is conducted face to
face it gives a better opportunity to read the body language of the respondents and match
the responses.
Qualitative Research Methods
Ethnography:
The researcher focuses on people and their behaviors in the natural environment. The
researcher involves, observes, engages, and embeds themself into the subject’s daily life
and routine.
The researchers experience the customs, traditions, and situations, etc., as a participant
or observer. Moreover, the researcher takes notes and describes their experience in the
descriptive form, so the reader also experiences the same culture.
It is a way of studying the culture of a group of people and describe in the form of
writing. This method aims to describe the cultural challenges in the same way as the
researcher observes.
Qualitative Research Methods
Focus groups:
The main aim of the focus group is to find answers to the “why” “what”
and “how” questions. One advantage of focus groups is, you don’t
necessarily need to interact with the group in person. Nowadays focus
groups can be sent an online survey on various devices and responses can
be collected at the click of a button.
Qualitative Research Methods
Case Study:
The case study method has evolved over the past few years and developed
into a valuable qual research method. As the name suggests it is used for
explaining an organization or an entity.
Cross-sectional surveys
Longitudinal surveys
Correlational Research
Casual-Comparative Research
Experimental Research
Quantitative Research Methods
Survey Research is the most elementary tool for all sorts of
quantitative research techniques. The most important purpose of the
research is to widely explain the characteristics of a particular group or
a bunch of population. This analysis is most typically employed by
both small and large organizations for a proper understanding of their
customers and to understand the merchandise and product views.
Cause and effect are not the basis of this type of observational research.
This sort of analysis is not restricted to the applied mathematics of two variables but
extends to analyzing different variables and groups.
Once analysis and conclusions are made, deciding about the causes should be done
fastidiously, as other different variables, each far-famed and unknown, might still have
an effect on the result.
Examples of this type of research include the effect of preschool attendance on social
maturity at the end of the first grade, the impact of drugs on a teenager.
Quantitative Research Methods
Experimental Research is usually based on one or more theories. It is
called true experimentation, uses the scientific method to establish the
cause-effect relationship among a group of variables that make up a study.
Multiple theories can be used to conduct this research. The components of
the experimental research design are prescribed below.
In prospective studies, individuals are followed over time and data about
them is collected as their characteristics or circumstances change. Birth
cohort studies are a good example of prospective studies.
Occurs when two events/measurements are correlated and the assumption is made
that one causes the other; however, there is a “lurking” variable that is actually
contributes to the occurrence of both events/measurements.
This fallacy occurs more often than the others in the media and horticulture. Applying a
fertilizer to the trees does not guarantee the high yield, while it is depending on many other
factors.
When two events are a result of bidirectional causation, one event causes another while the
other event causes the first. For example:
Example:
The number of lions in Kenya affects the number of gazelles in Kenya (lions eat gazelles). But it is
also true that the number of gazelles in Kenya affect the number of lions in Kenya (if lions don’t have
food, they will begin to die off).
This is called the predator/prey model.
Coincidence
Belief: XZ
Reality: YZ
Many times the fact that two events are correlated (linked) is
pure coincidence and there is no causal relationship that
exists between the two.
c) Reliability and Validity
Reliability and Validity
Systematic
True Score Error Random Error
More Reliable:
Less Reliable:
How do we determine whether a measure is
reliable?
Types of reliability
Test-retest
Internal Consistency
Split-half
Cronbach’s alpha: average of all possible split-half
reliabilities
Cronbach’s alpha
Number of items
High variation among individuals being tested
Clear instructions
Optimal testing situation
Validity
Denotes the extent to which an instrument is measuring
what it is supposed to measure.
How do we determine whether our
measures are valid?
Types of Validity
Content
Criterion
Construct
Content Validity
Whether the individual items of a test represent what you actually want to assess
Criterion-Related Validity
Simple random sampling: As the name indicates, simple random sampling is nothing but a
random selection of elements for a sample. This sampling technique is implemented where
the target population is considerably large.
Stratified random sampling: In the stratified random sampling method, a large population
is divided into groups (strata), and members of a sample are chosen randomly from these
strata. The various segregated strata should ideally not overlap one another.
Cluster sampling: Cluster sampling is a probability sampling method using which the main
segment is divided into clusters, usually using geographic and demographic segmentation
parameters.
Systematic sampling: Systematic sampling is a technique where the starting point of the
sample is chosen randomly, and all the other elements are chosen using a fixed interval. This
interval is calculated by dividing the population size by the target sample size.
There are 4 types of administrative errors, which may cause
problems on surveys.. And researcher should consider them:
Survey Methods
In-Home Fax
Interview
E-Mail
Computer- Outside…
Assisted
Telephone
Interviewing
Computer-Assisted Internet
Personal Interviewing
Telephone Surveys
ITEM Telephone Personnal Mail surveys
surveys interviews
Speed of data collection Very fast Fast Slow
Geographical flexibility High Moderate High
Respondent cooperation Poor Excellent Moderate
Versatility of questioning Moderate Quite High
Questionnaire length Short Mod-long Mod-long
Item nonresponse Medium Low High
Possibility of misunderstanding Average Lowest Less
Degree of interviewer influence on the Moderate High None-to-less
answer
Supervision of interviewers High Moderate None
Anonymity of respondent Moderate Low High
Ease of call back or follow-up Easy Difficult Easy
Costs Low-to-high High Moderate
Methods of Improving Response Rates
Prior
Incentives Follow-up
Notification
Monetary Nonmonetary
Prepaid Promised
e) Experimental Research designs
Experimental
Research
Designs
True- Quasi-
Pre-experimental Statistical
experimental Experimental
Solomon four
Statis group Factorial design
group
Pre-experimental design
Pre-Experimental Designs are the simplest form of
experimental research designs.
Pre-experimental designs have little or no control over
extraneous variables.
And, these designs do not randomly assign subjects to different
treatments.
The results of a test using a pre-experimental design are
difficult to interpret.
These designs are often used in testing television commercials
because they are simple and relatively inexpensive.
Pre-experimental design
One-Shot Case Studies: With a one-shot case study, test units—
people, test markets, etc.
Interaction(s)
Error(s)
Total
The Completely Randomized design
The area should be % 100 homogeneous to do this
design
Effect
Example of a Randomized Complete Block Design
with three blocks, and seven treatments for one
replicate randomized within each block (thus n=3).
A B C D Sources of Variance
Treatment
B C D A Row
Column
C D A B
Error
D A B C Total
Factorial designs
Factorial design involves having more than one independent variable, or factor, in a study.
Factorial designs allow researchers to look at how multiple factors affect a dependent
variable, both independently and together.
Simple Factorial
Example; 3 conditions of A and 4 conditions of B
1. Replication 2. Replication
A2 – B2 A0 – B0 Sources of Degrees of
A1 – B1 A1 – B3 Variance Fredom
A0 – B0 A1 – B1
Replication 2-1=1
A0 – B1 A2 – B2
Factor A 3-1=2
A2 – B0 A0 – B2
A1 – B3 A2 – B0 Factor B 4-1=3
A0 – B2 A1 – B0 Interaction 2 (Fac. A) x (Fac.
A2 – B1 A0 – B3 B) 2 =6
A0 – B3 A2 – B1 Error 24-6-3-2-1=12
A1 – B2 A0 – B1 Total 24
A1 – B0 A2 – B3
A2 – B3 A1 – B2
Split Plots
1. Replication 2. Replication
B0 B0
Sources of Degrees of
B1 B2 Variance Fredom
A0 A2
B2 B1
B3 B3 Replication 2-1=1
B2 B3 Factor A 3-1=2
B0 B2 Error-1 24-6-2-1=13
A1 A0
B1 B0 Factor B 4-1=3
B3 B1 Error-2 24-3-3-1=12
B1 B3 Interaction 3x2=6
B3 B0
A2 A1 Total 24
B2 B2
B0 B1
Thanks
Q/A