ASSIGNMENT No.

02
Methods of Social Research (4688) Msc Sociology
Spring, 2022
Q.1 How do we perform the content analysis in research? What are the advantages and
disadvantages of using content analysis techniques? (25)

Content analysis is a qualitative research tool or technique that is used widely to analyze the content
and its features. It is an approach used to quantify qualitative information by sorting data and
comparing different pieces of information to summarize it into useful information. Holsti (1969) has
defined content analysis as, “Any technique for making inferences by objectively and systematically
identifying specified characteristics of messages. The content can vary from a simple word, text,
picture to social media data, books, journals, and websites. The objective of content analysis is to
present the qualitative content in the form of objective and quantitative information.

In content analysis, qualitative data that is collected for research will be analyzed systematically to
convert it into quantitative data. Content analysis is different from the other research, as it does not
collect data from people directly. Instead, it is the study of data that is already recorded in social
media, text, or books or any other physical or virtual forms. Content analysis has been used
increasingly by organizations to surpass surface-level analysis by using computers and machine
learning for automatic labeling and coding of the text.

Example of Content Analysis

Research (Cruz and Lee 2014) was conducted to recognize the challenges that many companies are
facing in developing Twitter campaigns. Content analysis was conducted to analyze the Twitter feeds of
internationally recognized companies. Various terms were grouped based on Aaker’s five brand
personality dimensions framework that is used to describe the traits of a given brand into five
dimensions:

Sincerity, excitement, competence, sophistication, and ruggedness. Sentiment analysis was also
conducted using the Lexicoder Sentiment Dictionary that is used to perform simple content analysis.
The results of the content analysis highlighted two essential factors, word choice and media type, for
the success of a marketing campaign on Twitter. These two factors should be taken into consideration
while developing a social media marketing plan. Content analysis has seen rapid growth and
acceptance due to the computer-aided text analysis. It has become easier to perform content analysis
due to the easy availability of electronic messages, thereby making it easier to analyze with precision
and speed.

Development of Content Analysis

Content analysis can be dated back to 1920s in the United States of America, where a large quantity of
data from mass media such as radio and newspaper was analyzed. For example, the number of times a
text, such as the name of a political party, was repeated in the newspaper was counted and analyzed.
However, this was not foolproof as it could not identify the latent meaning, and it just counted the
number of times a word repeated. Later in 1972, Jurgen Ritsert developed a process that was able to
identify the latent meaning and ideological contents by applying quantitative analysis. Ever since then,
content analysis has been used to interpret the text and to arrive at a valid conclusion With the advent
of the internet and technological advancement, content analysis has gained particular significance.
Over the years, many things have changed, and a few have remained constant. Computers are now
used to gather, analyze, and present a massive amount of data with lightning speed and accuracy.

Content analysis of the big data produced by social media, online content, and mobile devices has
taken higher significance. Content analysis has taken over as the most popular method compared to
surveys, interviews, and other forms of analysis. Never has content analysis received more
considerable attention in many fields of research than at present. It has been embraced and is
extending far and wide into many disciplines.

1
Goals/Objectives of Content Analysis

The purpose of content analysis is to ‘read between the lines.’ It aims to determine answers to
questions where the text implies something, and not necessarily explicit. Content analysis is a research
that can analyze human communications, how people plan their lives, what people know about
something, and how people react to something. Content analysis has become an alternative to the
traditional inquiries of the mass media, which was then used for public opinion research. The content
analysis employs methods to examine the data, images, printed text, sounds, social media, articles,
books, journals and the web – mainly to understand what people mean, what people enable and what
does the information conveyed by them say to the business or the society at large.

Example

The content analysis helped Nescafé Dolce Gusto to improve their campaign performance by 400%.
The goal of content analysis was to find and create a multi-channel marketing strategy that can
attract coffee lovers. They started by conducting content analysis. They rolled out market research into
the coffee lover community online. They collected insights from the coffee lovers and used the
information to design a suitable marketing campaign that took into account factors such as the taste
and needs of the coffee lovers. As a result of this content analysis, Nescafé Dolce Gusto was able to
increase its Facebook engagement by 400%.

The objective of content analysis:

 To Identify the implied aspects of the content

 To sketch the characteristics of the content

 To analyze and present significant findings of content, clearly and effectively

 To simplify unstructured content

 To identify trends and relationships

 To spot the intentions of individuals or groups of people or an institution

 To describe attitudinal and behavioral responses to communications

 To determine the psychological or emotional state of a group of people

 To justify an argument

To summarize, content analysis is conducted to yield inferences from different kinds of content that
could be text, pictures, and social media data.

Sources of Content Analysis

Content analysis forms the bridge between quantitative and qualitative research methods, where some
of the organizational issues that are very difficult to study, such as the organization behavior, human
resources, and customer issues can be considered. By analyzing the presence of certain words and text
within a given qualitative data, the relationship of words and pictures, the researchers can make
inferences to many vital aspects such as the audience, behavior, culture, and the level of satisfaction.

The sources of data for content analysis are primarily two types:

1. Offline

The offline content analysis is based on books, journals, essays, interviews, research notes, open-
ended questions, and directories. The sample from offline sources will represent the whole universe.
However, in many cases, offline data can be outdated.

2
2. Online

With the rapid growth of the internet, online data sources have acquired significance. The online
conversations, social media comments, product reviews, and customer feedback is collected from the
most recent and updated references, thereby making the data source more relevant. With the above
information, it will be much easier to analyze the post and take decisions about the next steps.

Uses of Content Analysis

Content analysis can be applied to analyze any piece of content that is written or verbal. Content
analysis is involved in a variety of fields such as politics, human behavior, marketing, literature, health,
psychology, and much more. Content analysis is also displaying a close relation between the linguistic
factors and psychological aspects, thereby leading to the development of artificial intelligence.

Examples of the Uses of Content Analysis

For example, a brand can discover emerging trends with the use of content analysis. Content from
online conversations is obtained from various sources such as news, feedbacks, blogs, tickets, online
discussion, social media, and reviews.
Once the data is available, the data has to be sliced and diced using algorithms and proven
mathematical models. Topics, relationships, and tone intensities are analyzed to identify patterns,
correlations, and inferences at multiple levels.

As content analysis deals with text, numbers, comments, statistics, and much more measurable facts,
it is used for forecasting, trend analysis, and drawing logical strategies. It is used widely to remove the
ambiguity factor and to get rid of opinions and guesswork. Content that you gather is subjective, and
hence using it to analyze and define it more quantitatively helps to arrive at decisions. Therefore,
content analysis is essential. It has the following benefits:

 Establishes proof of the reliability of the data

 Allows both quantitative and qualitative analysis

 Offers valuable insights into history by analyzing information

 Provides analytical insight into human thought and language

 To Identify the trends and intentions of an individual or a group

Understands both human behavior and the use of language, and their relationship

The use of content analysis depends on how you use it. For example, when you release an article on
your blog page, content analysis will help to understand the further journey. How many people read it,
how many liked it, how many shared it, how many people visited your website after reading your
article, and how much sales increased post releasing this article. When you look at the content analysis
reports, you can identify several areas that are doing well and the specific regions where you will have
to devote attention to its improvement. All this would not have happened without content analysis.

Approaches to Content Analysis

Content analysis can be performed in three different methods: conventional, directed, and summative.
Though there are three different approaches, they intend to understand and analyze the meaning of
content. They do have specific differences, which is predominantly in the coding system.

1. Conventional Content Analysis

Also called inductive category development, this approach is used when the existing theory or research
on any given subject is limited. Here data is used as a source to arrive at categories rather than using
any of the pre-existing categories. In this approach, the researches rely entirely on the data to arrive
at new insights. Most of the qualitative analysis methods use this approach to study and analyze.

3
2. Directed Content Analysis

In this approach, research is based on an existing theory. This approach of content analysis is used to
validate or further analyze the already existing theory. This method can be done in two ways. One
way is to start coding the data based on the predetermined codes from the earlier approach. Another
way is to review the existing codes and assign new codes for the text that could not be categorized in
the previous method. The directed content analysis aims to focus and extend the pre-existing theory
to determine the key concepts.

3. Summative Content Analysis

In this approach, the words of text will be initially counted and compared, followed by further
interpretation of the content. The summative content analysis aims at finding the underlying meanings
of the text or words. In this approach, the study starts by searching for a particular text and counting
the number of times it appears and further tries to understand the fundamental context for the use of
the words either explicit or in its indirect terms. Summative content analysis is a nonreactive method
of studying the phenomenon of interest. The approaches of content analysis depend on the research
purposes that may need different research designs and various techniques of analysis. The research
should take the choice of using a conventional or summative or directed approach after considering the
purpose and the methods.

Q.2 When researcher needs measurement and testing in social science research? Briefly
explain different levels of measurements with examples. (25)

Not all data is created equally. Some is quantitative, and some is qualitative. Some is continuous and
some is discrete. Another way to separate data is to look at what is being measured. To do this there
are four levels of measurement: nominal, ordinal, interval and ratio. Different levels of measurement
call for different statistical techniques. For example, it makes no sense whatsoever to find the mean,
and median of a list of Social Security numbers.

Nominal Level of Measurement

The nominal level of measurement is the lowest of the four ways to characterize data. Nominal means
"in name only" and that should help to remember what this level is all about. Nominal data deals with
names, categories, or labels. Data at the nominal level is qualitative. Colors of eyes, yes or no
responses to a survey, and favorite breakfast cereal all deal with the nominal level of measurement.
Even some things with numbers associated with them, such as a number on the back of a football
jersey, are nominal since it is used to "name" an individual player on the field. Data at this level can't
be ordered in a meaningful way, and it makes no sense to calculate things such as means and
standard deviations.

Ordinal Level of Measurement

The next level is called the ordinal level of measurement. Data at this level can be ordered, but no
differences between the data can be taken that are meaningful. Here you should think of things like a
list of the top ten cities to live. The data, here ten cities, are ranked from one to ten, but differences
between the cities don't make much sense. There's no way from looking at just the rankings to know
how much better life is in city number 1 than city number 2. Another example of this are letter grades.
You can order things so that A is higher than a B, but without any other information, there is no way of
knowing how much better an A is from a B. As with the nominal level, data at the ordinal level should
not be used in calculations.

Interval Level of Measurement

The interval level of measurement deals with data that can be ordered, and in which differences
between the data does make sense. Data at this level does not have a starting point. The Fahrenheit
and Celsius scales of temperatures are both examples of data at the interval level of measurement.
You can talk about 30 degrees being 60 degrees less than 90 degrees, so differences do make sense.
However 0 degrees (in both scales) cold as it may be does not represent the total absence of
temperature.

4
Data at the interval level can be used in calculations. However, data at this level does lack one type of
comparison. Even though 3 x 30 = 90, it is not correct to say that 90 degrees Celsius is three times as
hot as 30 degrees Celsius. Ratio Level of Measurement

The fourth and highest level of measurement is the ratio level. Data at the ratio level possess all of the
features of the interval level, in addition to a zero value. Due to the presence of a zero, it now makes
sense to compare the ratios of measurements. Phrases such as "four times" and "twice" are meaningful
at the ratio level. Distances, in any system of measurement give us data at the ratio level. A
measurement such as 0 feet does make sense, as it represents no length. Furthermore 2 feet is twice
as long as 1 foot. So ratios can be formed between the data. At the ratio level of measurement, not
only can sums and differences be calculated, but also ratios. One measurement can be divided by any
nonzero measurement, and a meaningful number will result.

Think Before You Calculate

Given a list of Social Security numbers, it's possible to do all sorts of calculations with them, but none
of these calculations give anything meaningful. What's one Social Security number divided by another
one? A complete waste of your time, since Social Security numbers are at the nominal level of
measurement. When you are given some data, think before you calculate. The level of measurement
you're working with will determine what it makes sense to do.

The central methodological problem in any type of research concerns how the relevant data is to be
collected. In the field of second language research, a very popular data collection technique has been
the use of various types of questionnaires. But as Zoltan Dornyei of the University of Nottingham
points out in his brand new book, Questionnaires in Second Language Research, "there does not seem
to be sufficient awareness in the profession about the theory of questionnaire design and processing.
The usual--and in most cases false--perception is that anybody with a bit of common sense can
construct a good questionnaire.

Erroneous perception

Dornyei believes that this erroneous perception is based on a lack of recognition "that there is
considerable relevant knowledge and experience accumulated in various branches of the social sciences
(e.g. psychometrics, social psychology, sociology. The purpose of his book is to present practical
information on how to make and use quality questionnaires and thereby "save ourselves a lot of
trouble.

The book is short, but very much to the point, and consists of four chapters. The first 'Questionnaires
in Second Language Research,' is a general introduction to questionnaires. Dornyei (quoting Brown,
2001) defines a questionnaire as "any written instruments that present respondents with a series of
questions or statements to which they react either by writing out their answers or selecting from
among existing answers." Dornyei also notes that questionnaires should not be confused with tests or
any type of discourse completion. He next shows that questionnaires yield three types of data about
the respondent: attitudinal, behavioral, and factual.

Questionnaires are especially valuable

Dornyei believes that questionnaires are especially valuable because they are efficient "in terms of (a)
researcher time, (b), researcher effort, and (c), financial resources." (p. 9) He also examines the major
drawbacks of questionnaires: the simplicity of answers yielded, the problem of respondents who are
unmotivated or unreliable, the famous halo effect, the acquiescence and prestige biases, issues
concerning self-deception and respondent literacy, and the effect of fatigue in cases where the
questionnaire is long.
'Constructing the Questionnaire,' discusses the main parts and general features of a questionnaire.
Dornyei also explains rating scales, both multi-item and closed-ended, as well as open-ended
questions. He then draws up a set of simple rules, which describe how to write good items, how to
group and order the items, and how to pilot a questionnaire and perform item analysis.The next
chapter, 'Administrating the Questionnaire,' concerns the topics of sample size, questionnaire
confidentiality, and administration. Dornyei also gives tips on how to maximize both the quantity and
quality of participant responses.

5
In chapter four, 'Processing Questionnaire Data,' Dornyei tells how to code questionnaire data, and put
data into a computer file, as well as ways to process closed questions and do content analysis of
questions that are open-ended. He also lists several computer programs that are good for processing
data. He closes the chapter by writing about how to report questionnaire data, covering the areas of
"general guidelines about what to report and how. The technical information about the survey that
needs to be included in the professional report to accompany the actual findings. (And) Presentation
methods that can make the data more reader-friendly and digestible.

The final section of the book is a 39-part checklist of what Dornyei thinks are the most important
elements of questionnaire design, implementation, and analysis. The book also contains an appendix
which lists six pages of published L2 questionnaires, on diverse topics ranging from attitudes and
language anxiety, to language learning styles and teacher self-evaluation.

Second Language

All in all, Questionnaires in Second Language Research is a brisk, no nonsense introduction to a

complex subject. Dornyei writes clearly and concisely and can explain complicated technical matters in
a simple manner. two problems with the book's content. First, it would have been extremely helpful if
Dornyei had written a chapter that had taken the reader through the entire process of making,
administrating, and processing a questionnaire. Dornyei's numerous pointers and checklists are very
important and informative, but they are ultimately not enough, he should have also supplemented his
presentation with more concrete and detailed examples. The book jacket states that Dornyei has been
using questionnaires in his research on language learning motivation for over 15 years, that he has
surveyed almost 10,000 respondents, and that he has published over 30 articles based upon this work.
It would have been very informative if Dornyei would have described in detail how he went about
designing and using one of his research questionnaires.

In addition, Dornyei fails to mention anywhere in this work a key problem which plagues any second
language researchers who use questionnaires and conducts their research overseas or writes the
questionnaires in the native language of the participants: namely the question of how does one go
about making sure that the items used are correctly translated into the native language? Since cross-
cultural questionnaires and questionnaires done in the first language of the respondents have recently
become quite common in research journals, the topic is of general concern. To use a personal example,
for the past two and a half years, involved with a colleague in Japan in an extended research project
investigating Chinese and Japanese attitudes towards English in relation to issues concerning language
purism and identity. One major and continual problem we have had with the construction of our
questionnaires involves ensuring that each of our questionnaires precisely match the other in terms of
meaning for all of the items we use. Thus, it is problematic that Dornyei has neglected to discuss this
important aspect of questionnaire design.

Q.3 How a good report writing skill can make the thesis document more effective? What are
the different elements of a report in social sciences? Explain. (25)

Report writing formats

The techniques you use in writing reports for the most part include the same kinds of techniques used
in producing any written document. There are exceptions, such as the instance of a lab report that may
include nothing except raw data. Generally, however, you will follow the same writing process as you
would use to write any kind of document.

Stages

The first technique you need to understand in writing an effective report is recognizing the four stages
of the report-writing process. These stages are report preparation; report planning or arrangement of
the report order of presentation; writing the report; and honing or revising the report so that the final
draft effectively and efficiency conveys the message you need the report to convey, as described at the
School of Computing and Technology at the Eastern Mennonite University website.

Preparation

Certain reports with limited purposes may involve only two preparation stage activities both focused on
an identification process. These techniques involve identifying the type of report you need to write and

6
then identifying the report’s audience along with the audience’s needs in terms of the information you
will communicate. You will likely write a report in a different style for a college professor than you will
for a technical manager in a business environment. Other reports, however, can require additional
preparation activities. Particularly, you may need to do such research activities as reading material the
report requires or interviewing experts in a professional discipline. You need a good sense of the data
you expect to include to advance from the preparation to the planning, or arrangement, phase of the
report-writing process.

Planning

Some professional business writers regard the planning phase as equal to or even greater than the
actual report writing phase when it comes to writing effective business documents. Good planning can
help a writer more easily write an effective report by laying out the sections in advance in a logical,
orderly way. This allows the writer to in a sense simply “fill in the blanks” during the writing process.
When you have all, or even most, of the data on which you will base your report in hand, plan how you
want to organize that data, and list the section headings you need to incorporate in your report.

Writing

Group your research data according to the section headings, and write each section to convey a
message consistent with the purpose of the report and the purpose of each section.

Revision

During revision, review the original report draft for any flaws that make the report less effective than it
could be. In this stage, look for excess verbiage, insufficient clarity in the sentences and paragraphs,
jargon or excessively technical language and use of passive versus active voice. Rewrite your report as
many times as necessary until you eliminate problems in any of these areas. The report then should
flow together so as to communicate the report's purpose in clear, concise phrasing along with an
orderly presentation of information.

Ten Tips for Writing Reports Efficiently

Try using these 10 tips the next time you write a police report, and you’ll be able to complete your
paperwork more quickly and efficiently. And that’s only one of the benefits. Anyone who reads your
report (a lieutenant, reporter, or attorney) will be impressed by your professionalism and writing
ability. You will have avoided outdated (and time-wasting) wordiness that characterizes so much police
writing.

1. Use names and pronouns (I, he, her) when you write about yourself and others at the scene. Avoid
outdated expressions like “this officer” and “the abovementioned witness” or “victim 1.”

In the past some officers were taught that impersonal terminology guaranteed objectivity and
accuracy. Not true! You have the same integrity whether you’re calling yourself “I” or “this officer.”
And think about this: if you were testifying in court, and sworn to tell the truth, you would use
everyday language (“I,” “me”) in your testimony. Follow the same practice in your reports.

2. Limit yourself to one idea per sentence.

Short, straightforward sentences are easy to read and understand, saving time for everyone. (You’ll
especially appreciate this time-saving tip when you’re reviewing a report to prepare for a court
hearing.) The longer a sentence is, the more likely you are to make an error.

3. Start every sentence with a person, place, or thing.

Normal sentence structure in English begins with a noun, and the grammar is simple: Just put a period
at the end. Complicated sentences, on the other hand, require complicated punctuation, and they open
the door to sentence errors.

4. Try to limit yourself to three commas per sentence.

7
If a sentence has more than three commas, it’s probably too complicated to be read easily, and it may
contain usage or punctuation errors.

5. Be as clear and specific as possible.

“Contacted” is vague: Did you visit, phone, or email the witness? “Residence” is just as confusing:
House, apartment, mobile home, condo? Always strive for clarity.

6. Use simple language.

“Since” is easier to understand (and write) than “inasmuch as.” “Pertaining to” is a fancy (and time-
wasting) way to write “about.”

7. Stick to observable facts.

Conclusions, guesses, hunches, and other thought processes do not belong in a report. Stick to the
facts. A statement like “He was aggressive” won’t stand up in court. You can, however, write “Jackson
clenched his fists and kicked a chair.”

8. Write in paragraphs.

Organizing information in groups (what each witness told you, what actions you did, what evidence you
collected) has two important benefits: Your report is more logical, and it’s easier to read and
understand later on.

9. Use active voice.

A widespread (and mistaken) notion in law enforcement says that passive voice guarantees objectivity
and accuracy. False. Writing a sentence like “A revolver was seen under the nightstand” does not
guarantee that you’re telling the truth. It’s much simpler just to write “I saw a revolver under the
nightstand.” That’s what you would say in court, isn’t it?

10. Use bullet style.

You’ve probably been writing shopping lists all your life. Use the same format when you’re recording
several pieces of related information, like this:

Larry Holden told me:

 He and Sharon have been “fighting a lot”

 She was drunk when he came home from work
 She threw a package of frozen chicken at him
 He didn’t touch her

These 10 tips can transform your report writing, making you more professional, more up-to-date, and
more efficient. Don’t try to follow all 10 right away. Choose one or two to focus on until they become
second nature; then go on to one or two more. Keep learning and growing until you’ve become
proficient with all 10.

One more suggestion: Share what you’re learning with other officers: Your entire agency will benefit,
and you’ll be developing your leadership skills. When report writing improves, everyone, especially
you, benefits.

Q.4 Write comprehensive note on following: (13+12)

i. Presentation of research data

Data comes in a variety of forms. Learn different types of data, including analog vs. digital, character
strings, numeric, boolean, date and time, and multimedia.

8
Data Types
Computer systems work with different types of digital data. In the early days of computing, data
consisted primarily of text and numbers, but in modern-day computing, there are lots of different
multimedia data types, such as audio, images, graphics and video. Ultimately, however, all data types
are stored as binary digits. For each data type, there are very specific techniques to convert between
the binary language of computers and how we interpret data using our senses, such as sight and
sound.

Analog vs. Digital Data

There are two general ways to represent data: analog and digital. Analog data are continuous. They
are 'analogous' to the actual facts they represent. Digital data are discrete, broken up into a limited
number of elements. Nature is analog, while computers are digital. Many aspects of our natural world
are continuous in nature. For example, think of the spectrum of colors. This is a continuous rainbow
of an infinite number of shades.
Computer systems, on the other hand, are not continuous, but finite. All data are stored in binary
digits, and there is a limit to how much data we can represent. For example, a color image on a
computer has a limited number of colors - the number might be very large, but it is still finite.
Consider the example of color in a bit more detail. The very first monitor displays were essentially text
terminals with only a single color. White or light green text appeared on a black background.
Newer monitors used more colors, enough to represent basic images, but were still quite limited.
Modern displays have millions of colors and look much more natural. Still, the number of colors is
finite. The finite nature of data stored on a computer influences how different types are stored as
binary digits. You will see examples of this as the different types are discussed.

Character Strings
One of the most basic data types is plain text. In database terminology, this is referred to as
a character string, or simply a string. A string represents alphanumeric data. This means that a
string can contain many different characters, but that they are all considered as if they were text and
not put into calculations, even if the characters are numbers.

All of these fields are strings. Fields like the first and last name consist only of text characters, so it
makes sense they are stored as a string. The field for the street address contains both numbers and
characters and is also stored as a string. The student ID looks like a number, but it really represents a
code. It is not a number you want to do any calculations with, so it is stored as a string. Similarly, the
ZIP code looks like a number, but is also stored as a string.

Numeric Data Types

The second most important data type is numeric data. As a general rule, you store numbers only as a
numeric data type if they represent a count or measurement of some kind and if it makes sense to
perform calculations with them. A ZIP code is a number assigned to a geographic area by the postal
service. It would not make much sense to determine the average value for multiple ZIP codes.
There are several different types of numeric data. An integer is a numeric value without a decimal.
Integers are whole numbers and can be positive or negative. In a database, a distinction is made
between short and long integers, referring to how much data storage is used for the number. A short
integer is typically stored using 16 bits, which means that you can store up to 2^16, or 65,536 unique
values. For any number larger than that, you would need to use a long integer, which uses 32 bits or
more.
A number with a decimal is referred to as a decimal, a float or a double. The terminology varies
somewhat with the software being used. The term float comes from 'floating point,' which means you
can control where the decimal point is located. The term double refers to using double the amount of
storage relative to a float.
In the example table of students below, the field credits completed is an integer, while GPA is a
decimal. In both these examples, it would make sense to do calculations. For example, you could use

9
credits completed to calculate how many more credits a student needs to graduate. Or, you could
determine the average GPA for all the students.

Data analysis involves collecting and organizing data in order to reach a conclusion. Explore the
methods and techniques of data analysis, including qualitative and quantitative data analysis.

A Beginning Look at Data Analysis

Let's imagine that you have just enrolled in your first college course. After two days of class, your
professor assigns you a research assignment. You are to research which type of school system is
better, private or public. Immediately, you have an opinion of which system you feel is better, but you
realize that conducting research is not about your own personal opinion. Research is about gathering
data that you can analyze and use to come to some sort of conclusion. So, before you begin your data
collection, you realize that you have a lot to learn about the various methods and techniques of
gathering data.
Before we look at the methods and techniques of data analysis, lets first define what data analysis
is. Data analysis is the collecting and organizing of data so that a researcher can come to a
conclusion. Data analysis allows one to answer questions, solve problems, and derive important
information. So, for your assignment, you now know that the purpose of your assignment is to gather
enough information to come to a conclusion about which school system is better.

Methods of Data Analysis

Okay, you have decided to prove that public school is better than private school, but now you need to
figure out how you will collect the information and data needed to support that idea. There are two
methods that a researcher can pursue: qualitative and quantitative.
Qualitative research revolves around describing characteristics. It does not use numbers. A good
way to remember qualitative research is to think of quality.
Quantitative research is the opposite of qualitative research because its prime focus is numbers.
Quantitative research is all about quantity.

Qualitative Data Analysis Techniques

Qualitative research works with descriptions and characteristics. Let's look at some of the more
common qualitative research techniques:
Participant observation is when the researcher becomes part of the group that they are observing. This
technique can take a long period of time because the researcher needs to be accepted into the group
so that they observe data that is natural.

Boolean Data
The Boolean data type can only represent two values: true or false. Typically, a zero is used to
represent false and a one is used to represent true. In the example table of students, the field
Financial Aid is stored as a Boolean, since a student is classified as having financial aid or not.

Visual Representations of Data

Fields as diverse as finance, science, and politics all rely heavily on interpreting large amounts of
numerical data. This can be challenging for many people, so tables and graphs are used to present the
data in a way that is easy to understand.
Although there are many different types of tables and graphs, you can interpret them all if you
remember some basic guidelines. First, look at the title and any other labels. Some graphs may have a
legend that will give you important information. The axis labels can also tell you a lot about the
information presented in the graph. Next, look carefully at the graph to identify important information
that you need to know and then use that information to answer the question.
Let's look at a few different types of tables and graphs and see how to understand the information
presented in each.

10
Tables
Tables are the simplest way to represent data. A table compiles s all the data into columns and rows so
that it can be easily interpreted. This table below shows the average number of visitors per day at two
different lakes over an 8-year period.

By looking at the row and column titles, you see that each row corresponds to a different year, and
each column to a different lake. You could use a table like this to answer a lot of different questions.
For example, which lake had more visitors in Year 5?
The table shows that in Year 5, Lake Itawamba had an average of 74 visitors per day, and Lake
Tishomingo had an average of 85 visitors per day, so Lake Tishomingo had more visitors in Year 5.

Pictographs
A pictograph uses images to represent a certain number of items. For example, suppose you had a
lemonade stand andd wanted to record how much lemonade you sold each day. From this pictograph,
you can determine how many glasses of lemonade you sold on Thursday:

Did you get 40 glasses of lemonade? Great!

Looking at the pictograph, you can see that there are four lem
lemons
ons on the line that says Thursday, and
you also know that each lemon represents 10 glasses of lemonade. This means that you sold 40
glasses of lemonade on Thursday.
Line Graphs & Bar Graphs
A line graph plots individual data points as dots and connects th
them
em with lines. These are great if you
want to see how a quantity is changing over time. This line graph represents the distance an object
travels over time.

Even though there is no title, you can look at the axis labels and see that the x--axis measures time
and the y-axis
axis measures distance. Using this graph, you could find the position of the object at any
time. You could also use this graph to determine that the object was speeding up, because the
difference in distance between any two consecutive seco
seconds is increasing.
Bar graphs use bars of different heights to represent data. As with the other types of graphs, you first
want to read the title and the axis labels to make sure you understand what data is being presented.
In this bar graph, you can see that the x-axis has months and the y-axis
axis says temperature. Therefore,
this graph is showing you the average temperature during each month of the year at a certain location.

Now you are ready to analyze and use the data presented in the graph. Can you u use the graph to
calculate the temperature difference between the warmest and the coldest month of the year?

ii. Types of tests

Types of Statistical Tests

Statistics are the arrangement of statistical tests which analysts use to make inference from the data
given. These tests enables us to make decisions on the basis of observed pattern from data. There is a
wide rangeof statistical tests. The choice of which statistical test to utilize relies upon the structure of
data, the distribution of the data, and variable type.There are many different types of tests in statistics
like t-test,Z-test,chi-square
square test, anova test ,binomial test, one sample median test etc.
Choosing a Statistical test-

Parametric tests are used if the data is normally distributed .A parametric statistical test makes an
assumption about the population parameters and the distributions that the data came from. These types
of test includes t-tests,z-tests and anova tests, which assume data is from normal distribution.

Z-test- A z-test is a statistical test used to determine whether two population means are different when
the variances are known and the sample size is large. In z-test mean of the population is compared.The
parameters used are population mean and population standard deviation. Z-test is used to validate a
hypothesis that the sample drawn belongs to the same population.

Ho: Sample mean is same as the population mean(Null hypothesis)

Ha: Sample mean is not same as the population mean(Alternate hypothesis)

z = (x — μ) / (σ / √n),

where , x=sample mean, u=population mean, σ / √n = population standard deviation.

If z value is less than critical value accept null hypothesis else reject null hypothesis.

T-test-In t-test the mean of the two given samples are compared. A t-test is used when the population
parameters (mean and standard deviation) are not known.

Paired T-Test-Tests for the difference between two variables from the same population( pre- and post
test score). For example- In a training program performance score of the trainee before and after
completion of the program.

Independent T-test- The independent t-test which is also called the two sample t-test or student’s t-
test, is a statistical test that determines whether there is a statistically significant difference between
the means in two unrelated groups.For example -comparing boys and girls in a population.

One sample t-test- The mean of a single group is compared with a given mean. For example-to check
the increase and decrease in sales if the average sales is given.

t = (x1 — x2) / (σ / √n1 + σ / √n2),

where x1 and x2 are mean of sample 1 and sample 2 respectively.

ANOVA Test- Analysis of variance (ANOVA) is a statistical technique that is used to check if the means
of two or more groups are significantly different from each other. ANOVA checks the impact of one or
more factors by comparing the means of different samples. If we use a t-test instead of ANOVA test it
won’t be reliable as number of samples are more than two and it will give error in the result.

The hypothesis being tested in ANOVA is

13
Ho: All pairs of samples are same i.e. all sample means are equal

Ha: At least one pair of samples is significantly different

In anova test we calculate F value and compare it with critical value

F= ((SSE1 — SSE2)/m)/ SSE2/n-k, where

SSE = residual sum of squares

m = number of restrictions

k = number of independent variables

Non parametric statistical test- Non parametric tests are used when data is not normally distributed.
Non parametric tests include chi-square test.

Chi-square test( χ2 test)- chi-square test is used to compare two categorical variables. Calculating
the Chi-Square statistic value and comparing it against a critical value from the Chi-Square distribution
allows to assess whether the observed frequency are significantly different from the expected
frequency.

The hypothesis being tested for chi-square is-

Ho: Variable x and Variable y are independent

Ha: Variable x and Variable y are not independent.

Chi-square formula

where o=observed , e=expected.

14