Topic: INTRODUCTION OF DATA COLLECTION AND Methods for Quantitative Data Collection

METHODS / TECHNIQUES OF DATA COLLECTION 1. Surveys/Questionnaires

➢ Defined as objectives or closed-ended questions used to gain
DATA
detailed insights from respondents about a certain research topic.
➢ Data is various kinds of information formatted in a particular
These questions form the core of a survey and are used to gather
way.
numerical data to determine statistical results.
➢ Is a collection of facts such as numbers, word, measurements,
observation or just description of things. a. Online Survey (Google Forms, Survey Monkey, QuestionPro,
Jotform, Amazon MTurk, SurveyGizmo, Qulatrics, Google
DATA COLLECTION
Consumer Surveys).
➢ Is the process of gathering, measuring, and analyzing accurate
b. Offline Survey (In-person Paper-pencil survey, Mail survey).
data from a variety of relevant sources to find answers to research
problems, answer questions, evaluate outcomes, and forecast 2. Interviews
trends and probabilities. ➢ Interviews are another popular way of researching polling a
population. They can be thought as a survey but in verbal, in-
Two General/Sources of Data
person, or virtual face-to-face format. The online format of
Primary Source of Data Collection
interviews is becoming more popular nowadays, as it is cheaper
➢ As the name implies, this is original, first-hand data collected by
and logistically easier to organize them face-to-face interviews,
the data researcher. This process is the initial information
yet still allows the interviewer to see the respondent.
gathering step, performed before anyone carries out any further
➢ An interviewer runs through a survey with the respondents,
or related research. Primary data results are highly accurate
asking mainly closed-ended questions (yes or no, multiple choice
provided the researcher collects the information. However,
questions, or questions with rating scale) that ask how strongly
potentially time-consuming, and expensive.
the respondents agrees with the statements.
Methods on primary data collection
3. Observation
• Interviews ➢ The researcher asks questions of a large sampling
➢ Is a technique that focuses on recording the number or types of
of people, either by direct interviews or means of mass
people who do a certain action such as choosing a specific product
communication such as by phone or email. This method is by far
from a grocery shelf, speaking to a company representative at an
the most common means of the data gathering.
event, or how many people pass through a certain are within a
• Focus Group ➢ Focus group, like interviews are a commonly given time frame.
used technique. The group consists of anywhere from a half dozen ➢ Observation studies in quantitative research are similar in
to a dozen people, led by a moderator, brought together to nature to a qualitative ethnographic study (in which a researcher
discuss the issue. also observes consumers in their natural habitats), yet observation
studies for quantitative research remain focused on the numbers
• Questionnaire ➢ Are a simple straight forward data collection – how many people do an action, how many of a products
method. Respondents get a series question, either open or close- consumer pick up etc.
ended, related to the matter at hand.
4. Review Existing Documents
Secondary Data Collection ➢ Reviewing existing research to see how it can contribute to
➢ Unlike primary data collection, there are no specific collection understanding a new issue in question.
methods. Instead, since the information has already been ➢ There are numerous documents that can be analyzed to
collected, the researcher consults various data sources such as: support primary data or used as an end in themselves. Secondary
Journal articles, Websites, Full-blown research proper, data collection can include reviewing public records, governments
Government publications, Books and Internal record. research, company data bases, existing reports, paid for research
publication, magazines, journals, case studies, websites, and
➢ It is a second-hand data which is based in tried and tested data, more.
which is previously analyzed, filtered and already available. ➢ Aside from using secondary research alone, document review
Two Categories for Primary Data Collection Methods Qualitative can also be used in anticipation of primary research, to
Methods understand which knowledge gaps need to be filled and to nail
➢ It is the process of collecting and analyzing non-numerical data down the issues that might be important to explore in a primary
(e.g., text, video, or audio) to understand concepts, opinion, or research study.
experiences. It can be used to gather in-depth insights into a Why data collection is necessary / important?
problem or generate new ideas for research. ➢ Data collection plays a significant asset in any research study.
Quantitative Methods Research work caried can vary from field to field but all the
➢ Refers to the collection of numerical data that can be analyzed research work based on data which is analyzed and interrupted to
using statistical method. get information. Data is one of the vital resources for all research
➢ This type of data collection is often used in surveys, experiment, since it drawn conclusions based on the results of statistical
and other research methods. ➢ It analysis. The quality of data collection methods improved the
measures variables and establish relationships between variables. accuracy or validity of study outcomes or findings.
➢ The data collection through quantitative methods is typically in
the form of numbers, such as response frequencies, means, and
standard deviation, and can be analyzed using statistical software. Module 1
➢ It answered the “what and who” questions rather than “why”. Fundamental Concepts of Statistics
What is Statistics? describe the amount or number of something. This variable can be
Statistics is the science deals with the methods of collecting, categorized as either a discrete quantitative variable or a
organizing, summarizing, presenting, analyzing, and interpreting continuous quantitative variable.
quantitative data to support the creation of a more effective, Examples. Number of Students, Length of time to solve a
efficient, reliable, and valid decisions in a particular problem. statistical problem, grades in Statistics subject, monthly salary.

Two Areas of Studying Statistics Discrete Variable

A variable that can only assume certain values and there are
Descriptive Statistics usually “gaps” between values. Typically, discrete variables result
These are statistical methods that organize, summarize, and from counting (Abocejo & Pañares, 2014).
describe data, providing an organized visual presentation of the Examples. The number of students in the family, number of newly
data collected without drawing generalization (inferences) about a structured school buildings in EVSU Tanauan, number of COVID-19
large group or population data. Ex. Measures of central tendencies deaths recorded.
(mean, median mode) and measures of variability (range,
interquartile range, variance, semi-quartile range, and standard Continuous Variable
deviation). A variable that can be derived from a measuring process. Which
we can assume any values within a specific range of values.
Inferential Statistics Examples: minutes remaining in class, student’s height, patient's
These are statistical techniques used to estimate or predict a body temperature, student's general weighted average in
population parameter from a sample statistic. It includes making a Statistics.
decision, estimate, prediction, or generalization about a
population based on a sample. The idea is we study the Other Variables
characteristics of a sample data set and out of the results we have These are variables commonly utilize in doing a research project.
derived based on the sample allows us to make a conclusion and You need to be familiarized with the distinction and uses of each
generalization about the population where the sample data are variable.
taken.
Constant variable is any quantity or number whose value does not
The term population is referring to a complete set of all possible change.
elements, individuals, objects, observation, or measurements of
interest. It is a group of individuals/subjects that comprise the Depended Variable is a variable that changed because of the
same characteristics. independent variable. In other words, it is the response to the
change in the independent variable. Also called as the Outcome of
Example. The total number of students studying in a school or an experiment.
college, the total number of books in a library, the total number of
houses in a village or town are some examples of the population. Independent Variable is a variable that is not affected by anything
that is set or manipulated by the researcher.
The term sample is referring to a portion, part, or a subset of the
population of interest that is chosen for analysis. It is a data set of Intervening Variable is a hypothetical variable used to explain
interest which the investigators or researcher plans to study for causal links between other variables.
the objective of making a generalization about the population. Control Variable is a variable that the researcher wants to remain
Also, the terms Parameter and Statistic are key terms need to constant and unchanging. Commonly used when doing an
familiarize when study statistics. A parameter is a summary experimental study.
measure used to describe the characteristics of a population. A Nuisance Variable is an extraneous variable that increases
value calculated using the data from a population. Whereas variability overall.
Statistic is a summary measure used to describe the
characteristics of a sample. A value calculated using the data from Moderator Variable is a moderator is a qualitative (e.g., sex, race,
a sample such as mean, median, standard deviation, variance, etc. class) or quantitative (e.g., level of reward) variable that affects
A statistic is a datum (singular form of data) that can be the direction and/or strength of the relation between an
represented numerically. The collection of more than one figure is independent or predictor variable and a dependent or criterion
called statistics [plural]. variable (Baron & Kenny, 1986).

What is Variable? Mediator Variable is a given variable may be said to function as a

A variable is a condition or characteristic that the researcher mediator to the extent that it accounts for the relation between
manipulates, controls, or observes. Example: Age, sex, business the predictor and the criterion (Baron & Kenny, 1986).
income and expenses, country of birth, capital expenditure, class
grades, eye color, and vehicle type are examples of variables. In a correlational study, the Moderator variable that influences
the strength of a relationship between two other variables, while
Qualitative Variable the Mediator variable is one that explains the relationship
It is a general category for any variable no numerical between the two other variables.
characteristics, and it is gathered through categorical responses.
Simply describe the quality or characteristics of something.
Examples. Gender, religious affiliation, class sections, place of
birth, hair color, eye color.

Quantitative Variable
It is a general category of any variable that can be counted,
measured, or has a numerical value associated with it. Simply
Types of Data (Abocejo & Pañares, 2014) Ratio Scale The
Most of the students commit a mistake by using the word “data” ratio scale is the highest-level data measurement that has all the
as a singular noun whenever they use and apply it in their thesis characteristics of the interval scale data measurement. Unlike to
manuscripts. The word “datum” is the singular form and “data” is interval scale, it has true zero which indicates a total absence of
the plural form. So, a datum is one information while data are the measurement or property being measured. Also, it permits
much information, known facts, figures, observations, statistics, comparisons such as being twice as high. The ratio between the
records, and reports, among others. They can be classified in two data values is meaningful.
different ways.
Parametric and Non-Parametric Statistical Measurements
Ungrouped Data Parametric Statistical Measurements These are measurements
An ungrouped data is given as raw, individual data points, which hold the following assumptions about the population
unorganized information. parameters and characteristics in order to be used for data
Example. Examination scores of 12 education students on a 10- analysis:
item test. 9, 9, 9, 8, 10, 10, 7, 8, 7, 5, 9, 10 5.1.3.2
The population data where the sample data set are taken to be
Grouped Data analyzed is normally distributed.
Grouped data are data formed by aggregating individual  The level of measurements of data/variables are in interval or
observations or data values of a variable into groups so that a ratio level of measurements; *Note: The computation of the
frequency distribution formed out of these groups serves as a sample size may also depend on the complexity of the statistical
convenient means of summarizing or analyzing the data. test to be used.
Example.  The sample size computed is well representative of the
Age Frequency population and the selection process of samples is random.
22 20  Parametric tests are applicable only for variables; and
21 30  Parametric tests are powerful if it exists.
20 15
19 15 Non-Parametric Statistical Measurements

 It is distribution-free;
 The level of measurements of data/variables are in interval or
ratio level of measurements;
*Note: The computation of the sample size may also depend on
the complexity of the statistical test to be used.
 The sample size computed is well representative of the
population and the selection process of samples is random;
 It is less powerful compare with Parametric tests.
 Can be used for nominal or ordinal scales of data/variables
measurements;
 It can be applied to both variables and attributes.

Variable/Data Level of Measurements

Nominal Scale The

nominal scale is the lowest level of data measurement. It is merely
used for classification or identification purposes and no
measurements of data are involved, only counting. There is an
ordering of categories, and the sizes of the categories have no
meaning. These data categories are mutually exclusive and
exhaustive, which means a value of observation contains one and
only one category.

Ordinal Scale The

ordinal scale is a level of data measurement that has the
characteristics of a nominal scale but this time the data/variables
have naturally ordered categories. The order of measurements is
possible, however, the distance between data values cannot be Sample Size and Sampling Design in Research
determined or meaningless.
STRATEGIES FOR DETERMINING SAMPLE SIZE
Interval Scale The 1. Census or Complete Enumeration (Small Population)
Interval scale is a data level of measurement that possesses all the 2. Through Literature or Published Tables of Sample Size selection
characteristics of the ordinal scale level of measurement. It is 3. Adopting the Sample Size of similar Study.
possible to differentiate between any two classes based on the 4. Sample Size Determination Formula - Level of Confidence -
degree of differences. Distances between any two points are of Degree of Variability of Subjects - Margin of Precision or Error
known size, the unit of measurement is constant (but arbitrary), 5. Sample Size Calculator/Software
and the zero points is arbitrary (there is no natural zero point). 6. Sample size recommendation based on somewhat with
Besides, with interval measurements, operations of addition, and authority.
subtraction have meaning.
Sample Size Computation Criteria
Level of Confidence
Level of Precision or Margin of Error (Sampling error)
Degree of Variability of Subjects Sampling
Sample Size Determination Formula  Sampling is a process of selecting a sample from a population.
 Sampling is the process of choosing individual members or a
Level of Confidence portion of the population in order to draw statistical conclusions
The key idea encompassed in the Central Limit Theorem is that and estimate the characteristics of the entire population.
when a population is repeatedly sampled, the average value of the
attribute obtained by those samples is equal to the true Two General Methods of Sampling
population value.  Probability sampling involves random selection, allowing you to
*Widely Used Confidence Level 90%, 95%, & 99% make strong statistical inferences about the whole group.
 Non-probability sampling involves non-random selection based
Margin of Precision or Error on convenience or other criteria, allowing you to easily collect
- It is the range which the true value of the population is estimated data.
to be. This range is often presented in percentage form (e.g. ± 5%)
Probability Sampling
Degree of Variability of attributes of Being Measured Simple Random Sampling
refers to the distribution of attributes in the population. The more In a simple random sample, each member/subject of the target
heterogeneous a population, the larger the sample size required population has equal chance of being selected or chosen to be
to obtain a given level of precision. The less variable (more part of the target number of samples. Your sampling frame
homogeneous) a population, the smaller the sample size. (Complete lists of individual or elements—depends on the subject
*Commonly used percent of variability is: 20%, 80%, 50% of interest) should include the whole population.
Common techniques: Lottery technique, Fish ball Technique
Sample Size Determination Formula for Finite Population
(Cochran, 1977) Systematic Random Sampling
Using systematic random sampling, researchers select sample
members from a population at regular intervals. It involves the
selection of a starting point for the sample and a repeatable
sample size at regular intervals.

Stratified random sampling

is a technique in which the researcher divides the population into
subgroups (strata) that do not overlap but are representative of
the full population. During sampling, it is possible to organize
these groups and then draw samples from each group
independently.

Cluster Random sampling

also entails splitting the population into subgroups, but each
subgroup should have characteristics that are similar to those of
the entire sample. Instead of randomly selecting individuals from
each subgroup, entire subgroups are selected.

Multi-Stage Random Sampling

Using combination of random sampling designs in order to obtain
needed samples based on a large population.
 clothes worn by participants and site of data collection (Dearnley,
2005).
Non-Random Sampling
A novice researcher may not feel confident in conducting
1. Convenience Sampling - This type of sampling, researchers qualitative interviews with either limited or no experience in
prefer participants as per their own convenience. The researcher qualitative research (Dearnley, 2005; Hoskins & White, 2013).
selects the closest live persons as respondents. In convenience Also, a novice researcher may feel isolated from peers and other
sampling, subjects who are readily accessible or available to the qualitative researchers with the length of time of conducting
researcher are selected. interviews and possible travel during data collection (Dearnley,
2. Purposive Sampling - In this type of sampling, the researcher 2005). Additionally, a researcher faces challenges on how to
chooses the participants as per his/her own judgment, keeping choose participants, how to locate possible participants, how to
back in mind the purpose of the study. It uses the judgment of an convince them to participate, and then making them comfortable
expert in selecting cases or it selects cases with a specific purpose to talk in the interview (Dearnley, 2005; Hoskins & White, 2013).
in mind. More specific challenges occur with vulnerable populations and
dealing with sensitive topics such as sexuality in health research
3. Quota Sampling - You select your sample according to some (Ashton, 2013; Tarzia, Bauer, Fetherstonhaugh, & Nay, 2013).
fixed quota This type of sampling is somehow related to stratified These challenges related to the researcher’s perceptions about
sampling. Make sure that the sample represents each group or data collection are inclusive of early career researchers.
stratum of the population. Unlike the stratified sampling, the
researcher in quota sampling method selects the subjects PARTICIPANTS
available immediately fulfilling the criteria. Second, participants’ perceptions during data collection
4. Snowball Sampling - Also called "chain referral sampling,” in were also identified. Participants reported feeling uncomfortable
this method, the sample is actually collected in various stages. This with the researcher during data collection (Bonevski et al., 2014).
method is appropriate when the members of a special population Examples of researcher-related discomfort during data collection
are difficult to locate. It begins by the collection of data from one included: 1.
or more contacts usually known to the person collecting the data. concern whether health information would remain confidential
At the end of the data collection process (e.g., questionnaire, (Bonevski et al., 2014);
survey, or interview), the data collector asks the respondent to 2. being hungry or thirsty prior to data collection (Dearnley, 2005;
provide contact information for other potential respondents. Easton et al., 2000);
These potential respondents are contacted, interviewed and 3. perceived anxiety (Ashton, 2014; Mansell, Bennett, Northway,
further asked to provide more contacts. This process goes on till Mead, & Moseley, 2004; Smithson, 2000).
the purpose of the researcher is achieved. Participants may have concern about the confidentiality of their
Course: Quantitative Methods/Elementary Statistics health information if they are revealing personal information to
Topic: Issues encountered in data collection the researcher in the interview (Bonevski et al., 2014). Participants
may feel hungry or thirsty if they were sick, could not afford to
Data Collection Challenges buy food to eat that day, skipped eating breakfast, or at the clinic
Our search of data collection challenges in the literature were before lunchtime (Dearnley, 2005; Easton et al., 2000).
organized into the following categories: Participants may have perceived anxiety due to the location of the
1. researcher, interview, uneasiness or lack of rapport with the researcher, or
2. participant, stressors in their daily life (Ashton, 2014; Mansell et al., 2004;
3. data collection environment, and Smithson, 2000). Also, participants may have other concerns
4. interview guide design. beyond researcher-related discomfort. Participants with low
health literacy reported misunderstanding questions in surveys or
Manuscripts were found in the nursing, public health, social interviews (Bonevski et al., 2012).
science, and methodology literature (Bonevski et al., 2014;
Bournet & Robson, 2015; Dearnley, 2005; Easton, McComish, & DATA COLLECTION ENVIRONMENT
Greenberg, 2000; Hebert, Loxton, Bateson, Weisberg, & Locke,
2013; Johnson & Clarke, 2003). A third issue prevalent in the literature included
challenges related to the data collection environment. One
RESEARCHER example is the room selection for data collection. Selection of the
interview room in a neutral location, was a challenge in research
First, the authors identified data collection challenges studies (Dearnley, 2005; Easton et al., 2000). However, data
related to the researcher. Data collection challenges have collection may not take place in a room, but an open space or
included: public environment. Other environmental issues during the
1. participants who were resistant to participate; interview included loud background noises from another room
2. how to dress for an interview, such as wearing formal or and a cold or warm room temperature (Dearnley, 2005; Easton et
informal clothes; al., 2000). Also, the presence or nearness of family members or
3. lack of experience conducting qualitative interviews; and significant others is a potential issue related to the data collection
4. feelings of isolation from peers and other researchers during environment (Haahr, Norlyk, & Hall, 2014).
data collection (Dearnley, 2005; Hoskins & White, 2013; Johnson
& Clarke, 2003). Novice researchers may have difficulty INTERVIEW GUIDE DESIGN
establishing rapport with participants unlike them and encounter
participants who are resistant to participate in the interview Fourth, data collection challenges related to the
(Hoskins & White, 2013). A researcher may also be unsure to dress interview design and conducting interviews were identified in the
formally or casually in the interview depending on the type of literature. Challenges in qualitative data collection are the process
of designing an interview guide, how to formulate questions in an Sensitive Information
interview for the participant and staying focused on the research ➢ Another challenge we encountered was the participants
topics during the interviews (Ashton, 2014; Hoskins & White, comfort level during data collection. For Martinez, Davis, and Parr,
2013). Previous research reported challenges of preparing for all three studies focused on sexual experiences and attitudes
interviews on sensitive topics, writing interview questions on towards sex. In Martinez’s study, early adolescent participants
sensitive information such as end of life care, sexual activity, or (aged 12-13 years) expressed embarrassment when probed about
contraceptive use (Ashton, 2014; Bourne & Robson, 2015; Hebert sexual health and romantic relationships; she noted participants
et al., 2013). appearing uncomfortable with the topic, which halted the
interview.
The authors identified a number of data collection
challenges for dissertation research studies. While the authors of Sampling error
this manuscript conducted mixed methods, quantitative and ➢ A sampling error is the difference between a population
qualitative studies, the challenges we present focus on what is parameter and a sample statistic.
relevant for qualitative studies. The data collection challenges are ➢ As a result, the results found in the sample do not represent the
reported below under the following themes: results that would be obtained from the entire population.
1) location,
2) health literacy and language of data collection instrument, • A sampling error occurs when the sample used in the study is
3) duration of data collection, not representative of the whole population.
4) researcher fatigue, and • Sampling is an analysis performed by selecting a number of
5) sensitive information. observations from a larger population.
• Even randomized samples will have some degree of sampling
Location error because a sample is only an approximation of the population
➢ Is a critical component of the of data collection process (Gill, from which it is drawn.
Stewart, Treasure, & Chadwick, 2008; National Science • The prevalence of sampling errors can be reduced by increasing
Foundation, 1997). The location influenced data collection in three the sample size.
of the studies, one study in a school setting and two in a clinical • In general, sampling errors can be placed into four categories:
setting. In all three cases (Christiana, Namageyo-Funa, & population-specific error, selection error, sample frame error, or
Rimando), the data collection method was interviews. non-response error.

➢ The purpose of Christiana’s study was to examine youths’ Missing data

participation in noncompetitive outdoor physical activity. ➢ Missing data are problematic because, depending on the type,
Christiana conducted group and one-on-one interviews with they can sometimes cause sampling bias. This means your results
middle school students in the school’s main administration office’s may not be generalizable outside of your study because your data
copy/fax room. come from an unrepresentative sample.

Health Literacy and Language of Data Collection How to prevent missing data
➢ Participants’ health literacy and language of interview questions ➢ Missing data often come from attrition bias, nonresponse, or
and survey instrument can impact data collection. Participants’ poorly designed research protocols. When designing your study,
low health literacy may influence their understanding of questions it’s good practice to make it easy for your participants to provide
in a data collection instrument consequently impacting the data data.
collection process (Bonevski et al., 2014; Mayer & Villaire, 2007).
Martinez and Rimando both noted that low literacy levels may Here are some tips to help you minimize missing data:
have impacted data collection. • Limit the number of follow-ups
• Minimize the amount of data collected
Duration of Data Collection • Make data collection forms user friendly
➢ The data collection process can be impacted by the length of • Use data validation techniques
the data collection instrument or by how long a participant will be • Offer incentives
engaged in the process of providing data. Pilot testing the data
collection instrument can allow the researcher to determine how Questionnaire design
long participants will be engaged in data collection so they can be ➢ Is the process of designing the format and questions in the
informed of the time commitment. This can increase participant survey instrument that will be used to collect data about a
comfort in the data collection process, otherwise participants may particular phenomenon. In designing a questionnaire, all the
feel uncomfortable during data collection and not want to various stages of survey design and implementation should be
complete the survey questions or need assistance completing the considered.
survey. These include the following nine elements:
Researcher Fatigue 1. determination of goals, objectives, and research questions;
➢ In addition to factors discussed previously that cause 2. definition of key concepts;
participant fatigue, it is important to consider the fatigue that a 3. generation of hypotheses and proposed relationships;
researcher may experience. Completing a survey may be mentally 4. choice of survey mode (mail, telephone, face-to-face, Internet);
taxing for the participant, while conducting focus groups and 5. question construction;
interviews caused fatigue on the researcher collecting the data. 6. sampling;
Researcher fatigue can influence the information gathered during 7. questionnaire administration and data collection;
research. Researcher fatigue is a crucial component to the flow 8. data summarization and analysis;
and success of a focus group and interviews (Dickson-Swift, James, 9. conclusions and communication of results.
Kippen, & Liamputtong, 2007; Fern, 1982).
COURSE: QUANTITATIVE METHODS TOPIC: DESIGNING SURVEY generation
QUESTIONNAIRE (for research) – Things to consider 2.2.What are the differences in daily calorie intake
between Filipino women and men?
SURVEY QUESTIONNAIRE Dependent Variable: Calorie intake
➢ Are defined as objective questions used to gain detailed insights Comparison Groups: Filipino men and women
from respondents about a survey research topic. 2.3.What is the difference in food preferences in Filipino
➢ These answers received for those quantitative survey questions children and adults?
are analyzed and a research report is generated on the basis of Dependent Variable: Food preference
this quantitative data. Comparison groups: Filipino children and adults
➢ These questions form the core of a survey and are used to
gather numerical data to determine statistical results. ✓ The various groups mentioned in the above-mentioned option
➢ Feedback received to quantitative survey questions is related indicate independent variables. These independent variables
to, measured by or measuring a “quantity” or statistic and not the could be based on gender, ethnicity or education. It is the
“quality” of the parameter. dependable variable that determines the complexity of
comparative survey questions. 3. Relationship Survey Questions
Types of quantitative survey questions Are used to understand the association, trends, and causal
Quantitative survey questions should be such that they offer relationship between two or more variables. When discussing
respondents a medium to answer accurately on the basis of this research topics, the term relationship/casual survey questions
factor, quantitative survey questions are divided into three types: should be carefully used since it is a widely used type of research
1. Descriptive Survey Questions design, i.e. experimental research-where the cause and effect
- Are used to gain information about a variable or multiple between two or more variables. These questions start with “what
variables to associate a quantity to the variable. is the relationship [between or amongst] followed by a string of
- It is the simplest type of quantitative survey questions and helps independent [gender or ethnicity] and dependent variables
researchers in qualifying the variables by surveying a large sample [career, political beliefs etc.]?”
of their target objective. Examples:
- Most widely implemented descriptive questions start with “what
is this…”, “how much…”, “what percentage of…” and such similar 3.1. What is the relationship between age and usage of
questions. gadgets in EVSU-TC?
Dependent Variable: Usage of gadgets
Examples: Independent Variable: Age
1.1. What is the most preferred food of the Filipinos? Relationship group: EVSU-TC
Variable: Food 3.2. Do grades have an impact on the future success of
Target group: Filipinos BSIT students?
1.2. What are the facets that transform VSU student Dependent Variable: Future success
thoughts and motivate them to join EVSU University? Independent Variable: Grades
Variable: facets that transform career decisions Relationship group: BSIT students
Target group: VSU students 3.3. What is the relationship between college students’
1.3. How many Filipino citizens choose to leave the study time and their scores in exams?
country to find a better avenues? Dependent Variable: Scores in Exams
Variable: Number of citizens looking for better Independent Variable: Study time
opportunities Relationship group: College students
Target group: Filipino citizens What is a good questionnaire?
✓ In every example mentioned above, researchers should focus A good questionnaire should be valid, reliable, clear, interesting
on quantifying the variable. The only factor that changes is the and succinct.
parameter of measurement. They are different quantitative
sample problem which need to be measured by different Valid – a valid questionnaire should ask what it intends to ask, i.e.
parameters. the questions should be phrased in such a way that the
respondent understands, the objective of the questions. To
2. Comparative Survey Questions achieve this, the questionnaire should be reviewed by the
- Are used to obtain a comparison between 2 groups of variables. “content expert” during the pilot test. (e.g. if the target
The aim is to establish a clear difference between one or more respondents are a diabetic patient, then a diabetic patient should
variables belonging to different target groups. comment whether he understands the questionnaire). Any
- The questions usually begin with “what is the difference in uncertainties and queries should be clarified till the question is
[dependable variable] between [two or more variables]?” This clearly understand.
question will be enough to realize that the main objective of
comparative questions is to form a comparative relationship Reliable – a reliable questionnaire should yield the same answer if
between the groups under consideration. the same question is posed to the respondent repeatedly in a
short span of time. This can be achieved by performing a “test-
Examples: retest” i.e. administer the same questionnaire to the respondent a
2.1.What are the differences in the usage of mobile second time and check for consistency of the answer. Any
between young and older people? discrepancy in the answers could be due to lack of clarity of the
Dependent Variable: Usage of mobile devices questions and this should be reviewed and rephrased.
Comparison groups: Young generation and Older
Interesting – an interesting questionnaire is more likely to be easily accessible data from surveys shared from common account
complete by the respondent and hence yields a better response thereby compromising confidentiality. Most of these tools rely on
rate. This requires the researcher to put some thoughts into the researchers’ ingenuity in setting up the survey settings to limit
asking questions that are relevant to the respondents and in a for instance participants’ IP address or access to their own
logical sequence. personal data.
c. Right to withdrawal or omission of items
Succinct – a succinct questionnaire asks questions that aim to ✓ Study participants should have a right to withdrawal from the
answer only the research objectives. Any questions beyond that survey in addition to the choice of opt out sharing the data
scope of the research should be excluded. It is common for already provided on an online questionnaire. Researchers should,
researchers to “cast the wider” so that they will collect more data, therefore, ensure that the opportunity to erase or skip questions
regardless of whether these data are important or not. This or backtrack through the survey is provided in order to maintain
usually happens when the researcher has not properly thought ethnically sound conduct.
through the research objectives. It runs the risk asking too many
questions and the questionnaire runs into many pages. 6. Piloting tools
Once the survey tools, contents, and platforms are decided, it is
Six Methodological Components in Designing an Online Survey very important to carry out a pilot with potential participants.
1. User-friendly design layout Pilot studies can help ensure the adequacy of the questions and
The format of the questionnaire should be easy for the comprehensiveness of the contents. Some aspects such as
participants to navigate around and should need only a minimum download time require piloting an electronic version of survey.
of computer skills. Items should be short, clear and easy to read Six Methodological Components in Designing Offline Survey
by the participants, elderly people might need larger fonts.
Sensitive or personal questions should be placed at the end. 1. Questionnaire design in terms of content
All process participants develop a questionnaire that meets the
2. Selecting survey participants survey goals and easy to process.
An easy access to surveys for all participants is essential in any 2. Creation of the questionnaire print template
online questionnaire survey. An online questionnaire may be The completed questionnaire is adapted to the layout of the
appropriate only for a certain age groups. For example, an online corporate identity and finalized as a print template.
study among elderly population would not be appropriate if the 3. Wording of the letter on invitation
proportion of elderly who access/use internet is low. To encourage participant, a cover letter is prepared, which is also
3. Avoiding multiple responses designed in accordance with the corporate identity and corporate
Another important feature of the online survey design is the design of the study.
ability to avoid multiple responses. This is a particularly 4. Print
challenging when incentives are provided to the survey The cover letter and the questionnaire can be printed internally or
participants. In order to minimize this problem, online survey handed over to an appropriate service provider for printing.
design should able to include a feature that enables to register However, the costs of copies of the instruments is expensive. If
interested participants (through their email) in the first stage so replies are to be sent by standard mail, a reply envelope is also
that the online tool will be able to assign a unique participant required.
number which will minimize the chance of multiple enrolments 5. Field phase
into the study. It is very important that the email should be used The questionnaires will be completed and collected within a
for sharing the survey link only to ensure participants details are certain period of time.
protected. Restriction through an IP address could be another 6. Evaluation
strategy to avoid multiple enrolments. The questionnaires are digitized. The results are then prepared
and evaluated externally or internally according to the study’s
4. Data management specifications.
Generally, online survey platforms offer convenient and reliable
data management. Online survey format protects against the loss Closed-ended Survey Questions
of data and facilitates data transfer into a data (e.g. excel or SPSS) Respondents’ answers are limited to fixed set of responses. Most
for analysis. It can be argued the overall ease of use for well- scales are closed ended. Types of closed-ended questions include:
designed questionnaires for both study participants and a. Yes/No questions – the respondent answers the questions with
researchers potentially improves the reliability and validity of the a ‘yes’ or a ‘no’.
data collection process. Example: have you ever visited EVSU-TC?
5. Ethical issues b. Multiple choice – the respondent has several options from
Online administration of surveys raises unique ethical questions which to choose.
regarding key ethical components including: Example: how would you rate the performance of BSIT students?
a. Informed consent (i)Excellent (ii)Good (iii)Fair (iv)Poor
✓ In most online survey tools, it is not possible to explain the
study or to take verbal consent from the participants. Researchers c. Scaled questions – responses are graded on a continuum
therefore have turned to ensuring that all information regarding (example: Rate the appearance of the performance of BSIT
the study, participants’ rights and researcher’s contact details are students on a scale from 1to 10, with 10 being the most preferred
provided on the first page of the survey. performance). Examples of scales include the likert scale and
b. Privacy and confidentially semantic differential scale. Likert scale is a psychometric scale
✓ There have been concerns regarding the ability of online commonly involved in research that employs questionnaires to
administration tools ability to facilitate primary and confidentially. measure social attitudes.
Survey tools such as SurveyMonkey have been associated with Example: Are you happy with your grades in the first semester?
(1) Strongly disagree (2) disagree
(3) don’t know (4) agree (5) strongly agree

A semantic differential scale is a list of opposite adjectives to

measure psychological meaning of an object to an individual.
Example: Rate the park on the following dimensions:
Safe ____: ____: ____: Dangerous
Dirty ____: ____: ____: Clean
Quiet ____: ____: ____:

Noisy Open-ended Survey Questions

The options or predefined categories are not suggested. The

respondent replies in their own words without being constrained
by a fixed set of possible responses. Types of Open-ended
questions include:

a. Completely unstructured — The questionnaire in this case has

a basic structure and some branching questions but nothing that
limits the responses of a respondent. Example: ‘What is your
opinion on questionnaires?’

b. Word association — Words are presented, and the respondent

mentions the first word that comes to mind. Example: If
interviewer says cold, the respondent may say hot and the like
ones.

c. Sentence completion — Respondents complete an incomplete

sentence. Example: ‘The most important consideration in my
decision to buy a new house is . . .’

d. Story completion — involves participants writing stories about

hypothetical scenarios created by the researcher, through
responding to a stimulus consisting of at least one complete
sentence which represents the beginning of a story plot. Example:
David has decided to start removing his body hair….”

e. Picture completion — Respondents fill in an empty

conversation balloon.

f. Thematic apperception test — Respondents explain a picture or

make up a story about what they think is happening in the picture.