INTRODUCTION TO

STATISTICS
STATISTICS
 In statistics, a sample:
a. can be used for inferences but not
for predictions.
b. is another word for population.
c. is only used in descriptive statistics.
d. is a set of data taken from the
population to represent the
population.
Question 1 of 5:
 Answer: d

Answer 1 of 5:
 How do descriptive and
inferential statistics differ?
a. Inferential statistics only attempt to
describe data, while descriptive statistics
attempt to make predictions based on data.
b. Inferential statistics are more
computationally sophisticated than
descriptive statistics.
c. Descriptive statistics are more
computationally sophisticated than
inferential statistics.
d. Descriptive statistics only attempt to
describe
Question 2 of 5:
data, while inferential statistics
attempt to make predictions based on data.
 Answer: d

Answer 2 of 5:
 Which two are examples of
descriptive statistics?

a. Hypothesis testing and histograms.

b. Median and correlation.
c. Mean and standard deviation.
d. Variance and regression analysis.
Question 3 of 5:
 Answer: c

Answer 3 of 5:
 What is statistical
estimation?
a. Methods for reducing errors in descriptive
statistics.
b. Methods for reducing errors in inferential
statistics.
c. Methods for rounding answers in statistical
calculations.
d. Methods to determine the best graph to
represent statistical data.
Question 4 of 5:
 Answer: b

Answer 4 of 5:
 What are two examples of
inferential statistics?
a. Regression analysis and hypothesis
testing.
b. Variance and correlation.
c. Range and percentiles.
d. mean and probability distributions.

Question 5 of 5:
 Answer: a

Answer 5 of 5:
Statistics
A branch of mathematics that examines and investigates way to
process and analyze the data gathered
Provides procedure in data collection, presentation, organization, and
interpretation to have a meaningful data that is useful to business
Statistics is generally understood as the subject dealing with number
and data, more broadly it involves activities such as collection of data
from survey or experiment, summarization or management of data,
presentation of results in a convincing format, analysis of data or
drawing valid inferences from findings.
Kinds of Statistics
Descriptive Statistics – is the totality of methods and
treatments employed in the collection, description, and
analysis of numerical data
◦ To tell something about the particular group of observation
Inferential Statistics – logical process from sample analysis
to a generalization or conclusion about a population
(statistical inference/inductive statistics)
Population and Sample

Sample

Population
Population
Population – refers to the totality of observations or elements from a
set of data.
◦ Example. Suppose a teacher conducts a study on the correlation of the
students’ entrance examination scores and their respective academic
performance. To ensure the validity of his findings, he decided to include all
the students who are enrolled for the current school year under a certain
program or course, hence the entire population.
Sample
Sample– refers to one or more elements taken from the population
for a specific purpose.
◦ Example. Because of the budget issues and feasibility concerns, the teacher
decided to include only a group of 200 students to participate in his study.
Parameter versus Statistic
Parameter – a numerical measure that describes the whole
population
◦ If all students in a school are surveyed about their heights and an
average height of 65 inches was determined, then 65 inches is
called a population parameter.
Statistic – a numerical description of the sample
◦ 65 inches will be called a sample statistic when only 50 students
out of 230 students are surveyed to determine the average height.
Sources of Data
Primary Data are data that come from an original source, and
are intended to answer specific research questions, can be
taken by interview, mail-in questionnaire, survey, or
experimentation.
Secondary Data are data that are from previously recorded
data, such as information in research conducted, industry
financial statements, business periodicals, and government
reports. It can also be taken electronically like internet
websites or compact disk.
Characteristics of Objects, people or
events
Constant is a characteristic of object, people or events that does not
vary like temperature at which water boils (100 degrees Celsius)
Variable is characteristic of objects, people, events that can take of
different values. It can vary in quantity like weight of people, or in
quality like hair color.
Basic Types of Variables/Data
Qualitative
◦ is conceptualized and analyzed as distinct categories, with no continuum
implied.
◦ Categorical variable
◦ Observations that are put in the same or different classes, each class being
considered as possessing some common characteristic that is not shared by
those in other classes.
Example: eye color, gender, occupation, religious preference, etc.
Basic Types of Variables/Data
Quantitative Data
◦ Also termed as numerical variable;
◦ Variates that yield frequencies when counted, giving rise to discrete
variable or when measured yield frequencies when counted, giving
rise to discrete variable or when measured yield variates that yield
metric or continuous variables
◦ variable that is conceptualized and analyzed along continuum
implied.
◦ Differs in amount of degree
Example: height, weight, math aptitude, salary, etc.
Types of Variables
Variables

Qualitative Quantitative

Discrete Continuous
Mathematical Classification
Continuous variable – is a variable which can assume any of
an infinite number of values, and can be associated with
points on a continuous line interval.
◦ Example. Height, weight, volume, etc.
Discrete variable – is a variable which consist of either a
finite number of values or countable number of values
◦ Gender, courses, olympic games, etc.
Experimental Classification
Independent variables – are variables controlled by the
experimenter/researcher, and expected to have an effect on the
behavior of the subjects.
◦ Also called explanatory variable

Dependent variable – is some measure of the behavior of subject

and expected to be influenced by the independent variable.
◦ Also called as outcome variable
Example.
To predict the value of fertilizer on the growth of plants, the
dependent variable is the growth of the plants; while the
independent variable is the amount of fertilizer used.
Levels of Measurement
Nominal level of measurement
◦ Mutually exclusive and exhaustive meaning
◦ Used to differentiate classes or categories for purely classification
or identification purposes
◦ MUTUALLY EXCLUSIVE is a property of a set of categories such that an individual or object
is included in only one category.
◦ Exhaustive is a property of a set of categories such that each individual or object must
appear in a category
Levels of Measurement
Ordinal level of measurement – Is used in ranking
◦ It is somewhat stronger form of measurement, because an
observed value classified into one category is said to posses more
of a property being scaled than does an observed value classified
into another category.
Levels of Measurement
Interval level of measurement – is used to classify order and
differentiate between classes or categories in terms of
degrees of differences (either discrete or continuous)
Levels of Measurement
Ratio level of measurement – it differs from interval
measurement only in one aspect; it has true zero point
(complete absence of the attitude being measured)
Classification of Numerical Data
Numerical
Data

Qualitative Quantitative

Nominal Ordinal Interval Ratio

 Level of Properties
Measureme
nt
Characteristics of
Nominal Indicates a distinction
Levels of
Measurement Ordinal Indicates a distinction
Indicates the direction of the distinction
Interval Indicates a distinction
Indicates the direction of the distinction
Indicates the amount of distinction (in equal
intervals)
Ratio Indicates a distinction
Indicates the direction of the distinction
Indicates the amount of distinction
Indicates an absolute zero
 Scale Basic Common Marketing Permissible Statistics
Characteristics Examples Examples
Descriptive Inferential
Nominal Numbers identify & Social security Brand numbers, percentages, Chi-square,
classify objects numbers, store types mode binomial test
numbering of
football players
Ordinal Numbers indicate Quality rankings, Preference Percentile, Rank-order,
the relative rankings of rankings, market median correlation,
positions of objects teams in a position, social Friedman,
but not the tournament class ANOVA
magnitude of
differences
between them
Interval Differences Temperature Attitudes, Range, mean, Product-moment
between objects opinions, index standard
deviation
Ratio Zero point is fixed, Length, weight Age, sales, Geometric mean, Coefficient of
ratios of scale income, costs harmonic mean variation
values can be
compared
 Types of Data
Goal Measurement Rank, Score, or Binomial (Two Survival Time
Gaussian Population Measurement (Non- possible outcomes)
Gaussian
Population)
Describe one group Mean, standard deviation Median, interquartile range Proportion Kaplan Meier survival curve

Compare one group to a One sample t-test Wilcoxon test Chi-square or Binomial Test
hypothetical value
Compare two unpaired Unpaired t-test Mann-Whitney test Fisher’s test (Chi-square for Log-rank test or Mantel-
groups large samples Haenszel
Compare two paired groups Paired t-test Wilcoxon test McNemar’s test Conditional proportional
hazards regression
Compare three or more One-way ANOVA Kruskal-Wallis test Chi-square test Cox proportional hazard
unmatched groups regression
Compare three or more Repeated-measures Friedman test Cochrane Q Conditional proportional
matched groups ANOVA hazard regression
Quantify association Pearson correlation Spearman rho correlation Contingency coefficients
between two variables
Predict value from another Simple linear regression or Nonparametric regression Simple logistic regression Cox proportional hazard
measured variable nonlinear regression regression
Predict value from several Multiple linear regression or Multiple logistic regression Cox proportional hazard
measured or binomial multiple nonlinear regression
variables regression
Sampling Techniques
Probability Sampling
◦ Each member of the population has known probability of being selected in
the sample
Nonprobability Sampling
◦ There is bias in the selection and there is no recognized probability that one
member will be included in the sample
Sampling Techniques
Sampling Simple Random Sampling
Techniques Probability
Systematic Sampling
Sampling
Stratified Sampling
Cluster Sampling

Convenience Sampling
Nonprobability
Purposive Sampling
Sampling
Snowball Sampling
Quota Sampling
Simple Random Sampling
(Probability Sampling)
Most commonly used sampling technique
Each member of the population has an equal chance to be selected
as a participant
Done by choosing the members of the sample one by one, using
either the lottery method or the tables of random numbers
Systematic Random Sampling
(Probability Sampling)
It considers every nth element of the population in the sample with
the selected random starting point from the first k members
Systematic Random
Sampling
1 26
51 76
N = 100 2 27
52 77
3 28
53 78
4 29
54 79
5 30
55 80
6 31
56 81
7 32
57 82
8 33
58 83
9 34
59 84
10 35
60 85
11 36
61 86
12 37
62 87
13 38
63 88
14 39
Systematic Random
Sampling
1 26
51 76
N = 100 2 27
52 77
3 28
53 78
Want n = 20 4 29
54 79
5 30
55 80
6 31
56 81
7 32
57 82
8 33
58 83
9 34
59 84
10 35
60 85
11 36
61 86
12 37
62 87
13 38
63 88
14 39
Systematic
Random Sampling
1 26
51 76
N = 100 2 27
52 77
3 28
53 78
want n = 20 4 29
54 79
5 30
55 80
6 31
N/n = 5 56 81
7 32
57 82
8 33
58 83
9 34
59 84
10 35
60 85
11 36
61 86
12 37
62 87
13 38
63 88
14 39
Systematic Random
Sampling 1
51
2
26
76
27
N = 100
52 77
3 28
53 78
Want n = 20 4 29
54 79
5 30
55 80
6 31
N/n = 5 56 81
7 32
57 82
8 33
Select a random number from 1-5: chose 4 58 83
9 34
59 84
10 35
60 85
11 36
61 86
12 37
62 87
13 38
63 88
14 39
Systematic Random
Sampling N = 100
1
51
2
26
76
27
52 77
3 28
53 78
Want n = 20 4 29
54 79
5 30
55 80
6 31
N/n = 5 56 81
7 32
57 82
8 33
Select a random number from 1-5: chose 4 58 83
9 34
59 84
10 35
60 85
11 36
Start with #4 and take every 5th unit 61 86
12 37
62 87
13 38
63 88
14 39
Stratified Sampling
(Probability Sampling)
Particularly useful only in conditions when the population is divided
into homogeneous groups (grouped based on a controlling variables
in the study such as gender, race, civil status, or nationality)
Homogeneous partitions are also called STRATA (singular form:
STRATUM).
 Example.
A sample of 100 students is to be selected from a junior
high school population of 1000 of which
◦ 250 are in Grade 7
◦ 200 are in Grade 8
◦ 300 are in Grade 9
◦ 250 are in Grade 10

If the sample size is to be

proportionally distributed, how many
samples are to be taken from each
stratum?
Solution
Partitions Size of the Number of
Partition Samples
Grade 7 250 250/1000 * 100 = 25
Grade 8 200 200/1000 * 100 = 20
Grade 9 300 300/1000 * 100 = 30
Grade 10 250 250/1000 * 100 = 25
Total 1000 100
Cluster Sampling
(Probability Sampling)
population is divided into groups (clusters)
clusters are heterogeneous groups of the population which means
◦ Grouped differently according to the controlling variables of the study
Convenience Sampling
(Nonprobability Sampling)
Also called haphazard sampling
Carried out on the matter of convenience or ease of implementation
on the part of the researcher, that is, samples taken are readily
available to participate in the study
Example:
◦ ambush interview
◦ Opinion poll
Purposive Sampling
(Nonprobability Sampling)
Also called judgmental or selective sampling,
Sampling done with a purpose wherein samples are taken based
on the judgment of the researcher
Its goal is to carefully choose the members of the population
which are best fitted to answer the research questions
Example:
◦ If a researcher wants to study the toothpaste brand mostly preferred by
people, then he would on purpose go to the nearby convenience store and
conduct an interview among all buyers of a particular brand of toothpaste
Snowball Sampling
(Nonprobability Sampling)
Also called Chain-referral sampling
Chooses a possible respondent for the study at hand, then, each
respondent is asked to give recommendations or referrals to other
possible respondents
Very effective sampling technique especially when the suitable
participants of the study are hard to find
Quota Sampling
(Nonprobability Sampling)
Equivalent of stratified random sampling in terms of nonprobability
sampling
Researcher starts by identifying quotas (predefined control
categories such as age, gender, education, or religion)
Sample chosen by the researcher should be of the same proportion
to this population
Example
◦ Researcher collects the sample which has the same proportion in which 30%
are single and 70% are married
In general, the probability sampling techniques are more preferred in
researches and studies than the nonprobability sampling techniques
as they ensure the representativeness of the whole population. Non
probability sampling techniques, however, are criticized for their lack
of randomization and representative quality.
Methods of Collecting Data
1. Direct or Interview Method
◦ It is a face-to-face encounter between the interviewer and the interviewee.
The interview may vary according to the preference of either or both parties.
However, this method is time-consuming, expensive, and has limited field
coverage.
Methods of Collecting Data
2. Indirect or Questionnaire Method
◦ Unlike direct method, this method utilized questionnaires to obtain
information. It can be done by mail or hand-carried to the intended
respondents.

3. Registration Method
◦ This method of gathering information is governed by laws.
Methods of Collecting Data
4. Observation Method
◦ This method is used to data that are pertaining to behaviors of an individual
or group of individuals at the time of occurrence of a given situation are best
obtained by observation. One limitation of this method is observation is made
only at the time or occurrence of the appropriate events.
Methods of Collecting Data
5. Experiment Method
◦ This is used to determine the cause and effect relationship of certain
phenomena under controlled conditions. This method usually employed by
scientific researchers.
Methods of Presenting Data
Textual Method – narrative and paragraph forms
Tabular Method – tables which are orderly arranged in rows and
columns for an easier and more comprehensive comparison of
figures
Graphical Method – visual or pictorial form to get a clear view of
data (histogram, pareto chart, pictograph, etc.)
Summation Notation, Sigma Σ
Example. Write the following expressions in
expanded form.

1.
2.
3.
Solution:

1.
2.
3.
Example. Evaluate the following notations using
the values below.X1 = 1 X2 = 3 X3 = 2 X4 = 5
y1 = 0 y2 = 8 y3 = 1 y4 = 6
z1 = 4 z2 = 7 z3 = -2 z4 = 3
Frequency Distribution
After collecting data, the first task for a researcher is to
organize and simplify the data so that it is possible to get a
general overview of the results.

This is the goal of descriptive statistical techniques.

One method for simplifying and organizing data is to construct

a frequency distribution.
Frequency Distributions
A frequency distribution is an organized tabulation showing
exactly how many individuals are located in each category on
the scale of measurement. A frequency distribution
presents an organized picture of the entire set of scores, and
it shows where each individual is located relative to others in
the distribution.
Frequency Distribution Tables
A frequency distribution table consists of at least two columns - one listing
categories on the scale of measurement (X) and another for frequency (f).
In the X column, values are listed from the highest to lowest, without skipping
any.
For the frequency column, tallies are determined for each value (how often each
X value occurs in the data set). These tallies are the frequencies for each X
value.
The sum of the frequencies should equal N.
Regular Frequency Distribution
When a frequency distribution table lists all of
the individual categories (X values) it is called a
regular frequency distribution.
Grouped Frequency Distribution
Sometimes, however, a set of scores covers a wide
range of values. In these situations, a list of all the X
values would be quite long - too long to be a “simple”
presentation of the data.
To remedy this situation, a grouped frequency
distribution table is used.
Grouped Frequency Distribution (cont.)
In a grouped table, the X column lists groups of scores,
called class intervals, rather than individual values.
These intervals all have the same width, usually a simple
number such as 2, 5, 10, and so on.
Each interval begins with a value that is a multiple of the
interval width. The interval width is selected so that the
table will have approximately ten intervals.
Frequency Distribution Graphs
In a frequency distribution graph, the score
categories (X values) are listed on the X axis and the
frequencies are listed on the Y axis.
When the score categories consist of numerical
scores from an interval or ratio scale, the graph
should be either a histogram or a polygon.

71
Histograms
In a histogram, a bar is centered above each score (or
class interval) so that the height of the bar
corresponds to the frequency and the width extends
to the real limits, so that adjacent bars touch.

72
Polygons
In a polygon, a dot is centered above each score so
that the height of the dot corresponds to the
frequency. The dots are then connected by straight
lines. An additional line is drawn at each end to bring
the graph back to a zero frequency.

74
Bar graphs
When the score categories (X values) are measurements
from a nominal or an ordinal scale, the graph should be a
bar graph.
A bar graph is just like a histogram except that gaps or
spaces are left between adjacent bars.

76
Relative frequency
Many populations are so large that it is impossible to know
the exact number of individuals (frequency) for any specific
category.
In these situations, population distributions can be shown
using relative frequency instead of the absolute number of
individuals for each category.

78
Smooth curve
If the scores in the population are measured on an
interval or ratio scale, it is customary to present the
distribution as a smooth curve rather than a jagged
histogram or polygon.
The smooth curve emphasizes the fact that the
distribution is not showing the exact frequency for
each category.

80
Frequency distribution graphs
Frequency distribution graphs are useful because they show
the entire set of scores.
At a glance, you can determine the highest score, the lowest
score, and where the scores are centered.
The graph also shows whether the scores are clustered
together or scattered over a wide range.

82
Shape
A graph shows the shape of the distribution.
A distribution is symmetrical if the left side of the graph is (roughly)
a mirror image of the right side.
One example of a symmetrical distribution is the bell-shaped normal
distribution.
On the other hand, distributions are skewed when scores pile up on
one side of the distribution, leaving a "tail" of a few extreme values
on the other side.

83
Positively and Negatively
Skewed Distributions
In a positively skewed distribution, the
scores tend to pile up on the left side of the
distribution with the tail tapering off to the
right.
In a negatively skewed distribution, the
scores tend to pile up on the right side and
the tail points to the left.

84
Stem-and-Leaf Displays

A stem-and-leaf display provides a very efficient

method for obtaining and displaying a frequency
distribution.
Each score is divided into a stem consisting of the first
digit or digits, and a leaf consisting of the final digit.
Finally, you go through the list of scores, one at a time,
and write the leaf for each score beside its stem.
The resulting display provides an organized picture of
the entire distribution. The number of leaves beside
each stem corresponds to the frequency, and the
individual leaf identify the individual scores.
86
Steps in constructing frequency
distribution
❖Find out the highest score and the lowest score. Then determine the Range which is highest
score minus lowest score.
❖Second step is to decide the number and size of the groupings to be used. (It should be
between 5 and 20.)
❖Prepare the class intervals. It is natural to start the intervals with their lowest scores at
multiples of the size of the intervals. For example when the interval is 3, to start with 9, 12, 15,
18 etc. when the interval is 5, to start with 5, 10, 15, 20 etc.
❖Once we have adopted a set of class intervals, we have to list them in their respective class
intervals. For that we have to put tallies in their proper intervals.
❖Make a column to the right of the tallies headed ‘f (frequency). Write the total number of
tallies on each class in­terval under column ‘f. The sum of the f column will be total number of
cases—’N’.
Example:
Tabu­late the scores into frequency distribution using a class interval of 5
units.
Find out the highest score and
the lowest score. Then
determine the Range which is
highest score minus lowest
score.
Interval Tally
Qualitative Data

Qualitative data uncovers valuable insights that can be used to

improve the user and customer experience. But how exactly do you
measure and analyze data that isn't quantifiable?

There are different qualitative data analysis methods to help you make
sense of qualitative feedback and customer insights, depending on
your business goals and the type of data you've collected.
What is qualitative data analysis?
Qualitative data analysis (QDA) is the process of organizing, analyzing, and interpreting
qualitative data—non-numeric, conceptual information and user feedback—to capture
themes and patterns, answer research questions, and identify actions to take to improve
your product or website.
Qualitative data often refers to user behavior data and customer feedback.
5 Qualitative Data Analysis Methods
Here are five methods of qualitative data analysis to help you make sense of
the data you've collected through customer interviews, surveys, and
feedback:
1. Content analysis
2. Thematic analysis
3. Narrative analysis
4. Grounded theory analysis
5. Discourse analysis
Content analysis
Content analysis is a research method that examines and quantifies the
presence of certain words, subjects, and concepts in text, image, video, or
audio messages. The method transforms qualitative input into quantitative
data to help you make reliable conclusions about what customers think of
your brand, and how you can improve their experience and opinion.

You can conduct content analysis manually or by using tools like Lexalytics
to reveal patterns in communications, uncover differences in individual or
group communication trends, and make connections between concepts.
How content analysis can help your team?
Content analysis is often used by marketers and customer service specialists, helping them
understand customer behavior and measure brand reputation.
For example, you may run a customer survey with open-ended questions to discover users’
concerns—in their own words—about their experience with your product. Instead of having to
process hundreds of answers manually, a content analysis tool helps you analyze and group
results based on the emotion expressed in texts.
Some other examples of content analysis include:
Analyzing brand mentions on social media to understand your brand's reputation
Reviewing customer feedback to evaluate (and then improve) the customer and user experience
(UX)
Researching competitors’ website pages to identify their competitive advantages and value
propositions
Interpreting customer interviews and survey results to determine user preferences, and setting
the direction for new product or feature developments
Content analysis benefits and
challenges
Content analysis has some significant advantages for small teams:
You don’t need to directly interact with participants to collect data
The process is easily replicable once standardized
You can automate the process or perform it manually
It doesn’t require high investments or sophisticated solutions

On the downside, content analysis has certain limitations:

When conducted manually, it can be incredibly time-consuming
The results are usually affected by subjective interpretation
Manual content analysis can be subject to human error
The process isn’t effective for complex textual analysis
Thematic analysis
Thematic analysis helps to identify, analyze, and interpret patterns in qualitative data, and can be
done with tools like Dovetail and Thematic.

While content analysis and thematic analysis seem similar, they're different in concept:
Content analysis can be applied to both qualitative and quantitative data, and focuses on
identifying frequencies and recurring words and subjects.
Thematic analysis can only be applied to qualitative data, and focuses on identifying patterns
and ‘themes’.
How thematic analysis can help your
team?
Thematic analysis can be used by pretty much anyone: from product marketers, to customer
relationship managers, to UX researchers.

For example, product teams can use thematic analysis to better understand user behaviors
and needs, and to improve UX. By analyzing customer feedback, you can identify themes
(e.g. ‘poor navigation’ or ‘buggy mobile interface’) highlighted by users, and get actionable
insight into what users really expect from the product.
Thematic analysis benefits and
challenges
Some benefits of thematic analysis:
It’s one of the most accessible analysis forms, meaning you don’t have to train your teams on it
Teams can easily draw important information from raw data
It’s an effective way to process large amounts of data into digestible summaries

And some drawbacks of thematic analysis:

In a complex narrative, thematic analysis can't capture the true meaning of a text
Thematic analysis doesn’t consider the context of the data being analyzed
Similar to content analysis, the method is subjective and might drive results that don't
necessarily align with reality
Narrative analysis
Narrative analysis is a method used to interpret research participants’ stories—things like
testimonials, case studies, interviews, and other text or visual data—with tools like Delve and
AI-powered ATLAS.ti.

Some formats narrative analysis doesn't work for are heavily-structured interviews and
written surveys, which don’t give participants as much opportunity to tell their stories in
their own words.
How narrative analysis can help your
team?
Narrative analysis provides product teams with valuable insight into the complexity of
customers’ lives, feelings, and behaviors.

In a marketing research context, narrative analysis involves capturing and reviewing

customer stories—on social media, for example—to get more insight into their lives,
priorities, and challenges.

This might look like analyzing daily content shared by your audiences’ favorite influencers on
Instagram, or analyzing customer reviews on sites like G2 or Capterra to understand
individual customers' experiences.
Narrative analysis benefits and
challenges
Businesses turn to narrative analysis for a number of reasons:
The method provides you with a deep understanding of your customers' actions—and the
motivations behind them
It allows you to personalize customer experiences
It keeps customer profiles as wholes, instead of fragmenting them into components that can be
interpreted differently

However, this data analysis method also has drawbacks:

Narrative analysis cannot be automated
It requires a lot of time and manual effort to make conclusions on an individual participant’s story
It’s not scalable
Grounded theory analysis
Grounded theory analysis is a method of conducting qualitative research to develop theories
by examining real-world data. The technique involves the creation of hypotheses and
theories through the collection and evaluation of qualitative data, and can be performed
with tools like MAXQDA and Delve.

Unlike other qualitative data analysis methods, this technique develops theories from data,
not the other way round.
How grounded theory analysis can help
your team?
Grounded theory analysis is used by software engineers, product marketers, managers, and
other specialists that deal with data to make informed business decisions.

For example, product marketing teams may turn to customer surveys to understand the
reasons behind high churn rates, then use grounded theory to analyze responses and
develop hypotheses about why users churn, and how you can get them to stay.

Grounded theory can also be helpful in the talent management process. For example, HR
representatives may use it to develop theories about low employee engagement, and come
up with solutions based on their findings.
Grounded theory analysis benefits and
challenges
Here’s why teams turn to grounded theory analysis:
It explains events that can’t be explained with existing theories
The findings are tightly connected to data
The results are data-informed, and therefore represent the proven state of things
It’s a useful method for researchers that know very little information on the topic

Some drawbacks of grounded theory are:

The process requires a lot of objectivity, creativity, and critical thinking from researchers
Because theories are developed based on data instead of the other way around, it's considered
to be overly theoretical, and may not provide concise answers to qualitative research questions
Discourse analysis
Discourse analysis is the act of researching the underlying meaning of qualitative data. It
involves the observation of texts, audio, and videos to study the relationships between the
information and its context.

In contrast to content analysis, the method focuses on the contextual meaning of language:
discourse analysis sheds light on what audiences think of a topic, and why they feel the way
they do about it.
How discourse analysis help your team
In a business context, the method is primarily used by marketing teams. Discourse analysis
helps marketers understand the norms and ideas in their market, and reveals why they play
such a significant role for their customers.

Once the origins of trends are uncovered, it’s easier to develop a company mission, create a
unique tone of voice, and craft effective marketing messages.
Discourse analysis benefits and
challenges
Discourse analysis has the following benefits:
It uncovers the motivation behind your customers’ or employees’ words, written or spoken
It helps teams discover the meaning of customer data, competitors’ strategies, and
employee feedback

But it also has drawbacks:

Similar to most qualitative data analysis methods, discourse analysis is subjective
The process is time-consuming and labor-intensive
It’s very broad in its approach
Which qualitative data analysis method
should you choose?
While the five qualitative data analysis methods we list above are aimed at processing data
and answering research questions, these techniques differ in their intent and the approaches
applied.

Choosing the right analysis method for your team isn't a matter of preference—selecting a
method that fits is only possible when you define your research goals and have a clear
intention. Once you know what you need (and why you need it), you can identify an analysis
method that aligns with your objectives.
References
https://qph.cf2.quoracdn.net/main-qimg-288dd9a2cd15d63a171646bcd1371e6f-lq
https://www.hotjar.com/qualitative-data-analysis/methods/#
https://www.slideshare.net/donthuraj/basics-of-statistics-53905627
https://www.scribbr.com/methodology/external-validity/
https://www.scribbr.com/methodology/internal-validity/
https://www.scribbr.com/frequently-asked-questions/correlational-vs-experimental-research/
#:~:text=In%20an%20experimental%20design%2C%20you,without%20manipulating%20any
%20of%20them.
https://www.slideshare.net/menhaz/stat-1163-statistics-in-environmental-science-116147156
https://1.bp.blogspot.com/-QRSuSJ8kMbo/WPvYUzbRDAI/AAAAAAAACCU/
nja0lKnSINk2eNjqpd11JVYJ7hbMcgAdgCLcB/s1600/ronald%2Bfisher%2B2.jpg