TYPES OF DISCRIMINANT

ANALYSIS
March 27, 2018

What is Discriminant Analysis?

A very commonly used method of classification is the Discriminant Analysis.
According to this method, distinct classes’ product data are based on
distinct Gaussian distributions. For the purpose of creating a classifier, the
parameters for the Gaussian distribution are estimated by the fitting function
for every class. In order to predict new data classes, the class having the
lowest cost of misclassification is found by the trained classifier.
Named after the inventor, R.A. Fisher, Linear Discriminant Analysis is also
called Fisher Discriminant. It is basically a technique of statistics which
permits the user to determine the distinction among various sets of objects in
different variables simultaneously. Discriminant Analysis deals with the
matrices that are applied in Multivariate Analysis of Variance (MANOVA).
When to Use Discriminant Analysis?
There are certain requirements for using this analysis:

 Data must be from different groups. Membership of group should be already

known before the starting analysis.
 It is used for the analysis of differences in groups.
Discriminant Analysis Examples

To use discriminant analysis, one needs to ensure that the data cases should
be members of two or more mutually exclusive groups. These may be
persons, animals, economic growth of a country at different points in time etc.
For example, a research team has been organized to study the outcomes of
buildings on fire when residents are involved. The purpose of the study is to
predict what elements can ensure the safe release of residents even before
the fire security team arrives. The Hypothesis is that many variables may be
good predictors of safe evacuation versus injury to during evacuation of
residents. These variables may be: number of residents, access to fire station,
number of floors in a building etc.
The research team will examine the previous incidents and come up with a
prediction equation which will be helpful in taking precautionary measures for
future incidents.

Discriminant Analysis Example in Education

Discriminant analysis techniques are helpful in predicting admissions to a
particular education program. Psychologists studying educational testing
predict which students will be successful, based on their differences in several
variables.

Discriminant Analysis Example in Political Sciences

Political scientists who study court case dispositions use techniques derived
from this analysis. These techniques are also used to examine voting
behavior among citizens or among legislators.

Discriminant Analysis in Social Sciences

Researchers have used discriminant analysis in a wide variety of analysis. In
social sciences, researchers have used these techniques in psychological and
educational testing. Another usage is in personnel testing. This technique is
useful in analyzing experimental data when assignment to a ‘treatment’ group
is presumed to effect scores on different criterion variables. C.O. Klecka’s
study of sex role stereotypes in children is an example of this.
Types of Discriminant Analysis
Linear Discriminant Analysis
This is used for performing dimensionality reduction whereas preserving as
much as possible the information of class discrimination. It is basically a
generalization of the linear discriminant of Fisher. This is a technique used in
machine learning, statistics and pattern recognition to recognize a linear
combination of features which separates or characterizes more than two or
two events or objects. The combination that comes out as a result might be
applied as linear classifier as well as for dimensionality reduction prior to later
classification.
Linear Discriminant Analysis and Analysis of Variance
LDA has a close link with Analysis of Variance (ANOVA) as well as regression
analysis that also tries to show each dependent variable as a linear
combination of rest of the measurements or features. But, analysis of variance
makes use of independent categorical variables along with a continuous
dependent variable, while Discriminant Analysis has continuous independent
variables along with the categorical dependent variable which is the class
label. Similar to LDA and Analysis of Variance are probity regression and
logistic regression. This is because they too elaborate a categorical variable
through continuous independent variables’ values. These other techniques
are used in applications where it is not accurate to make assumptions that the
independent variables have normal distributions, that is fundamentally
assumed for LDA technique.
Linear Discriminant Analysis and Principal Component Analysis
Linear Discriminant Analysis (LDA) has a close linked with Principal
Component Analysis as well as Factor Analysis. Here both the methods are in
search of linear combinations of variables that are used to explain the data.
LDA clearly tries to model the distinctions among data classes. On the other
hand, Principal Component Analysis does not consider the distinctions among
classes and the factor analysis method creates the feature combinations on
the basis of distinctions instead of similarities.
Discriminant Analysis also differs from factor analysis because this technique
is not interdependent: a difference between dependent and
independent variables should be created. LDA is applied min the cases where
calculations done on independent variables for every observation are
quantities that are continuous. While working on categorical independent
variables, a technique which is equivalent is discriminant correspondence
analysis.
Multiple Discriminant Analysis
Mutliple Discriminant Analysis is a technique used to compress a multivariate
signal for producing a low dimensional signal that is open to classification.
Multiple Discriminant Analysis does not perform classification directly. It only
helps classification is producing compressed signals that are open to
classification.

Uses of Multiple Discriminant Analysis

Mutliple Discriminant Analysis is useful as majority of the classifiers have a
major affect on them through the curse of dimensionality. This means that
when signals are shown in spaces that extremely high dimensional, the
performance of classifier is impaired catastrophically through the over-fitting
issue. This issue is lessened by compressing of signals down to a space that
is low dimensional as done by Multiple Discriminant Analysis. The technique
is also used for revealing neural codes.

It is referred to as a method used for reducing the distinction among variables

for the purpose of classifying them into a given number of broad groups. This
method is used in finance for compressing the variance among securities
while also permitting the person to screen for a number of variables. It is
linked with Discriminant Analysis that attempts in classification of a data set by
developing a rule which will give the most meaningful separation.

Simplicity of the Method

Despite the fact that this method needs a little of mathematical implications, it
is quite simple. Multiple Discriminant Analysis permits the analyst to consider
various stocks and emphasize on data pints which are very significant to a
particular kind of analysis, reducing down the other distinctions among stocks
without completely factoring them out. For instance, Multiple Discriminant
Analysis can be applied in selecting securities in accordance with the portfolio
theory based on statistics and put forward by Harry Markowitz. When this
technique is applied accurately, it helps in factoring our variables such as
price in favor of values which calculate historical consistency and volatility.
The Altman-Z score in Multiple Discriminant Analysis is used by Edward
Altman for which he is famous.

Quadratic Discriminant Analysis

The assumption of groups with matrices having equal covariance is not
present in Quadratic Discriminant Analysis. Similar to the Linear Discriminant
Analysis, an observation is classified into the group having the least squared
distance. But, the squared distance does not reduce to a linear function as
evident from the name, Quadratic Discriminant Analysis.
Quadratic distance, unlike linear distance is not symmetric. This means that
the quadratic discriminant function of a group, ‘i’ evaluated with the group ‘j’
mean is not equal to the group ‘j’ quadratic discriminant function evaluated
with the group ‘i’ mean. Quadratic distance, on the results, is known as the
generalized squared distance. In the cases where the sample group
covariance matrix’s determinant is less than one, there can be a negative
generalized squared distance.

Quadratic Discriminant Analysis and Linear Discriminant Analysis

The Mahalonobis distances are calculated by Minitab through the use of
covariance matrices of individual class. However, a quadratic discriminant
function is not calculated by Minitab. Quadratic Discriminant Analysis is linked
closely with the Linear Discriminant Analysis in which the assumption is made
that the calculations are distributed normally. In Quadratic Discriminant
Analysis, unlike Linear Discriminant Analysis, it is not assumed that the
covariance of every class is same. To calculate the parameters needed in
quadratic discrimination further data and computation is needed as compared
to linear discrimination. If there is less distinction in group covariance
matrices, the latter will perform in a similar way to quadratic discrimination.
Quadratic Discrimination is also known as a general type of Bayesian
discrimination.
Canonical Discriminant Analysis
Canonical Discriminant Analysis is a method of dimension-reduction liked with
Canonical Correlation and Principal Component Analysis. If a classification
variable and various interval variables are given, Canonical Analysis yields
canonical variables which are used for summarizing variation between-class
in a similar manner to the summarization of total variation done by principal
components.

First and Second Canonical Correlation

If more than two or two observation groups are given having measurements
on various interval variables, a linear combination of variables is derived by
Canonical Analysis which has the greatest possible multiple correlation with
groups. First Canonical Correlation is the name given to this highest multiple
correlation. Canonical weights or canonical coefficients are the linear
combination’s coefficients. In order to obtain the second canonical correlation
the linear combination which is uncorrelated with the initial canonical variable
is found which has the maximum multiple correlation with groups. The
procedure of digging out canonical variables could be done over and over
again till the amount of canonical variables is equal to the amount of original
variables or minus one from the number of classes; whatever is smaller.
The first canonical correlation must be as large as the multiple correlation
among any original variables and groups. In the case where original variables
have high correlations within the group, the first canonical correlation could be
bigger even though every multiple correlation is small. This implies that the
first canonical variable can demonstrate major distinctions between classes,
even though this is not done by any original variables.

For every canonical correlation, tests of Canonical Analysis hypothesize that

all smaller canonical correlations and this one are zero in population. The
variables must have an average multivariate normal distribution in every class,
having a common covariance matrix for the purpose of validating the levels of
probability.

Gaussian Discriminant Analysis

Gaussian Discriminant Analysis also known as GDA, is used when data can
be approximated with normal distribution. Two variables are mean and
standard deviations are important while computing this type of analysis.

In a nutshell it can be observed that Discriminant Analysis is a long-standing

technique used to derive dimensions among the groups that are different from
one another. It is seen that the method is sometimes the first technique
applied while approaching a problem of classification. Therefore, in order to
make use of this technique we should have in place a training data set. This is
not required by any other methods.

Bivariate statistical techniques:

What is Bivariate Analysis?
Bivariate analysis means the analysis of bivariate data. It is one of the simplest forms of statistical
analysis, used to find out if there is a relationship between two sets of values. It usually involves
the variables X and Y.
 Univariate analysis is the analysis of one (“uni”) variable.
 Bivariate analysis is the analysis of exactly two variables.
 Multivariate analysis is the analysis of more than two variables.
The results from bivariate analysis can be stored in a two-column data table. For example, you might
want to find out the relationship between caloric intake and weight (of course, there is a pretty strong
relationship between the two. You can read more here.). Caloric intake would be your independent variable,
X and weight would be your dependent variable, Y.

Bivariate analysis is not the same as two sample data analysis. With two sample data analysis (like a two
sample z test in Excel), the X and Y are not directly related. You can also have a different number of data
values in each sample; with bivariate analysis, there is a Y value for each X. Let’s say you had a caloric
intake of 3,000 calories per day and a weight of 300lbs. You would write that with the x-variable followed
by the y-variable: (3000,300).
Two sample data analysis
Sample 1: 100,45,88,99
Sample 2: 44,33,101
Bivariate analysis
(X,Y)=(100,56),(23,84),(398,63),(56,42)

Types of Bivariate Analysis

Common types of bivariate analysis include:

1. Scatter plots,
These give you a visual idea of the pattern that your variables follow.
A simple scatterplot.

2. Regression Analysis
Regression analysis is a catch all term for a wide variety of tools that you can use to determine how your
data points might be related. In the image above, the points look like they could follow an exponential
curve (as opposed to a straight line). Regression analysis can give you the equation for that curve or line.
It can also give you the correlation coefficient.

3. Correlation Coefficients
Calculating values for correlation coefficients are using performed on a computer, although you can find
the steps to find the correlation coefficient by hand here. This coefficient tells you if the variables are related.
Basically, a zero means they aren’t correlated (i.e. related in some way), while a 1 (either positive or
negative) means that the variables are perfectly correlated (i.e. they are perfectly in sync with each other).

Data editing
From Wikipedia, the free encyclopedia

Jump to navigationJump to search

Data editing is defined as the process involving the review and adjustment of collected survey data.
The purpose is to control the quality of the collected data.[1] Data editing can be performed manually,
with the assistance of a computer or a combination of both.[2]

Editing methods[edit]
Interactive editing[edit]
The term interactive editing is commonly used for modern computer-assisted manual editing. Most
interactive data editing tools applied at National Statistical Institutes (NSIs) allow one to check the
specified edits during or after data entry, and if necessary to correct erroneous data immediately.
Several approaches can be followed to correct erroneous data:

 Recontact the respondent

 Compare the respondent's data to his data from previous year
 Compare the respondent's data to data from similar respondents
 Use the subject matter knowledge of the human editor
Interactive editing is a standard way to edit data. It can be used to edit
both categorical and continuous data.[3] Interactive editing reduces the time frame needed to
complete the cyclical process of review and adjustment.[4]
Selective editing[edit]
Selective editing is an umbrella term for several methods to identify the influential errors, [note
1]
and outliers.[note 2] Selective editing techniques aim to apply interactive editing to a well-chosen
subset of the records, such that the limited time and resources available for interactive editing are
allocated to those records where it has the most effect on the quality of the final estimates of
publication figures. In selective editing, data is split into two streams:

 The critical stream

 The non-critical stream
The critical stream consists of records that are more likely to contain influential errors. These critical
records are edited in a traditional interactive manner. The records in the non-critical stream which
are unlikely to contain influential errors are not edited in a computer assisted manner.[5]
Macro editing[edit]
There are two methods of macro editing:[6]
Aggregation method[edit]
This method is followed in almost every statistical agency before publication: verifying whether
figures to be published seem plausible. This is accomplished by comparing quantities in publication
tables with same quantities in previous publications. If an unusual value is observed, a micro-editing
procedure is applied to the individual records and fields contributing to the suspicious quantity.[7]
Distribution method[edit]
Data available is used to characterize the distribution of the variables. Then all individual values are
compared with the distribution. Records containing values that could be considered uncommon
(given the distribution) are candidates for further inspection and possibly for editing.[8]
Automatic editing[edit]
In automatic editing records are edited by a computer without human intervention[9]. Prior
knowledge on the values of a single variable or a combination of variables can be formulated
as a set of edit rules which specify or constrain the admissible values

Coding (social sciences)

From Wikipedia, the free encyclopedia

Jump to navigationJump to search

In the social sciences, coding is an analytical process in which data, in both quantitative form (such
as questionnaires results) or qualitative form (such as interview transcripts) are categorized to
facilitate analysis.
 One purpose of coding is to transform the data into a form suitable for computer-aided analysis. This
categorization of information is an important step, for example, in
preparing data for computer processing with statistical software.
Some studies will employ multiple coders working independently on the same data. This minimizes
the chance of errors from coding and is believed to increase the reliability of data.

Contents

 1Directive
 2Quantitative approach
 3Qualitative approach
 4Memos
 5See also
 6References

Directive[edit]
One code should apply to only one category and categories should be comprehensive. There should
be clear guidelines for coders (individuals who do the coding) so that code is consistent.

Quantitative approach[edit]
For quantitative analysis, data is coded usually into measured and recorded as nominal or
ordinal variables.
Questionnaire data can be pre-coded (process of assigning codes to expected answers on designed
questionnaire), field-coded (process of assigning codes as soon as data is available, usually
during fieldwork), post-coded (coding of open questions on completed questionnaires) or office-
coded (done after fieldwork). Note that some of the above are not mutually exclusive.
In social sciences, spreadsheets such as Excel and more advanced software packages such
as R, Matlab, PSPP/SPSS, DAP/SAS, MiniTab and Stata are often used.

Qualitative approach[edit]
For disciplines in which a qualitative format is preferential, including ethnography, humanistic
geography or phenomenological psychology a varied approach to coding can be applied. Iain Hay
(2005) outlines a two-step process beginning with basic coding in order to distinguish overall
themes, followed by a more in depth, interpretive code in which more specific trends and patterns
can be interpreted.[1]
Much of qualitative coding can be attributed to either grounded or a priori coding.[2] Grounded coding
refers to allowing notable themes and patterns emerge from the document themselves, where as a
priori coding requires the researcher to apply pre-existing theoretical frameworks to analyze the
documents. As coding methods are applied across various texts, the researcher is able to apply
axial coding, which is the process of selecting core thematic categories present in several
documents to discover common patterns and relations.[3]
Prior to constructing categories, a researcher must apply a first cycle coding method. There are a
multitude of methods available, and a researcher will want to pick one that is suited for the format
and nature of their documents. Not all methods can be applied to every type of document. Some
examples of first cycle coding methods include:
 In Vivo Coding: codes terms and phrases used by the participants themselves. The objective is
to attempt to give the participants a voice in the research.
 Process Coding: this method uses gerunds ("-ing" words) only to describe and display actions
throughout the document. It is useful for examining processes, emotional phases and rituals.
 Versus Coding: uses binary terms to describe groups and processes. The goal is to see which
processes and organizations are in conflict with each other throughout the document. These can
be both conceptual and grounded objects.
 Values Coding: codes that attempt to exhibit the inferred values, attitudes and beliefs of
participants. In doing so, the research may discern patterns in world views.
The process can be done manually, which can be as simple as highlighting different concepts with
different colours, or fed into a software package. Some examples of qualitative software
packages include Atlas.ti, MAXQDA, NVivo, and QDA Miner.
After assembling codes it is time to organize them into broader themes and categories. The process
generally involves identifying themes from the existing codes, reducing the themes to a manageable
number, creating hierarchies within the themes and then linking themes together through theoretical
modeling.[4]

Memos[edit]
Creating memos during the coding process is integral to both grounded and a priori coding
approaches. Qualitative research is inherently reflexive; as the researcher delves deeper into their
subject, it is important to chronicle their own thought processes through reflective or methodological
memos, as doing so may highlight their own subjective interpretations of data [5] It is crucial to begin
memoing at the onset of research. Regardless of the type of memo produced, what is important is
that the process initiates critical thinking and productivity in the research. Doing so will facilitate
easier and more coherent analyses as the project draws on [6] Memos can be used to map research
activities, uncover meaning from data, maintaining research momentum and engagement and
opening communication

Difference between Qualitative Analysis and

Quantitative Analysis
• Categorized under Mathematics & Statistics,Science | Difference
between Qualitative Analysis and Quantitative Analysis
Qualitative and quantitative analysis are two fundamental
methods of collecting and interpreting data in research. The
methods can be used independently or concurrently since they all
have the same objectives. They have some errors, and so using
them concurrently can compensate for the errors each has and
then produce quality results.
Moreover, there are overlaps in quantitative and qualitative
analysis. This article uncovers the key differences between
these two research analysis methods.

What is quantitative
analysis?
Quantitative analysis is often associated with numerical analysis
where data is collected, classified, and then computed for certain
findings using a set of statistical methods. Data is chosen
randomly in large samples and then analyzed. The advantage of
quantitative analysis the findings can be applied in a general
population using research patterns developed in the sample. This
is a shortcoming of qualitative data analysis because of limited
generalization of findings.

Quantitative analysis is more objective in nature. It seeks to

understand the occurrence of events and then describe them
using statistical methods. However, more clarity can be obtained
by concurrently using qualitative and quantitative methods.
Quantitative analysis normally leaves the random and scarce
events in research results whereas qualitative analysis considers
them.

Quantitative analysis is generally concerned with measurable

quantities such as weight, length, temperature, speed, width, and
many more. The data can be expressed in a tabular form or any
diagrammatic representation using graphs or charts. Quantitative
data can be classified as continuous or discrete, and it is often
obtained using surveys, observations, experiments or interviews.

There are, however, limitations in quantitative analysis. For

instance, it can be challenging to uncover relatively new concepts
using quantitative analysis and that is where qualitative analysis
comes into the equation to find out “why” a certain phenomenon
occurs. That is why the methods are often used simultaneously.
What is a qualitative
analysis?
Qualitative analysis is concerned with the analysis of data that
cannot be quantified. This type of data is about the understanding
and insights into the properties and attributes of objects
(participants). Qualitative analysis can get a deeper
understanding of “why” a certain phenomenon occurs. The
analysis can be used in conjunction with quantitative analysis or
precede it.
Unlike with quantitative analysis that is restricted by certain
classification rules or numbers, qualitative data analysis can be
wide ranged and multi-faceted. And it is subjective, descriptive,
non-statistical and exploratory in nature.
Report this ad

Because qualitative analysis seeks to get a deeper understanding,

the researcher must be well-rounded with whichever physical
properties or attributes the study is based on. Oftentimes, the
researcher may have a relationship with the participants where
their characteristics are disclosed. In a quantitative analysis the
characteristics of objects are often undisclosed. The typical data
analyzed qualitatively include color, gender, nationality, taste,
appearance, and many more as long as the data cannot be
computed. Such data is obtained using interviews or observations.

There are limitations in qualitative analysis. For instance, it

cannot be used to generalize the population. Small samples are
used in an unstructured approach and they are non-
representative of the general population hence the method cannot
be used to generalize the entire population. That is where
quantitative analysis into the factor.
Key differences
between qualitative
and quantitative
analysis
Definition of qualitative and quantitative analysis

Qualitative data analysis is based on classification of objects

(participants) according to properties and attributes whereas
quantitative analysis is based on classification of data based on
computable values. Qualitative analysis is subjective whereas
quantitative is objective.

Data collection for qualitative and quantitative

analysis

In qualitative analysis, the data is collected in small,

unrepresentative samples in an unstructured way. Typical data
collected include color, race, religion, nationality, and many more.
In quantitative analysis, on the other hand, data is collected in
large, representative samples that can generalize the entire
population.
Research methodology involved in qualitative and
quantitative analysis

Qualitative analysis methodology is exploratory where the

analysis seeks to get a deeper understanding of why a certain
phenomenon occurs. The methodology in quantitative analysis
can be conclusive such as how much or how many times a certain
phenomenon occurs not why it does occur.

Research findings

In qualitative analysis, research findings are specific to the objects

being studied and are not applicable on the general population
whereas in quantitative analysis the findings can be applicable on
the general population.

Method of data collection

In qualitative analysis, researchers often ask open-ended

questions, conduct interviews, and observations whereas in
quantitative analysis researchers take measurements, conduct
surveys, experiments and observations.

Purpose of qualitative and quantitative analysis

Qualitative analysis seeks to get a deeper understanding into

social interactions while quantitative analysis seeks to test
hypotheses and even give future predictions
Qualitative analysis Vs. Quantitative analysis:
Comparison chart

Report this ad
 Summary of
Qualitative analysis
Vs. Quantitative
analysis
 Quantitative analysis quantifies data to test hypotheses or predict
the future whereas qualitative analysis seeks to get a deeper
understanding of why certain things occur
 The sample is small in qualitative analysis and cannot be used to
represent the whole population while in quantitative analysis the
sample is large and can represent the entire population
 The researcher conducts interviews or surveys to collect
qualitative data whereas in quantitative analysis the research
conducts experiments, observations and measurements
 Typical data include color, race, gender, in qualitative analysis
whereas in quantitative analysis include all measurable quantities
such as density, length, size, weight
 Qualitative analysis is exploratory and subjective
 Quantitative analysis is conclusive and objective

Material 2:
 BASIS FOR
QUALITATIVE DATA QUANTITATIVE DATA
COMPARISON

Meaning Qualitative data is the data Quantitative Data is the

in which the classification type of data which can be
of objects is based on measured and expressed
attributes and properties. numerically.

Research Exploratory Conclusive

Methodology

Approach Subjective Objective

Analysis Non-Statistical Statistical

Collection of data Unstructured Structured

Determines Depth of understanding Level of occurrence

Asks Why? How many or How much?

Sample Small number of non- Large number of

representative samples representative samples

Outcome Develops initial Recommends final course

understanding. of action.

Definition of Qualitative Data

Qualitative Data refers to the data that provides insights and understanding
about a particular problem. It can be approximated but cannot be computed.
Hence, the researcher should possess complete knowledge about the type of
characteristic, prior to the collection of data.
The nature of data is descriptive and so it is a bit difficult to analyze it. This
type of data can be classified into categories, on the basis of physical attributes
and properties of the object. The data is interpreted as spoken or written
narratives rather than numbers. It is concerned with the data that is
observable in terms of smell, appearance, taste, feel, texture, gender,
nationality and so on. The methods of collecting qualitative data are:

 Focus Group
 Observation
 Interviews
 Archival Materials like newspapers.

Definition of Quantitative Data

Quantitative Data, as the name suggests is one which deals with quantity or
numbers. It refers to the data which computes the values and counts and can
be expressed in numerical terms is called quantitative data. In statistics, most
of the analysis are conducted using this data.

Quantitative data may be used in computation and statistical test. It is

concerned with measurements like height, weight, volume, length, size,
humidity, speed, age etc. The tabular and diagrammatic presentation of data is
also possible, in the form of charts, graphs, tables, etc. Further, the
quantitative data can be classified as discrete or continuous data. the methods
used for the collection of data are:

 Surveys
 Experiments
 Observations and Interviews

Key Differences Between Qualitative and Quantitative

Data
The fundamental points of difference between qualitative and quantitative
data are discussed below:

1. The data type, in which the classification of objects is based on

attributes (quality) is called qualitative data. The type of data which can
be counted and expressed in numbers and values is called quantitative
data.
2. The research methodology is exploratory in qualitative data, i.e. to
provide insights and understanding. On the other hand, quantitative
data is conclusive in nature which aims at testing a specific hypothesis
and examine the relationships.
3. The approach to inquiry in the case of qualitative data is subjective and
holistic whereas quantitative data has an objective and focused
approach.
4. When the data type is qualitative the analysis is non-statistical. As
opposed to quantitative data which uses statistical analysis.
5. In qualitative data, there is an unstructured gathering of data. As against
this, data collection is structured in quantitative data.
6. While qualitative data determines the depth of understanding,
quantitative data ascertains the level of occurrence.
7. Quantitative data is all about ‘How much or how many’. On the
contrary, qualitative data asks ‘Why?’
8. In qualitative data the sample size is small and that too is drawn from
non-representative samples. Conversely, the sample size is large in
quantitative data drawn from the representative sample.
9. Qualitative data develops initial understanding, i.e. it defines the
problem. Unlike quantitative data, which recommends the final course
of action.