Data Collection Use Cases:
Definition: Data collection is the process of Market research
gathering and measuring information on Customer satisfaction
variables of interest in a systematic fashion, Public opinion polls
enabling one to answer research questions, test Surveys/Questionnaires
hypotheses, and evaluate outcomes. Clear and simple
Purpose: To obtain accurate and reliable data to Relevant
make informed decisions or conclusions in Neutral
statistical analysis. Logical Flow
Appropriate Question Type
2 Types of Data Collection Allow for Ethical Considerations
Primary Data Collection: Data collected directly Pretest/ Pilot
from original sources for the first time. Ease of Response
Examples: Interviews, Surveys/ Questionnaires, Allow Anonymity and Confidentiality
Registration, Experimental, Observation, and Be Culturally Sensitive
Texting methods.
Length Should be Appropriate
Secondary Data Collection: Data collected from
ADVANTAGES
existing sources.
Examples: Government Reports, Research One of the easiest methods.
Articles, Online Databases, and Historical Can reach a large audience quickly.
Records. Easy to analyze, especially with closed-
ended questions.
Interviews DISADVANTAGES
A qualitative data collection method Response bias.
involving direct, face-to-face or virtual Limited depth in responses.
conversation between the interviewer Cannot be accomplished by “illiterates”.
and the respondent.
Types: Registration
Structured: Pre-determined questions. Recording information about a
Semi-Structured: Guided by a set of
population or specific events through
questions but allows flexibility.
official records, forms, or registries. This
Unstructured: Open-ended,
conversational style. method is often used for collecting data
ADVANTAGES that requires ongoing, accurate, and
In-depth information. consistent recording over time.
Opportunity to clarify and probe Key Features:
responses. Ongoing Process: Involves continuous or
DISADVANTAGES periodic data collection, such as birth,
Time-consuming. death, marriage, and divorce records,
Potential for interviewer bias. voter registrations, and business
Surveys/Questionnaires Structured licenses.
forms used to collect data from a large Official Records: Gathered through legal
number of respondents. Qualitative or or official means, often mandated by
quantitative may be collected. law; governments and organizations
maintain these records as part of their
operations.
Standardized Forms: The information is The group that does not receive the
usually collected using standardized treatment; the group that is exposed to
forms or formats, ensuring consistency a standard condition. This group serves
and reliability in the data. as a baseline to compare against the
Large-Scale: Registration systems are experimental group.
often large-scale, covering entire TREATMENT GROUP
populations or sectors, and they allow The group that receives the treatment
for longitudinal studies where data is or exposure to the independent
tracked over time. variable.
ADVANTAGES ADVANTAGES
Because mandatory and legally Ability to establish cause-and- effect
required, the data is generally accurate relationships.
and reliable. High level of control over variables.
Comprehensive data coverage covering Precise results.
entire populations or specific groups. DISADVANTAGES
Time efficiency if a registration system is Can be costly and time consuming.
in place. Ethical concerns in some experimental
DISADVANTAGES designs.
Potential for Underreporting if Participants may change their behavior
individuals or organizations fail to because they know they are part of the
register. experiment.
Legal and privacy concerns
Delays in data availability due to the Observation
bureaucratic nature of registration the researcher systematically watches
processes. and records behaviors, actions, events,
or conditions as they occur in their
Experimental natural setting. This method is widely
The researcher manipulates one or used in fields such as anthropology,
more variables (independent variables) sociology, psychology, and education to
to observe their effect on another gather detailed and firsthand
variable (dependent variable). This information.
method is widely used in scientific Key Characteristics:
research, particularly in fields such as Directness: Unlike other methods, such
psychology, medicine, and the natural as surveys or interviews, observation
sciences, to establish cause-and- effect involves directly seeing and recording
relationships. what happens, reducing the risk of self-
Types: report biases.
Laboratory Experiments: Conducted in Non-Intrusive: Observational methods
controlled, indoor environments. can be unobtrusive, especially in
Field Experiments: Conducted in real- naturalistic observation, where the
world settings. observer tries not to interfere with or
influence the behavior being observed.
CONTROL GROUP
ADVANTAGES
Allows researchers to capture nuanced Market Research: To quickly gather
details and the context in which feedback on products or services.
behaviors occur, providing a deep Public Health: To collect health-related
understanding of the subject. data, especially in rural or
Offers insights into how people interact underdeveloped areas.
with their environment and with each Social Research: To collect data on social
other, often revealing dynamics that behaviors, opinions, or demographic
would be missed with other methods. information.
It can be more accurate in capturing How it works:
actual behavior since observation does 1. A survey or data collection tool is designed
not rely on participants’ self-reporting, specifically for text messaging. The questions
need to be concise due to character limits in
DISADVANTAGES text messages.
The researcher’s expectations or 2. The survey is sent out to participants via text
perspectives can influence what they message- can be done manually or through
notice and record, leading to biased automated systems that can handle large-scale
data collection. distributions.
Observation can be a slow process, 3. Participants respond directly to the text
researcher spends extended periods in messages. Their responses are typically stored
the field to gather sufficient data. in a database for later analysis.
Because observations often focus on 4. If necessary, follow-up messages can be sent
specific settings or groups, the findings to gather additional data or clarify previous
may not be easily generalizable to other responses.
contexts or populations.
Covert observation and recording Examples:
people without their consent can raise 1. Do you feel sick today? Reply with YES or NO.
significant ethical issues 2. Rate your stress level on a scale of 1 (low) to
When participants know they are being 5 (high).
observed (overt observation), they 3. Rate your satisfaction with our service on a
might change their behavior, known as scale of 1 to 5.
the Hawthorne effect, which can skew 4. Have you tried our new product? Reply with
results. YES or NO.
5. Which issue concerns you most? Reply with
Texting JOBS, HEALTH, or EDUCATION."
Involves gathering data through SMS 6. How safe do you feel in your neighborhood?
(Short Message Service) or other text- Reply with SAFE, UNSAFE, or UNSURE.
based messaging platforms. This 7. Did you enjoy the event? Reply with YES or
approach is increasingly popular, NO.
especially in areas where internet 8. How would you rate the event on a scale of 1
access may be limited but mobile phone to 5?
usage is widespread.
Applications: Format considerations:
Instructions: Keep instructions clear and Population refers to the entire group of
simple. For example, “Reply with YES or individuals or items that you're
NO” or “Reply with a number”. interested in studying. It includes all
Conciseness: Limit the number of possible subjects that fit a particular set
questions to avoid participant fatigue. of criteria.
Each message should be easy to read For example, if you're studying the
and respond to quickly. eating habits of teenagers in a specific
Response Options: Provide specific country, the population would include
response options to minimize confusion every teenager in that country.
and ensure that data is easy to analyze. Sample is a subset of the population
ADVANTAGES that is selected for the actual study.
High reach Instead of examining the entire
Cost-effective population, which is often impractical or
Quick turnaround impossible, researchers study a sample
Ease of use to draw conclusions or make inferences
about the population as a whole.
DISADVANTAGES For example, instead of surveying every
Limited data depth teenager in a country, a researcher
Response bias might survey a sample of 1,000
Technical issues teenagers and use the results to make
Privacy concerns generalizations about the entire
Population and Sample population.
Sampling
The process of selecting a subset (called
a sample) from a larger group (called a
population) to make inferences about
the population. Since it is often
impractical or impossible to collect data
from an entire population, sampling
allows researchers to study a smaller
group that represents the population.
The sample should be chosen in such a
way that it accurately reflects the
characteristics of the population,
ensuring that the conclusions drawn
from the sample are valid for the entire
population.
Slovin’s Formula
Used in statistics to determine the
sample size needed for a survey or
study when the population size is
known.
It is particularly useful in cases where
the researcher wants to ensure that the
sample accurately reflects the
population while keeping the sample
size manageable.
