STATISTICS MATTERS IN DATA SCIENCE

Introduction

Statistics is a branch of applied mathematics that helps one to collect, organize, analyze, interpret and present data.
Its primary goal is to extract meaningful insights from numerical facts and figures to understand trends, summarize information and make information and make informed decisions in the face of uncertainty.

Statistical and non-statistical analysis

This can be done in two ways;

1). Statistical analysis
This is used to collect, explore and present large amounts of data to identify patterns and trends. It is also called quantitative analysis.
2). Non statistical analysis
It provides generic information and includes text, sound, still images and moving images. It is also called qualitative analysis.

Major categories of statistics

1). Descriptive statistics
It helps organize data and focuses on the main characteristics of the data. It also provides a summary of the data numerically or graphically.
. Examples of descriptive statistics;
. Mean (average)
. Median
. Mode
. Range
. Standard deviation
An example; Finding the average score of students in a class.

2). Inferential statistics
It is used to make predictions and conclusions about a population based on a sample. It also helps in decision and forecasting.
. Examples of inferential statistics;
. Probability testing
. Regression analysis
. Hypothesis making
. Confidence intervals
An example in a polling station a sample of 2000 voters to predict the outcome of a national election.

Types of Data

What is data?
This is a collection of raw fact values measurements collected for analysis. The main types of data are;

1). Qualitative Data (Categorical Data)
This is the type of data that describes qualities or characteristics and cannot be measured numerically.

 Types of Qualitative Data

. Nominal Data
This is data grouped into categories with no specific order.
. Examples of nominal data;
. Gender
. Religion
. Eye color
. Ordinal Data
This is data grouped into categories that have an order or ranking.
. Examples of ordinal data;
. Customer satisfaction (poor, good, excellent)
. Class positions
. Education level

2). Quantitative Data (Numerical Data)
This is the type of data that consists of numbers and can be measured or counted.

Types of Quantitative Data

    . Discrete Data

These are countable numbers, usually whole numbers.
. Examples of discrete data;
. Number of students
. Number of cars
. Goal scored
. Continuous Data
This is data that has measurable values which take any value within a range.
. Examples of continuous data;
. Height
. Weight
. Temperature

Statistical Data analysis steps

1). Define the problem
2). Collect data
3). Organize and clean the data
4). Explore the data
5). Choose the appropriate statistical method
6). Analyze the data
7). Interpret Results
8). Draw conclusions and make decisions
9). Present the findings

Common mistakes

1). Using poor quality data
2). Confusing correlation with causation
3). Misinterpreting Probability
4). Data leaking
5). Poor data visualization

Conclusion
Without statistics, data scientists would find it hard to analyze data science models which will lead to lack of accuracy, reliability, and scientific validity.