Statistics for Data Science -1

Statistics for Data Science -1

Introduction and types of data

Usha Mohan

Indian Institute of Technology Madras

1/ 10
Statistics for Data Science -1

Learning objectives
1. What is statistics?
I Descriptive statistics, inferential statistics.
I Distinguish between a sample and a population.
2. Understand how data are collected.
I Identify variables and cases (observations) in a data set
3. Types of data-
I classify data as categorical(qualitative) or
numerical(quantitative) data.
I Understand cross-sectional versus time-series data.
I Measurement scales
4. Creating data sets; Downloading and manipulating data sets;
working on subsets of data.
5. Framing questions that can be answered from data.
2/ 10
Statistics for Data Science -1

Introduction
Basic definitions
Population and sample

Understanding data

Classification of data
Categorical and numerical
Cross-sectional versus time-series data
Scales of measurement

3/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Scales of measurement

I Data collection requires one of the following scales of

measurement: nominal, ordinal, interval, or ratio.

4/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Nominal scale of measurement

I When the data for a variable consist of labels or names used

to identify the characteristic of an observation, the scale of
measurement is considered a nominal scale.

5/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Nominal scale of measurement

I When the data for a variable consist of labels or names used

to identify the characteristic of an observation, the scale of
measurement is considered a nominal scale.
I Examples: Name, Board, Gender, Blood group etc.

5/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Nominal scale of measurement

I When the data for a variable consist of labels or names used

to identify the characteristic of an observation, the scale of
measurement is considered a nominal scale.
I Examples: Name, Board, Gender, Blood group etc.
I Sometimes nominal variables might be numerically coded.

5/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Nominal scale of measurement

I When the data for a variable consist of labels or names used

to identify the characteristic of an observation, the scale of
measurement is considered a nominal scale.
I Examples: Name, Board, Gender, Blood group etc.
I Sometimes nominal variables might be numerically coded.
I For example: We might code Men as 1 and Women as 2. Or
Code Men as 3 and Women as 1. Both codes are valid.

5/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Nominal scale of measurement

I When the data for a variable consist of labels or names used

to identify the characteristic of an observation, the scale of
measurement is considered a nominal scale.
I Examples: Name, Board, Gender, Blood group etc.
I Sometimes nominal variables might be numerically coded.
I For example: We might code Men as 1 and Women as 2. Or
Code Men as 3 and Women as 1. Both codes are valid.
I There is no ordering in the variable.

5/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Nominal scale of measurement

I When the data for a variable consist of labels or names used

to identify the characteristic of an observation, the scale of
measurement is considered a nominal scale.
I Examples: Name, Board, Gender, Blood group etc.
I Sometimes nominal variables might be numerically coded.
I For example: We might code Men as 1 and Women as 2. Or
Code Men as 3 and Women as 1. Both codes are valid.
I There is no ordering in the variable.
I Nominal: name categories without implying order

5/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Ordinal scale of measurement

I Data exhibits properties of nominal data and the order or rank

of data is meaningful, the scale of measurement is considered
a ordinal scale.

6/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Ordinal scale of measurement

I Data exhibits properties of nominal data and the order or rank

of data is meaningful, the scale of measurement is considered
a ordinal scale.
I Each customer who visits a restaurant provides a service
rating of excellent, good, or poor.
I The data obtained are the labels—excellent, good, or
poor—the data have the properties of nominal data.
I In addition, the data can be ranked, or ordered, with respect to
the service quality.

6/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Ordinal scale of measurement

I Data exhibits properties of nominal data and the order or rank

of data is meaningful, the scale of measurement is considered
a ordinal scale.
I Each customer who visits a restaurant provides a service
rating of excellent, good, or poor.
I The data obtained are the labels—excellent, good, or
poor—the data have the properties of nominal data.
I In addition, the data can be ranked, or ordered, with respect to
the service quality.
I Ordinal – name categories that can be ordered

6/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Interval scale of measurement

I If the data have all the properties of ordinal data and the
interval between values is expressed in terms of a fixed unit of
measure, then the scale of measurement is interval scale.

7/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Interval scale of measurement

I If the data have all the properties of ordinal data and the
interval between values is expressed in terms of a fixed unit of
measure, then the scale of measurement is interval scale.
I Interval data are always numeric. Can find out difference
between any two values.

7/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Interval scale of measurement

I If the data have all the properties of ordinal data and the
interval between values is expressed in terms of a fixed unit of
measure, then the scale of measurement is interval scale.
I Interval data are always numeric. Can find out difference
between any two values.
I Ratios of values have no meaning here because the value of
zero is arbitrary.

7/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Interval scale of measurement

I If the data have all the properties of ordinal data and the
interval between values is expressed in terms of a fixed unit of
measure, then the scale of measurement is interval scale.
I Interval data are always numeric. Can find out difference
between any two values.
I Ratios of values have no meaning here because the value of
zero is arbitrary.
I Interval:
numerical values that can be added/subtracted (no absolute zero)

7/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Example: temperature
I Suppose the response to a question on how hot the day is
comfortable and uncomfortable, then the temperature as a
variable is nominal.

8/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Example: temperature
I Suppose the response to a question on how hot the day is
comfortable and uncomfortable, then the temperature as a
variable is nominal.
I Suppose the answer to measuring the temperature of a liquid
is cold, warm, hot - the variable is ordinal.

8/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Example: temperature
I Suppose the response to a question on how hot the day is
comfortable and uncomfortable, then the temperature as a
variable is nominal.
I Suppose the answer to measuring the temperature of a liquid
is cold, warm, hot - the variable is ordinal.
I Example: Consider a AC room where temperature is set at
20°C and the temperature outside the room is 40°C. It is
correct to say that the difference in temperature is 20°C, but it
is incorrect to say that the outdoors is twice as hot as indoors.

8/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Example: temperature
I Suppose the response to a question on how hot the day is
comfortable and uncomfortable, then the temperature as a
variable is nominal.
I Suppose the answer to measuring the temperature of a liquid
is cold, warm, hot - the variable is ordinal.
I Example: Consider a AC room where temperature is set at
20°C and the temperature outside the room is 40°C. It is
correct to say that the difference in temperature is 20°C, but it
is incorrect to say that the outdoors is twice as hot as indoors.
I Temperature in degrees Fahrenheit or degrees centigrade is an
interval variable. No absolute zero.
Celsius Fahrenheit
Freezing point 0 32
Boiling point 100 212
8/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Ratio scale of measurement

I If the data have all the properties of interval data and the
ratio of two values is meaningful, then the scale of
measurement is ratio scale.

9/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Ratio scale of measurement

I If the data have all the properties of interval data and the
ratio of two values is meaningful, then the scale of
measurement is ratio scale.
I Example: height, weight, age, marks, etc.

9/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Ratio scale of measurement

I If the data have all the properties of interval data and the
ratio of two values is meaningful, then the scale of
measurement is ratio scale.
I Example: height, weight, age, marks, etc.
I Ratio: numerical values that can be added, subtracted,
multiplied or divided (makes ratio comparisons possible)

9/ 10
Statistics for Data Science -1
Classification of data
Scales of measurement

Summary

10/ 10