SESSION ONE: NATURE AND SCOPE OF STATISTICS

MEANING OF STATISTICS

In its initial concept, statistics is the collection of data on the population and
social – economic activities vital to the nation or state or province. The nation
requires information on numbers of taxable adults to allow a projection of
reliable total income. In this century statistics has advanced far beyond that
narrow conception. In general, the word “statistics” has three possible
connotations depending on the context of the application, namely: As a
SUBJECT; As a piece of Information and as a mathematical function (or entity).

Statistics as a subject is defined as the scientific methodology concerned

with the planning, collection, summarization, presentation, analyzing and
interpretation of data to meet a lot of specified objections.
Statistics as a piece of information is thought of as quantitative data such
as graduates data, export data, births and deaths statistics in Botswana etc.
Statistics as a mathematical function (entity) is a function of observations
(sampled).

In a nutshell, statistics can be broadly defined as a group of methods used to

collect, analyze, present and interpret data and to make decisions.

Examples of Statistics are;

An average 6 year old laughs 300 times each day.

An employed woman in Botswana spends an average of 13.2 hours per
week on housework.
In 2007-2008 great footballer Ronaldo earnings

TYPES OF STATISTICS

Like almost all fields of study, statistics has two aspects; Theoretical and
Applied. Theoretical or mathematical statistics deal with the development,
derivation, and proof of statistical theorems, formulas, rules and laws.

Applied statistics involves the application of those theorems, formulas, rules and
laws to solve real – world problems. This (course) book is concerned with
applied statistics and not with theoretical statistics. By the time you finish
1
studying this book, you will learn how to think statistically and to make
educated and scientific guesses.

Broadly speaking applied statistics can be divided into three areas;

Descriptive Statistics: This is a reduction of mass of data into few

members which are quantitative expressions of the salient characteristics
of data. Examples are measures of location, central tendency and
presentation of data in tabular and diagrammatic forms.
Deductive Inferential Statistics: This is an act of drawing inference about a
sample (part of population) using our knowledge of the population. The
process involves arguing from the general (the entire units) population to
specific (sample)). For example, in Swaziland the only language is Swazi
and so one can infer that Dr S. Dlamini from Swazi can speak Swazi
language fluently.
Inductive Inferential Statistics: This is the process of drawing inference
about the population from the sample, it is arguing from the specific to the
general. For example; J.J. Okocha is a great footballer, so all male
Nigerians can be said to be footballers.

STATISTICAL VARIABLE

Statistical variable is a characteristics which is capable of assuming values or

forms within a given range.

A quantitative variable is one that can assume numerical values: examples are
height, age, weight, parity, number of students in STA 112 and STA113 and so
on. The quantitative variable can further be classified into two major groups as

Discrete: Those variables that can take integers or whole numbers as

values; for example, Age last birthday, number of students, bacteria
counts and so on.
Continuous: Those variables that can take intervals as values; for example,
blood pressure, Heights, Weights and so on.

2
A qualitative variable is one that cannot be characterized by numerical values.
For example, educational status, marital status, gender and so on. This variable
form can be further classified in to two groups: (i) Those that can be
characterized by only two possible outcomes. Examples are No or Yes; Dead or
Alive, Present or Absent and so on. (ii) Those qualitative variables whose values
have more than two possible outcomes. For example, marital status has values as
never married, married, divorced or widows.

DATA SETS

Data are sets of numeric and non-numeric information recorded on statistical

variable. Data are facts or figures from which conclusions may be drawn. Data
are the raw material of statistics. Statistical analysis cannot proceed until the data
of interest in an investigation are assembled and organized in a useful manner.
By definitions, data collected for a particular study is referred to as DATA SET.
Data is called RAW DATA when it has not received all statistical treatment.

EXAMPLES

Examples of data sets are found all around us;

the financial section of our daily paper contains price data for securities
and commodities;
an economic report showing inflation rates for different countries;
A newspaper article contains data on achievement test scores for all
schools in a city.

Consider the following table showing 2006 sales revenues and industrial
classification for the companies included in a study:

2006 SALES REVENUES AND INDUSTRIAL CLASSIFICATIONS

COMPANY SALES INDUSTRIAL

(Coded Name) (P000,000) CLASSIFICATION

AFL 175.3 Pharmaceuticals

INC 485.5 Chemicals

3
ACC 243.9 Chemicals

WPL 1081.2 Electrical Equip.

PEC 656.6 Food Products

RET 819.2 Electrical Equip.

HWP 851.2 Pulp and Paper

CII 434.0 Textiles

MPL 107.8 Meat Packing

FLC 209.1 Food Products.

CHARATERISTICS OF DATA SETS

The data set in the above table represents a collection of facts and figures, more
technically, the data set contains observations on the variables of interest for the
elements, in the data set.

A data set represents a collection of elements, and for each element information
on one or more characteristics of interest is included in the data set. In the above
table, the element is a company and the characteristics of interest are the
company’s 2006 SALES REVENUES AND INDUSTRIAL.

DEFINITION: A characteristic that can take on different possible outcomes is

called variable. If the outcomes are expressed numerically, the variable is said to
be quantitative. If the outcomes refer to non-numerical qualities or attributes, the
variable is said to be qualitative. For example; variable INDUSTRIAL
CLASSIFICATION is a qualitative variable because its outcomes are non-
numerical, also the variable SALES is a quantitative variable because its
outcomes are numerical.

4
DATA SOURCES

Statistics is not only concerned with organizing and analyzing data once they are
assembled, but also with the sources of data and how data are collected for
study. The initial stage of any investigation involves a specification or definition
of the problem to be studied. From this specification comes an identified need for
particular types of data to clarify the problem. At this point, the question of
where to obtain the necessary data is posed.

PRIMARY AND SECONDARY SOURCES

Some data may be partially or totally available from a primary source(from an

internal source) such as manual or computerized records of operating and
accounting data, which most organisations maintain routinely. Indeed, many
organisations now have extensive computer data banks where large quantities of
internal ordinary data are stored and from which data are retrieved as needed.
Other data may be obtainable from a secondary source (or an external source).

Such secondary source may be unpublished in some case published in a

reference book or statistical periodical or data may be in a computerized form
such as tapes or on-line access to an external data bank. The secondary source
organisation (external source organisation) may be a government agency, a trade
association, or a private specialized service company.

ILLUSTRATION

PRIMARY SOURCE: Consider in case of a market researcher who wishes to

study the geographic distribution of customers that use a particular company
product. The researcher might find much of the required data in the internal
accounting records of the business.

SECONDARY SOURCE: Consider a store location analyst for a retail outlet chain
who is concerned with crime patterns in different cities might find the necessary
data in external sources such as government reports on criminal and judicial
statistics or computerized police records.

REMARK: Primary data are data collected directly from the field of inquiry that
is data collected by means of surveys or census. Whereas secondary data are data
obtained from published or unpublished results or reports of organisations.

5
LIMITATIONS IN THE USE OF DATA SOURCES

When using any data source, the user should be thoroughly acquainted with the
nature and limitations of the data. Limitations may include imperfect or
improper methods of data collection, recording and classification as well as
errors of omission or commission when data are transferred from one record to
another.

The user needs to determine whether the definitions employed in compiling the
data are appropriate for the purpose. Also, important is to check whether
changes in concepts, definition and data – collection methods have occurred over
the time period of interest, and, if so, to determine the effect of these changes on
the data.

The user should have access to pertinent information and explanatory comments
about the manner in which the data were compiled, the data’s accuracy and how
the data should be interpreted.

6
7