INEN 4063

New Trends in
Industrial Engineering
Research

Cayubit, Mark Christian M.

Gingo, Jomari T.
San Gabriel, John Jholo N.
Qualitative Research
A. What is Qualitative Research?

Qualitative Research is primarily exploratory research. It is used to gain an

understanding of underlying reasons, opinions, and motivations. It provides
insights into the problem or helps to develop ideas or hypotheses for
potential quantitative research. Qualitative Research is also used to uncover
trends in thought and opinions, and dive deeper into the problem. Qualitative
data collection methods vary using unstructured or semi-structured
techniques. Some common methods include focus groups (group
discussions), individual interviews, and participation/observations. The
sample size is typically small, and respondents are selected to fulfill a given
quota.

Qualitative Research describes social phenomena as they occur naturally - no

attempt is made to manipulate the situation - just understand and describe.
Understanding is sought by taking a holistic perspective / approach, rather
than looking at a set of variables. Qualitative research data is used to help us
to develop concepts and theories that help us to understand the social world
- which is an inductive approach to the development of theory, rather than a
deductive approach that quantitative research takes testing theories that
have already been proposed. Qualitative data is collected through direct
encounters i.e. through interview or observation and is rather time
consuming

B. When to use Qualitative Research?

These are the following methods that can be used in Qualitative Research:
1. Focus Group
= is a form of qualitative research in which a group of people are asked
about their perceptions, opinions, beliefs, and attitudes towards a product,
service, concept, advertisement, idea, or packaging. Questions are asked
in an interactive group setting where participants are free to talk with
other group members. During this process, the researcher either takes
notes or records the vital points he or she is getting from the group. Care
should be noted to select members of the group carefully for effective and
authoritative responses.

2. Individual Interviews
= The purpose of the research interview is to explore the views,
experiences, beliefs and/or motivations of individuals on specific matters
(factors that influence their attendance at the dentist). Qualitative
methods, such as interviews, are believed to provide a 'deeper'
understanding of social phenomena than would be obtained from purely
quantitative methods, such as questionnaires. 1Interviews are, therefore,
most appropriate where little is already known about the study
phenomenon or where detailed insights are required from individual
participants. They are also particularly appropriate for exploring sensitive
topics, where participants may not want to talk about such issues in a
group environment.
 3. Participations/Observations
participant observation is also one of the most demanding. It requires that
the researcher become a participant in the culture or context being
observed. The literature on participant observation discusses how to enter
the context, the role of the researcher as a participant, the collection and
storage of field notes, and the analysis of field data. Participant
observation often requires months or years of intensive work because the
researcher needs to become accepted as a natural part of the culture in
order to assure that the observations are of the natural phenomenon.

C. Sample Problem/Analysis using Qualitative Research

The sample analysis of Qualitative Research is about the writings by John Dewey
(2000 (1938]), another influential author. In his book Logic, the Theory of
Inquiry he very clearly outlines the process of research. Very reassuring for
beginning researchers, he states that research follows a uniform structure, which
applies to our everyday life as well as to science. In other words, there are
familiar elements in conducting research and we can draw on knowledge that we
already have gained in our everyday life. Dewey describes the research process
as follows:

The antecedent condition of inquiry that gets it all started is an indeterminate

or uncertain situation. It is a situation that makes us fell disturbed, troubled,
confused; it is ambiguous and contradictory. This leads us to formulate a
problem statement and to determine a way to solve this problem. Dewey puts it
very simply: We inquire when we question; and we inquire when we seek for
whatever will provide an answer to a question asked. (p. 105).

In consequence, research is and should be based on real life problems and

should not contain fictitious elements. Often questions are derived from the
personal biography or social context of the researcher. The connection between
social context and personal biography is for example obvious in the following
student projects I supervised in the past:
o being a boxing trainer himself, a student examined
the function of boxing as a way to help adolescents with a criminal record
to deal with aggression
o having provided voluntary work for elderly, a group
looked at personal benefits and down sights of the honorary office
o the soccer world championship took place in
Germany in 2006 and within this context some students looked at the new
German nationalism, others at public viewing events as a new form of
getting together
o based on the personal biography of a student who
served as soldier in the KFOR-Mission in Kosovo, he and his grouped
studied the individual consequences and effects on soldiers who take part
in military operations in foreign countries

After having come across an uncertain situation, the next step is to clearly
identify and formulate the problem. This is very important as the problem
statement is like a lens through which you look at reality, it reduces the
complexity of reality and structures the research field. Further, you derive more
detailed research questions and hypothesis from it and this can only work
successfully when the point of departure, the stated problem, is comprehensible
and unambiguously spelled out. See also the chapter on research design for
computer-assisted analysis in di Gregorio and Davidson (2008).
D. Steps in Solving Qualitative Research

1. Determine Research Questions: Focused questions are at the heart of

actionable qualitative research. In fact, they are at the heart of good
quantitative research as well and play a key role in Lean UX thinking. Are
users not using the mobile app because of usability, security concerns or
something else? How do users make decisions about how to invest: do they
ask a friend, use a financial advisor, or research on their own?
2. Design the Study: Getting input from users instead of just internal
discussions is an essential first step. With research questions defined, the
"What" of the study has been established. Now think Who, When, Where
and How. There are logistical advantages and challenges to collecting the
data you need. It's a matter of trading them off.
3. Collect Data: The qualitative researcher should assume the role of
an unobtrusive observer and have little impact on the settings being
observedwhether it be watching participants use existing products at
home or in a more controlled lab environment. Qualitative is often used
synonymously with small samples, but one can take a qualitative approach
to larger sample sizes (more than 50 participants) just as one can take a
quantitative approach to small sample sizes (less than 10).
4. Analyse Data: Most qualitative research studies generate a lot of data.
Creating a system for coding actions and notable quotes helps speed
through the process of turning utterances into actionable insights.
5. Generate Findings: What was learned from engaging users? This step
involves synthesizing the copious amount of notes, videos and artifacts.
As many of the responses from participants will be open-ended, there will
be a need to identify patterns. For example, when we were interviewing
users about why they didn't pay their credit card bill on their mobile phone,
we didn't ask users if they had security concerns. Instead, many of them
voiced the concern in their own words and stories.
6. Validate findings: One of the best ways to validate findings is
to triangulate using other methods, including surveys or additional sources.
One weakness of qualitative research is that it is hard to establish external
validity, that is, to provide corroborating evidence that the findings aren't
just the opinion of the researcher. Every researcher, of course, does bring
with her biases on the problems with a product or what deserves emphasis
in the interview.

One approach to minimize this researcher bias is to include a section on

the interviewer or principal investigator's background and how it might
influence their conclusions. Having recordings of sessions and detailed
notes helps other interested parties come to their own conclusions and can
help validate findings. Including verbatims along with the interpretation
also helps others see how the conclusions were drawn.

E. New Trends in Qualitative Research

Snap Survey Software

It is the ideal survey platform and online research software where structured
techniques such as large numbers of respondents and descriptive findings
 are required. Snap Survey Software has many robust features that will help
your organization effectively gather and analyze quantitative data.

Snap Surveys provides survey software and web based survey services to a
wide variety of organizations worldwide. We have offices in the UK in central
London and Bristol, and in the US in Portsmouth, NH.

It is precisely because of the inherent diversity of such requirements, and the

control that individual customers have over survey invitations, content and
presentation of questionnaires, together with the data gathered, that Snap
has chosen one overarching security standard to cover all systems.

Quantitative Research
F. What is Quantitative Research?
Quantitative methods emphasize objective measurements and the statistical,
mathematical, or numerical analysis of data collected through polls,
questionnaires, and surveys, or by manipulating pre-existing statistical data
using computational techniques. Quantitative research focuses on gathering
numerical data and generalizing it across groups of people or to explain a
particular phenomenon.

Quantitative research is about asking people for their opinions in a structured

way so that you can produce hard facts and statistics to guide you. To get
reliable statistical results, its important to survey people in fairly large
numbers and to make sure they are a representative sample of your target
market.

G. When to use Qualitative Research?

These are the following methods that can be used in Qualitative Research:

4. Survey research
= Survey research is a quantitative method whereby a researcher poses
some set of predetermined questions to an entire group, or sample, of
individuals. Survey research is an especially useful approach when a
researcher aims to describe or explain features of a very large group or
groups. This method may also be used as a way of quickly gaining some
general details about ones population of interest to help prepare for a
more focused, in-depth study using time-intensive methods such as in-
depth interviews or field research. In this case, a survey may help a
 researcher identify specific individuals or locations from which to collect
additional data.

5. Correlational research
= is a quantitative method of research in which you have 2 or more
quantitative variables from the same group of subjects, & you are trying
to determine if there is a relationship (or covariation) between the 2
variables (a similarity between them, not a difference between their
means). Theoretically, any 2 quantitative variables can be correlated (for
example, midterm scores & number of body piercings!) as long as you
have scores on these variables from the same participants; however, it is
probably a waste of time to collect & analyze data when there is little
reason to think these two variables would be related to each other.

6. Experimental research
= is based on a methodology that meets three criteria: (a) random
assignmentthe subjects (or other entities) are randomly assigned to
treatment groups, (b) experimental controlall features of the treatments
are identical except for the independent variable (i.e., the feature being
tested), and (c) appropriate measuresthe dependent measures are
appropriate for testing the research hypothesis. For example, in the class
size example, random assignment involves finding a group of students
and randomly choosing some to be in small classes (i.e, consisting of 15
students) and some to be in large classes (i.e., consisting of 30 students).
The researcher cannot use pre-existing small or large classes because
doing so would violate the criterion of random assignment. The problem
with violating random assignment is that the groups may systemically
differ; for example, students in the smaller classes may be at more
wealthy schools that also have more resources, better teachers, and
better-prepared students. This violation of the random assignment
criterion, sometimes called self-selection, is a serious methodological flaw
in experimental research.

7. Causal-comparative research
= is a research design that seeks to find relationships between
independent and dependent variables after an action or event has already
occurred. The researcher's goal is to determine whether the independent
variable affected the outcome, or dependent variable, by comparing two
or more groups of individuals. There are similarities and differences
between causal-comparative research, also referred to as ex post facto
research, and both correlational and experimental research. This entry
discusses these differences, as well as the benefits, process, limitations,
and criticism of this type of research design. To demonstrate how to use
causal-comparative research, examples in education are presented.

H. Sample Problem/Analysis using Quantitative Research

Example:
 Question: How old are you? (this question has the quality, of acquiring
different values)
Value 10-20
Value 21-40
Value 41-60
Value 61+

Variables are primarily used in surveys that are then statistically

evaluated and transformed into graphs.

Before you start creating a questionnaire, you should know that there
are different categories of variables that are not the same. Variables
can be classed into three main categories according to their acquired
values:

1. Interval (cardinal) Value is a number, which we can compare with

others without any difficulty and determine by how much it varies. This
category includes age or salary.
Nominal Nominal values are mostly expressed in words. These
include gender or marital status (male/female, single/married).
2. Ordinal Ordinal values can also be expressed verbally, but in
comparison to nominal values one can determine their order. However
it is not possible to determine exactly how and by how much they differ.
It is for example formal education (high school/college).

In our next episode we will focus on survey sample selection,

which is the last step before questioning respondents
themselves.

I. Steps in Solving Quantitative Research

7. Choose the type of Quantitative Research Questions: The type of
quantitative research question that you use in your dissertation needs to
be reflected in the way that you write out the research question; that is,
the word choice and phrasing that you use when constructing a research
question tells the reader whether it is a descriptive, comparative or
relationship-based research question. Therefore, in order to know how to
structure your quantitative research question, you need to start by
selecting the type of quantitative research question you are trying to
create: descriptive, comparative and/or relationship-based.
8. Identify the different types of variable you are trying to measure,
manipulate and/or control, as well as any groups you may be
interested in. Whether you are trying to create a descriptive,
comparative or relationship-based research question, you will need to
identify the different types of variable that you are trying
to measure, manipulate and/or control. If you are unfamiliar with the
different types of variable that may be part of your study, the article, Types
of variable, should get you up to speed. It explains the two main types of
 variables: categorical
variables (i.e., nominal, dichotomous and ordinal variables) and continuous
variables
(i.e., interval and ratio variables). It also explains the difference
between independent and dependent variables, which you need to
understand to create quantitative research questions.
9. Select the appropriate structure: The structure of the three types of
quantitative research question differs, reflecting the goals of the question,
the types of variables, and the number of variables and groups involved.
By structure, we mean the components of a research question (i.e., the
types of variables, groups of interest), the number of these different
components (i.e., how many variables and groups are being investigated),
and the order that these should be presented (e.g., independent variables
before dependent variables). The appropriate structure for each of these
quantitative research questions is set.
A. Structure of descriptive research questions
B. Structure of comparative research questions
C. Structure of relationship-based research questions
10.Analyze and Interpret Data: Most quantitative research studies
generate a lot of data. Take the data apart to look at individual responses.
Represent the data in tables, figures, and pictures. Explain conclusions
from the data that address the research questions. The population is the
group to be studied. The sample refers to specific people or events in the
population from which data will be collected.
11.Generate Findings: Sources of data may include people, literature,
documents, and findings (for example, from sources such as laboratory
data or measurements of vital signs).Data may be collected from
interviews, questionnaires, direct measurement, or examinations (such as
physical or psychological tests).
12.Validate findings: Through publications and presentations, the researcher
explains the results of the study and links them to the existing body of
knowledge in the literature. The researcher also describes the implications
of the study and suggests directions for further research.

o New Trends in Qualitative Research

Stata Software

This is an interactive program that can also be used for both simple
and complex analyses. It will also generate charts, graphs and plots of your
data and results. This program may seem a bit more complicated to some
researchers. It uses four different windows including the command window,
the review window, the result window and the variable window. While it is a
very useful program, the organization of this software may seem daunting.

You can combine and reshape datasets, manage variables, and collect
statistics across groups or replicates. You can work with byte, integer, long,
float, double, and string variables (including BLOBs and strings up to 2 billion
characters). Stata also has advanced tools for managing specialized data
 such as survival/duration data, time-series data, panel/longitudinal data,
categorical data, multiple-imputation data, and survey data.

Stata are programs which are used for the statistical analysis of the
collection, organization, analysis, interpretation and presentation of data. SAS
University Edition, GNU Octave, ADaMSoft, BV4.1, PSPP, R, pbdR, Shogun,
CSPro, CumFreq, Gretl.

Meta-Analysis Experimental Design

A. What is Meta-Analysis Design?

Meta-analysis is a statistical technique for amalgamating, summarising, and

reviewing previous quantitative research. By using meta-analysis, a wide
variety of questions can be investigated, as long as a reasonable body of
primary research studies exist. Selected parts of the reported results of
primary studies are entered into a database, and this "meta-data" is "meta-
analyzed", in similar ways to working with other data - descriptively and then
inferentially to test certain hypotheses. It can be used as a guide to answer the
question 'does what we are doing make a difference to X?', even if 'X' has been
measured using different instruments across a range of different people. Meta-
analysis provides a systematic overview of quantitative research which has
examined a particular question.

The appeal of meta-analysis is that it in effect combines all the research on

one topic into one large study with many participants. The danger is that in
amalgamating a large set of different studies the construct definitions can
become imprecise and the results difficult to interpret meaningfully.

Not surprisingly, as with any research technique, meta-analysis has its

advantages and disadvantages. An advantage is its objectivity, and yet like
any research, ultimately its value depends on making some qualitative-type
contextualization and understandings of the objective data.

B. When to use Quasi-Experimental Design?

 1. Publications

Many journals encourage researchers to submit systematic reviews and

meta-analyses that summarize the body of evidence on a specific question,
and this approach is replacing the traditional narrative review. Meta-analyses
also play supporting roles in other papers. For example, a paper that reports
results for a new primary study might include a meta-analysis in the
introduction to synthesize prior data and help to place the new study in
context.

2. Planning new studies

Meta-analyses can play a key role in planning new studies. The meta-analysis
can help identify which questions have already been answered and which
remain to be answered, which outcome measures or populations are most
likely to yield significant results, and which variants of the planned intervention
are likely to be most powerful.

3. Grant applications

Meta-analyses are used in grant applications to justify the need for a new
study. The meta-analysis serves to put the available data in context and to
show the potential utility of the planned study. The graphical elements of the
meta-analysis, such as the forest plot, provide a mechanism for presenting the
data clearly, and for capturing the attention of the reviewers. Some funding
agencies now require a meta-analysis of existing research as part of the grant
application to fund new research.

C. Sample Problem/Analysis using Quantitative Research

TITLE: The prevalence and incidence of coronary heart disease is significantly

increased in periodontitis: a meta-analysis.

Abstract

BACKGROUND:

Previous studies have shown conflicting results as to whether periodontitis

(PD) is associated with increased risk of coronary heart disease (CHD).
 The aim of the current study was to evaluate whether such an
association exists.

METHODS:

A systematic review of the literature revealed 5 prospective cohort studies

(follow-up >6 years), 5 case-control studies, and 5 cross-sectional
studies that were eligible for meta-analysis. Individual studies were
adjusted for confounding factors such as age, sex, diabetes mellitus,
and smoking. The 3 study categories were analyzed separately.
Heterogeneity of the studies was assessed by Cochran Q test. The
studies were homogeneous; therefore, the Mantel-Haenszel fixed-effect
model was used to compute common relative risk and odds ratio (OR).

RESULTS:

Meta-analysis of the 5 prospective cohort studies (86092 patients) indicated

that individuals with PD had a 1.14 times higher risk of developing CHD
than the controls (relative risk 1.14, 95% CI 1.074-1.213, P < .001).
The case-control studies (1423 patients) showed an even greater risk
of developing CHD (OR 2.22, 95% CI 1.59-3.117, P < .001). The
prevalence of CHD in the cross-sectional studies (17724 patients) was
significantly greater among individuals with PD than in those without
PD (OR 1.59, 95% CI 1.329-1.907, P < .001). When the relationship
between number of teeth and incidence of CHD was analyzed, cohort
studies showed 1.24 times increased risk (95% CI 1.14-1.36, P < .
0001) of development of CHD in patients with <10 teeth.

CONCLUSIONS:

This meta-analysis indicates that both the prevalence and incidence of CHD
are significantly increased in PD. Therefore, PD may be a risk factor for
CHD. Prospective studies are required to prove this assumption and
evaluate risk reduction with the treatment of PD.

D. Meta-Analysis Design Methodologies

1. Define the theoretical relationship of interest.

2. Collect the population of studies that provide data on the
relationship.
3. Code the studies and compute effect sizes.
 4. Examine the distribution of effect sizes and analyze the impact of
moderating variables.
5. Interpret and report the results.

J. New Trends in Meta-Analysis

Comprehensive Meta-Analysis Software

Is the world's most widely used program for meta-analysis. This software was
developed with funding from NIH, in collaboration with experts from around the
world.

The first is educational. The program was developed partly as an educational

tool, and it includes many features that help to explain the process of meta-
analysis. For example, we can use CMA to create a forest plot which shows
each of the individual studies and the combined effect size. Then, we can
manipulate the studies and see how these modifications impact the weight
assigned to each study, how they impact the summary effect, and so on.
Critically, we can also see how the selection of a model (fixed-effect vs.
random-effects) impacts the analyses. CMA allows us to open a window
(literally) on the calculations so that we can draw a direct link from the
concepts to the formulas.

The second is practical. While some participants may be interested solely in

understanding meta-analysis, others are also interested in performing meta-
analyses, and want to learn how to use CMA for this purpose. Therefore,
whenever we explain a procedure we also show how to perform that procedure
using CMA. The learning curve for CMA is extremely short, and so even when
working with CMA we can focus on concepts (which analysis should we choose,
how do we interpret the results) rather than the mechanics of actually using
the program.

Persons attending the workshop will be given a free license for CMA that starts
when you register for the workshop, and expires one month after the
workshop. We will offer a discounted rate for anyone who wants to obtain
license beyond the 30 days.
Quasi-Research/Quasi-Experimental Design

A. What is Quasi-Experimental Design?

Quasi-experimental research designs, like experimental designs, test causal

hypotheses. In both experimental (i.e., randomized controlled trials or RCTs)
and quasi-experimental designs, the programme or policy is viewed as an
intervention in which a treatment comprising the elements of the
programme/policy being evaluated is tested for how well it achieves its
objectives, as measured by a prespecified set of indicators (see Brief No. 7,
Randomized Controlled Trials). A quasi-experimental design by definition lacks
random assignment, however. Assignment to conditions (treatment versus no
treatment or comparison) is by means of self-selection (by which participants
choose treatment for themselves) or administrator selection (e.g., by officials,
teachers, policymakers and so on) or both of these routes. 1 Quasi-
experimental designs identify a comparison group that is as similar as possible
to the treatment group in terms of baseline (pre-intervention) characteristics.
The comparison group captures what would have been the outcomes if the
programme/policy had not been implemented (i.e., the counterfactual). Hence,
the programme or policy can be said to have caused any difference in
outcomes between the treatment and comparison groups.

B. When to use Quasi-Experimental Design?

Quasi-experimental methods that involve the creation of a comparison group

are most often used when it is not possible to randomize individuals or groups
to treatment and control groups. This is always the case for ex-post impact
evaluation designs. It may also be necessary to use quasi-experimental
designs for ex-ante impact evaluations, for example, where ethical, political or
logistical constraints, like the need for a phased geographical roll-out, rule out
randomization. Quasi-experimental methods can be used retrospectively, i.e.,
after the intervention has taken place (at time t+1, in table 1). In some cases,
especially for interventions that are spread over a longer duration, preliminary
impact estimates may be made at mid-term (time t, in table 1). It is always
highly recommended that evaluation planning begins in advance of an
intervention, however. This is especially important as baseline data should be
collected before the intended recipients are exposed to the programme/policy
activities
C. Quasi-Experimental Design Methodologies

1. Propensity score matching (PSM)

In PSM, an individual is not matched on every single observable

characteristic, but on their propensity score that is, the likelihood that the
individual will participate in the intervention (predicted likelihood of
participation) given their observable characteristics. PSM thus matches
treatment individuals/households with similar comparison
individuals/households, and subsequently calculates the average difference in
the indicators of interest. In other words, PSM ensures that the average
characteristics of the treatment and comparison groups are similar, and this
is deemed sufficient to obtain an unbiased impact estimate.

How to apply PSM involves the following five steps:

1. Ensure representativeness Ensure that there is a

representative sample survey of eligible participants and non-
participants in the intervention. Baseline data are preferred for
calculating propensity scores. This technique can, however, also be
used with endline data: the matching variables must be variables that
are unaffected by the intervention.

2. Estimate propensity scores The propensity scores are

constructed using the participation equation, which is either a logit or
probit regression with programme participation as the dependent
variable (in the programme = 1, not in the programme = 0). The
characteristics deemed to affect participation should be well
considered and as exhaustive as possible, but should exclude
characteristics that may have been affected by the intervention. For
this reason, it is best to use baseline data, where available, to estimate
the propensity scores.

3. Select a matching algorithm Each member of the treatment

group is then matched to one or more members of the comparison
group. There are different ways of doing this such as matching each
participant to their nearest neighbour non-participant. The mean of
the closest five neighbours is most commonly used. A single individual
in the comparison group may be matched to several different
individuals in the treatment group. In order for the matching to be
valid, it is essential to compare observed values for participants and
 non-participants with the same range of characteristics. Observations
in the comparison group with a propensity score lower than the lowest
observed value in the treatment group are discarded. Similarly,
observations in the treatment group with a propensity score higher
than the highest observed value in the comparison group are also
discarded. What remains is known as the region of common support

2. Regression discontinuity design (RDD)

This approach can be used when there is some kind of criterion that must be
met before people can participate in the intervention being evaluated. This is
known as a threshold. A threshold rule determines eligibility for participation
in the programme/policy and is usually based on a continuous variable
assessed for all potentially eligible individuals. For example, students below a
certain test score are enrolled in a remedial programme, or women above or
below a certain age are eligible for participation in a health programme (e.g.,
women over 50 years old are eligible for free breast cancer screening).
Clearly, those above and below the threshold are different, and the threshold
criterion (or criteria) may well be correlated with the outcome, resulting in
selection bias. Remedial education is provided to improve learning outcomes,
and therefore those with poorer learning outcomes are picked to be in the
programme. Older women are more likely to get breast cancer, and it is older
women who are selected for screening. So, simply comparing those in the
programme with those not in the programme will bias the results. Those just
either side of the threshold are not very different, however. If the threshold
for being enrolled in a remedial study programme is a test score of 60,
students enrolled in the programme who get a score of 58 to 59.9 are not
very different from those who get a score of 60 to 60.9 and are not enrolled.
Regression discontinuity is based on a comparison of the difference in
average outcomes for these two groups.
The first step is to determine the margin around the threshold and this is
done using an iterative approach. At first, a small margin can be set up, and
the resulting treatment and comparison groups can be tested for their
balance or similarity. If the match is good, the margin can be widened a little
and the balance checked again. This process must be repeated until the
samples start to become dissimilar. Although balancing is based on
observable characteristics, there is no reason to expect imbalance among
non-observable characteristics (this is different in the case of PSM, as
explained above). Once the sample is established, a regression line is fitted.
This is a line drawn through the data points that represents the best fit
between the variables being studied or that summarizes the relationship
between the selected variables that is, when the line slopes down (from top
left to bottom right) it indicates a negative or inverse relationship; when it
slopes up (from bottom left to top right) a positive or direct relationship is
 indicated. In this case, the regression line is fitted on the selected outcome of
interest (e.g., test scores). The sample for the regression is restricted to
observations just either side of the threshold. Often a major challenge for
RDD is the need for sufficient observations on either side of the threshold to
be able to fit the regression line. An example of a remedial education
programme is shown in figure 2. The selection criterion for eligibility to
participate in the programme is a pre-intervention test score, with a threshold
of 60. The outcome variable is a post-intervention test score. The scatter plot
shows that these two variables are, unsurprisingly, related. There is a positive
relationship between pre- and post-intervention test scores. Children with a
pre-intervention test score of below 60 received the remedial classes. The
sample used for the analysis is taken from just either side of the threshold
those included have pre-intervention test scores in the range of 50 to 70, i.e.,
10 units either side of the threshold. The fitted regression line has a jump;
this is the discontinuity. The size of this jump (which is 10) is the impact of
the programme that is, the remedial education programme increases test
scores by 10 points on average.
Bibliography:
https://www.snapsurveys.com/blog/what-is-the-difference-between-
qualitative-research-and-quantitative-research/
http://www.nature.com/bdj/journal/v204/n6/full/bdj.2008.192.html
http://www.socialresearchmethods.net/kb/qualmeth.php
http://atlasti.com/qualitative-research/
https://www.unicef-irc.org/publications/pdf/brief_8_quasi-experimental
%20design_eng.pdf
http://2012books.lardbucket.org/books/sociological-inquiry-principles-
qualitative-and-quantitative-methods/s11-survey-research-a-quantitative.html
http://libguides.usc.edu/writingguide/quantitative
https://www.capilanou.ca/programs-courses/psychology/student-
resources/research-guidelines/Correlational-Research-Guidelines/
https://www.education.com/reference/article/experimental-research/
https://www.orau.gov/cdcynergy/soc2web/Content/activeinformation/tools/too
lscontent/quantiativemethods.htm
http://methods.sagepub.com/reference/encyc-of-research-design/n42.xml
http://www.survio.com/en/blog/popular-series/quantitative-research-survey-
problem-formulation#.WIH8Hvl97IU
https://cirt.gcu.edu/research/developmentresources/research_ready/quantres
earch/whentouse
https://quizlet.com/9947628/steps-of-the-quantitative-research-flash-cards/
http://dissertation.laerd.com/how-to-structure-quantitative-research-
questions.php
https://www.meta-analysis.com/pages/why_do.php
http://www.wilderdom.com/research/meta-analysis.html