Introduction to Statistics

Statistics is the science of analyzing numerical data related to any subject or event. In the early days, it was used to collect data regarding state administration and military affairs. Today, we use it in almost every fields. Professor Gottfried Achenwall first used “Statistics” as a complete and separate subject. Sir Ronal Aylmer Fisher is known as the “Father of Statistics.” To sum up, statistics is the science of making decisions about an event by collecting, processing, presenting, and analyzing numerical data that is influences by various factors.

We can explain statistics in two different ways: as method and as data. As method, statistics refers to the techniques and methods used for collecting, presenting, analyzing, and interpreting nemurical data from a field of inquiry. For example, collecting exam marks, presenting marks in a table, analyzing the marks, and interpreting the exam results. Conversely, as data, statistics refers to the numerical expression of events or subjects. For example, 45% of students passed the exam, in which 45% is “statistics as data”.

Characteristics of Statistics

To be considered “Statistics”, data must possess some features. The features are explained below:

(a) Aggregate of Facts: Statistics must be a collection of facts, not an isolated number. For example, “The average height of 1st-year students is 5 feet” is statistics. Conversely, “Ahmed’s height is 5 feet” is not statistics.

(b) Numerically Expressed: Data must be expressed in numbers (e.g., weight, height, age). Qualitative attributes (e.g., intelligence, talent) are not statistics as these cannot be converted into numerical scores.

(c) Pre-Determined Purpose: Data must be collected with a specific, pre-set objective. Numbers collected without a goal do not qualify as statistics. For example, when a researcher counts random objects like ceiling tiles or red books without a specific question or objective, these figures remain isolated numbers lacking analytical value. Hence, it is not statistics.

(d) Affected by Multiple Causes: Data is usually influenced by by a variety of factors, not just a single case. For example, rice production is statistics because it is affected by soil fertifility, fertilizer, rainfall, floods, and pesticides simultaneously.

(e) Related to Each Other: Data must be related to a specific subject or context. For example, a random list like “50,60,70” is meaningless. It only becomes statistics we if specify that these are “marks obtained by students in a class.”

(f) Homogeneous: Data must be of the same type or category. For example, statistics does not allow mixing “height of students” with “height of trees” because these are not the same subject.

(g) Comparable: Data must be comparable, which means all data must be expressed in the same unit. For examples, if salaries of 3 employees are listed as 5000 taka, 400 dollars, and 200 pounds, this data is not statistics.

(h) Collected from Inquiry: Data must be gathered from a specific field of inquiry using scientific methods, not based on imagination or non-scientific estimation.

(i) Systematic Collection: Data must be collected according to a planned, systematic method to ensure accuracy and avoid wasting time/resources. Collecting data in haphazard way leads to error in analysis.

Stages of Statistical Investigation

Statistics is defined as the science of decision-making accomplished through five specific steps or activities. The five stages are explained below:

(1) Data Collection: Data is considered the “raw material” of statistical research. It involves collecting accurate data from a population via Census or Sample Surveys using scientific methods.

(2) Processing of Data: Raw data collected from the field is often random, disorganized, and error-prone. This stage involves editing, correcting errors, and organizing the data for storage and further use.

(3) Presentation of Data: In this stage, processed data is organized to be easily understood, concise, and attractive. The methods of data presentation are classification, tabulation, frequency distribution, and visualizations (graphs, charts, stem and leaf diagrams). The goal of this stage is to make data ready for analysis.

(4) Analysis of Data: In this stage, we apply statistical methods to the presented data. The statistical methods are central tendency, dispersion, skewness, kurtosis, correlation, regression, time series, index numbers, intrepolation, and extrapolation. The goal of this stage is to analyze the data to reveal patterns and simplify the information.

(5) Interpretation: This stage involves explaining the characteristics of the population based on the analytical results. The goal of this stage is to enable the researcher to make effective decisions.