1. Statistics Overview

 Let’s Learn Stats!  -by Dr. Prafulla Dikshit

6-8 minutes read

1.     Statistics Overview

Before even getting to know statistics, it is critical to get a basic understanding of research. Research is a set of concepts, principles, and methods which facilitates science and scientific inquiry. The Scientific field itself may be classified into hard (e.g. Physics, Chemistry, Biology, or Mathematics) and soft (Psychology, Sociology, Anthropology, Management, or political science), science subjects. While research in itself is a big topic, it is primarily of two types – namely 1) Qualitative and 2) Quantitative. Research and its methods may also vary according to the domain or subjects.

1.1  Importance of statistics and why we even need statistics

The purpose of statistics is to enable and facilitate research findings and decisions. It is interesting to note that statistics is all around us, and if we pay attention, we may often encounter statistics in our everyday lives. Usually, we come across scientific claims in news about one or the other condition, event, or process outcome. They report results supported with statistical information or facts. For example, you may have seen claims in news lookin
g something similar to the following (some of the claims may be true or false, and some are just made up for the purpose of illustration):

o   “Men are significantly more likely to need intensive care treatment for COVID -19."

o   "Based on research, 80 percent of women felt wrinkles were reduced within four weeks."

o   "The XYZ product makes you lose 20 pounds in a week."

o   "80% of dentists recommend our products."

o   “Lipsticks Contain Excessive Lead, Tests Reveal”

o   “Latest statistics show heart failures on the rise.”

o   "Exercise labels on drinks are better for keeping teenagers away from junk food!”

o   "Insects might be more sensitive to radiation than thought."

o   "38.2% of Americans accidentally share fake news."

o   "XYZ formula prevented children who took it from developing allergies."

o   “XYZ-cola’s Vitamin water could promote healthy joints."

I'm sure at least some of them sound familiar to you. Notice that the above claims come from diverse fields like business, life sciences, sports, fitness, etc., and not all of them provide statistical evidence, but most seem to make some scientific-looking claims in the headlines. Usually, such headlines are presented with some supporting statistical findings. However, whether the findings or claims made based on them are genuine must be ascertained. Many of the numbers expressed in these claims may not be accurate and may lead you to make decisions that you may find faulty or regret later. A good understanding of Statistics empowers us to evaluate such claims objectively. This helps us distinguish good from faulty reasoning and prevents vulnerability to manipulation and decisions which may not be in our interest. Therefore, equipping oneself with statistical know-how is a crucial step one can take, toward gaining better control of one’s life.

That was the importance of statistics from the point of view of a common person. However, from a research perspective knowledge of statistics at a basic level may be indispensable. Whoever is pursuing research in one or the other form, should have an elementary understanding of statistics and statistical concepts. Statistics is seen as critical in many research scenarios, especially, where measurable inputs and outcomes are involved.

Before taking a deep dive into statistics, let’s get to know some elementary terms or keywords of statistics, as presented below.

1.2    Research concepts important to know before learning Statistics

When we think of statistics what comes to our mind first is numbers. Taking a deeper look, we find that these numbers represent a well-known term – data.  Data itself may be either qualitative or quantitative, just like two types of research. This is because data is the critical link between Statistics and Research (see Figure 1).  For example, a research study that intends to ascertain the link between smoking and cancer has to collect data about smoking patterns and the incidence of cancer. Without the relevant data, research can't go anywhere. Further, without the relevant data, no statistics or statistical concept can produce any meaningful output to serve the purpose of research. In the smoking example, we can't possibly apply any statistics in absence of the smoking or cancer data, or even if we have some data, which is not relevant, we can’t produce any meaningful results to answer the questions posed by the said research. It follows that the type of research (qualitative or quantitative) guides the type of data required (qualitative or quantitative), which in turn shows the way for the application of statistics. However, not all research may require the application of statistics. Generally, quantitative research requires quantitative or numeric data, which may call for the application of statistics, while qualitative research mostly does not require statistical analyses to arrive at results.

Figure 1

An Overview of the Linkages between Research, Data, and Statistics.

 

Research by itself may also be described as either exploratory, descriptive, or explanatory. Exploratory research intends to explore ideas and factors toward formulating a problem. Descriptive research describes any individual, group, or phenomenon. Explanatory research describes and explains an event or establishes a relationship between entities leading to the event, for a specific or a well-defined problem area. Each of these three types of research may involve the use of different types and levels of statistics. For example, under exploratory research, the use of graphical statistical tools like box plots or histograms to explore the data and summarize their main characteristics may be applied to arrive at a research problem worth solving. Usually, descriptive statistics may be employed to meet the exploratory research objectives. Descriptive statistics involve measures of variability and the central tendency of the data. At the most basic level, variability and central tendency represent how different versus how similar one element of the data is likely to be from the others within the same set of data. Since exploratory research can most often be qualitative in nature, in such cases it may not involve the use of statistics.

Descriptive research may involve both descriptive statistics and what may be termed inferential statistics. While we'll discuss descriptive versus inferential statistics in detail in the forthcoming posts, for now, it suffices to know that descriptive statistics describe the characteristics of data whereas inferential statistics let us make deterministic conclusions about the relationships embedded in the data. Explanatory research also employs both descriptive and inferential statistics, however in this case the study statistics are used to inform a very specific and well-defined problem or a set of problems and assumptions.

So this was an overview of Elementary Statistics for you😊. In the next post, we will get to know a few basic keywords and definitions critical to understanding Statistics before we actually take a deep dive into the subject. Till then take care and have a great time!