 Quantitative research can be costly, difficult and

PRACTICAL RESEARCH 2 time-consuming.

LECTURE NOTES 1  Quantitative studies require extensive statistical
treatment, requiring stringent standards, more so
Lesson 1: Characteristics, Strengths, Weaknesses, and with confirmation of results.
Kinds of Quantitative Research  Quantitative research requires a large number of
respondents. It is assumed that the larger the
Inquiry – to look for information by asking various sample is, the more statistically accurate the findings
questions about the thing you are curious about. are.
Research – to discover truths by investigating on your  Many pieces of information are difficult to gather
chosen topic scientifically. using structured research instruments.
 If not done seriously and accurately, data from the
Inquiry and research are both seeking for information questionnaires may be incomplete and inaccurate.
to have extensive knowledge. While inquiry is an act  The information contextual factors to help interpret
of asking for information, research is a systematic the results or explain variations are usually ignored.
investigation and study of materials in order to
establish facts and reach new conclusions. Kinds of Quantitative Research
These are categorized based on the application of
Characteristics of Quantitative Research research method, purpose, and types of data needed.
 Objective – Focuses on the subject of the study
through measurable characteristics. This seeks Categories for the Kinds of Quantitative Research
accurate measurement and analysis of the target A. Based on Application of Research Method (Use of
subjects. Research)
 Clearly Defined Research Questions – The 1. Basic/Pure Research – deals with concepts,
researchers know in advance what they are principles, or abstract things. It aims to increase
looking for. The research questions are well- your knowledge about something.
defined for which objective answers are sought. 2. Applied Research – to apply your chosen
 Structured Research Instruments – Data are research to societal problems or issues, finding
normally gathered using structured research tools ways to make positive changes in society.
such as questionnaires to collect measurable
characteristics of the population. B. Based on Purpose of the Research
 Numerical Data – The data are in form of 1. Descriptive Research – defining or giving
numbers and statistics, often organized and portrayal or picture of a person, and etc.
presented using tables, charts, graphs, and figures 2. Correlational Research – shows relationships or
that consolidate large numbers of data to show connectedness of two factors.
trends, relationships, or differences among variables. 3. Exploratory/Formulative – to find out how
 Large Sample Sizes – A large population yields reasonable or possible it is to conduct a research
more reliable data, but principles of random study on a certain topic.
sampling must be strictly followed to prevent 4. Explanatory Research – elaborates the relationship
researcher’s bias. of two factors and the way which such
 Replication – Methods can be repeated to verify relationships exist.
findings in another setting, thus, reinforcing 5. Action Research – an ongoing practice for the
validity of findings purpose of obtaining results that will bring
 Future Outcomes – By using mathematical improvements in the system.
calculations and with the aid of computers, if-then
scenarios may be formulated to predict future C. Based on the Types of Data Needed
results. 1. Qualitative Research – requires non-numerical
data to express the results.
Strengths of Quantitative Research 2. Quantitative Research – involves measurement of
 It is most reliable and valid of concluding results, data.
giving way to a new hypothesis or to disproving it.
 Results or generalizations are more valid and Research design by Polit and Hungler (1999).
reliable because of a bigger number of the sample
of a population.
 It is objective. Since it provides numerical data, it
cannot be easily misinterpreted.
 The use of statistical techniques facilitates
sophisticated analyses and allows you to
comprehend a huge number of vital characteristics
of data.
 The numerical data can be analyzed in a quick and Let us focus on Quantitative Research Design:
easy way. The following are the various kinds of quantitative
 Quantitative studies are replicable. Standardized research design that a researcher may employ:
approached allow the study to be replicated in
different areas or over time with the formulation of 1. Experimental Research Design
comparable findings.  This allows the researcher to control the situation.
 “What causes something to occur?”
Weaknesses of Quantitative Research
 Allows the researcher to identify cause and effect
relationships between variables.
 Supports the ability to limit alternative
explanations and to infer direct causal relationships
in the study.
 The approach provides the highest degree level of
evidence for single studies.

A. Pre-Experimental Design
 Least internal validity Three Types of Quasi-Experimental Design:
 Pre-experiments are the simplest form of research
design. 1. Non-Equivalent Group Design
 Instead of comparing the pretest with the posttest  This refers to the chance failure of random
within one group, the posttest of the treated assignment to equalize the conditions by converting
groups is compared with that of an untreated a true experiment into this kind of design.
group.  The researcher chooses existing groups that
appear similar, but only one experiences the
 In a pre-experiment, either a single group or
treatment.
multiple groups are observed subsequent to some
 Most common type of quasi-experimental design.
agent or treatment presumed to cause change.
 If the groups are not random, they are non-
equivalent.
Three Types of Pre-Experimental Research Design:
 Example: You picked two people who are similar.
The Person 1 experienced the treatment, and the
a. One-shot case study design
other didn’t. After that, you can find out whether the
 A single group is studied at a single point in time
treatment has an impact by comparing them.
after some treatment that is presumed to have caused
change.
2. Regression discontinuity
 The carefully studied if the treatment can cause  Many potential treatments are designed around an
change or not. essentially arbitrary cutoff, where those above the
 No control or comparison group is employed, so the threshold received the treatment and those below
experimental is the only one existing. it did not.
 Example: You were asked about the outcome of the  The differences between the two groups are often
product they gave you without having a pre-test. minimal as to be nearly nonexistent. Therefore,
researchers can use individuals just below the
b. One-group pretest-posttest design threshold as a control group and those just above
 A single case is observed at two time points, one as a treatment group.
before the treatment and one after the treatment.
 No control or comparison group is employed, so the 3. Natural Experiments
experimental one is the only existing.  In a natural experiment, an external event or
 Example: You were tested before and after the situation results in the random or random-like
experiment to find out if the product caused changes. assignment of subjects to the treatment group.
 They are still not random assignments, since it is
c. Static-group comparison observational.
 A group that has experienced some treatment is  Example: There is an Ayuda, yet not anyone is able
compared with one that has not. to receive it even though it is for everybody. So, the
 Control group vs. Experimental group government will do a lottery of random names.

Two Classes of Experimental Design: Interrupted Time Series Design (continuation of no.1,
non-equivalent, in the orig. module)
B. Quasi-experimental Design  It employs multiple measures before and after the
 In this design, the researcher can collect more experimental intervention.
data, either by scheduling more observations or  It differs from the single-group pre-experiment
finding more existing measures. that has only one pretest and one posttest.
 It does not reply on random assignment; the
subjects are intentionally assigned or picked.
 Quasi-experimental design involves selecting
groups without any random pre-selection processes. C. True Experimental Design
 Like a true experiment, a quasi-experimental  It relies on statistical analysis to prove or disprove
design aims to establish a cause-and-effect a researcher’s hypothesis.
relationship between an independent and dependent  One of the most accurate forms of research
variable.  Only a true experimental design can establish a
 Example: There are doctors who wanted to try the cause-effect relationship within a group.
new therapy for the Patient 1, but there are also  True experimental design employs both treated
doctors who chose to use the old protocol for the and control groups to deal with time-related rival
same patient. There, they can identify the explanations.
progression between the new and old therapy.  In a true experiment:
  There is a control group that do not change and  Example: A comparative Study on the Health
an experimental group that changes. Problems among Rural and Urban People in Ilocos
 A variable can be manipulated by the Region, Philippines.
researcher.
 Random distribution of the variables. E. Normative
 It describes the norm level of characteristics for a
2. Non-Experimental Design given behavior.
 The researcher observes the phenomena as they  Example: If you are conducting research on the
occur naturally and no external variables are study habits of the high school students, you are to
introduced. use the range of score to describe the level of their
 The variables are not deliberately manipulated study habits.
nor is the setting controlled.
 Also called as Descriptive Research F. Evaluative
 Descriptive Research Design’s main purpose  It is a process used to determine what has
is to observe, describe and document aspects of happened during a given activity or in an
a situation. institution.
 The purpose of evaluation is to see if a given
Types of Descriptive Design: program is working.
 Example: In evaluation study, it will not just be
A. Survey considering the performance of the students who
 It is used to gather information from groups of were taught under modular instruction; instead, it is
people to see the general picture of the population. the rate of progress that happened among the
students who were exposed to modular instruction.
B. Correlational
 It is aim is to find out the direction, associations or G. Methodological
relationship between different variables.  In this approach, the implementation of a variety
of methodologies forms a critical part of achieving
Three Types of Correlational Research: the goal of developing a scale-matched approach,
where data from different disciplines can be
1. Bivariate Correlational Studies integrated.
 It obtains score from two variables for each
subject, and then uses them to calculate a
correlation coefficient.
 The term bivariate implies that the two variables
are correlated.
 Example: Children of wealthier (variable one),
better educated (variable 2) parents earn higher
salaries as adults.

2. Prediction Studies
 It uses correlation coefficient to show how one
variable (the predictor variable) predicts another
(the criterion variable).
 Example: Which high school applicants should be
admitted to college?

3. Multiple Regression Prediction Studies

 All variables in the study can contribute to the
over-all prediction in an equation that adds together
the predictive power of each identified variable.
 Example: Suppose the High School GPA is not the
sole predictor of college GPA, what might be other
good predictors? Answer: Strands

C. Ex-Post Facto or Causal Comparative

 This kind of research derives conclusion from
observations and manifestations that already
occurred in the past and now compared to some
dependent variables.
 Example: What is the Effect of Home Schooling on
the Social Skills of Adolescents?

D. Comparative
 It involves comparing and contrasting two or
more samples of study subjects.
 This design is used to compare two distinct groups
on the basis of selected attributes.