You are on page 1of 19

DATA TYPES

Dr Freddie Mupambireyi (BSc, MSc, PhD) UZ


Senior Lecturer: Business Enterprise and Management Dept
Faculty of Business Management Sciences and Economics
pfcmupa@gmail.com

Compiled by Dr F Mupambireyi 1
DATA TYPES

Why is it important to know the type of data


you are going to collect in any decision
making process?

Compiled by Dr F Mupambireyi 2
DATA TYPES
Knowledge of the data to be collected is important
because of the following reasons:
 How will you assess its quality

 Will be able to determine how its going to be

collected (tools to be used for collection)


 How is it going to be summarised

 How it is going to be analysed (what packages or

approaches are going to be employed)


 How is it going to presented/interpreted 3
Compiled by Dr F Mupambireyi
DATA TYPES
 The type of data is determined by the nature of the
random variable which the data represents
Qualitative data
 These are data which yield categorical or non-

numeric responses.
 These data are descriptive in nature.

 The numbers representing the categories are

arbitrary
 Coded values can not be manipulated arithmetically
4
Compiled by Dr F Mupambireyi
DATA TYPES
Random Variable Data Codes and Response Categories
Area of residence 1) Rural area, 2) Urban area, 3) Farm, 4) Mine

Colour 1) Blue 2) Red 3) Green 4) Black

Sex 1) Male 2) Female

Level of education 1) No schooling 2) Primary 3) Secondary 4) Tertiary

There has been much 1) Strongly agree 2) Agree 3) Neutral 4) Disagree


development in some 5) Strongly disagree
rural areas since
independence
Compiled by Dr F Mupambireyi 5
Quantitative Data
 These are data which yield numeric responses

which are mainly generated through counting or


measurement.
 The data generated from a quantitative random

variable can be meaningfully manipulated using


conventional arithmetic operations (addition,
subtraction, division and multiplications)

Compiled by Dr F Mupambireyi 6
Quantitative Data
Random Variable Response Range Data

Age of Employee 18 to 65 years eg 45years

Distance to work 1 to 50 km eg 18km

Size of class 1 to 100 eg 60 students

Revenue 0 to 2000USD eg 1500 USD

Number of OVC reached 0 to 3000OVC eg 1000OVC

Number of complaints received 0 to 10 eg 3 7


Data Measurement Scales
 Measurement is usually is to do with physical
objects( numeric--- weight, height et) or abstract
(non-numeric-----personalities, feelings)
Nominal Scale(data)----- Properties
 Nominal scale is simply a system of assigning

number symbols to events in order to label or


differentiate them.
 The lables are of equal weighting

 No statistical operations can be meaningfully done

 Is the least powerful level of measurement 8


Compiled by Dr F Mupambireyi
Nominal scale data
 It indicates no order or distance relationship and has
no arithmetic origin.
 Categories of responses should be mutually
exclusive (mutual exclusivity)
 Categories of responses should be exhaustive
(principle of exhaustiveness)
 Codes should be uni-dimensional in each question

Compiled by Dr F Mupambireyi etc9


Nominal Scale Data

 Chi-square is the most common test where these


types of data are used
 A nominal scale simply describes differences
between things by assigning them to categories.
 Eg Sex: 1) Male 2) Female Colour: 1) Red 2) Blue
 Nationality: 1) Zimbabwean 2) Zambian 3) Egyptian
 Product bar codes, District of birth numbers, Student
identity numbers etc
Compiled by Dr F Mupambireyi 10
Data Measurement Scales
Ordinal Scale/Data
 The ordinal scale places events in an order or

involves ranking of responses


 Codes are of different weights or importance

 Distance between the codes can not be measured

 The real differences between adjacent ranks may

not be equal

Compiled by Dr F Mupambireyi 11
Data Measurement Scales
 Examples of Ordinal Scale
 Management level: 1) junior 2) middle 3) senior
 Size of T-shirt: 1) small 2) medium 3) large
 4) X-large
 Likert Scale; the responses are from one extreme to
the other extreme
 1) Strongly agree 2) Agree 3) Neutral
 4) Disagree 5) Strongly disagree
Compiled by Dr F Mupambireyi 12
Data Measurement Scales
Interval scale/data
 They are associated with quantitative data

 Differences between values can be measured

 It possesses both the order or ranking and distance

properties
 Does not possess an absolute origin (zero is

arbitrary or hypothetical)
 More powerful statistical measures can be

 used with interval scales.


13
Compiled by Dr F Mupambireyi
Data Measurement Scales
 Mean is the appropriate measure of central
tendency, while standard deviation is the most
widely used measure of dispersion
Example
 Temperature is the common example for interval

scale data
 The ratio of the two temperatures, 10°C and 20°C

means nothing because zero is an arbitrary point.


Compiled by Dr F Mupambireyi 14
Data Measurement Scales
Ratio scale:
 Ratio scales have an absolute or true zero of

measurement.
 Contains all the properties of the scales of

measurement
 Generally, all statistical techniques are usable

with ratio scales


 Multiplication and division can be used with

this scale but not with other scales mentioned


above.
Compiled by Dr F Mupambireyi 15
Data Measurement Scales
 Examples of ratio scale data
 Age, income, time, distance,

Compiled by Dr F Mupambireyi 16
Figure 1: Variables and levels of measurement

Categorical variable
Artificial Nominal or ordinal level of measurement
Categorical
numerical cannot change into
(Calculation of an average does not make
value continuous
sense)

Continuous variable Continuous can


True change into
numerical Scale level of measurement
categorical
value
(Calculation of an average makes sense)

17
Compiled by Dr F Mupambireyi
Illustration
 Quantitative data can be converted into categorical
data and never the other way round
 Scenario 1: If you are to ask the question on a
respondent’s age like: Age-----------------years
(responses are in continuous form)
 Scenario 2: The same questions can be asked as:
 Age 1) less than 30 2) 30 to <50 3) 50+years
(responses are in categorical form)
Compiled by Dr F Mupambireyi 18
THE END

ASANTE SANA
THANK YOU

Compiled by Dr F Mupambireyi 19

You might also like