RESEARCH METHODOLOGY

DIGITAL ASSIGNMENT III

BIVARIATE ANALYSIS

INTRODUCTION:
There are three main statistical analysis procedures when it comes to the level of analysis. The following
are —

 Univariate Analysis
 Bivariate Analysis
 Multivariate Analysis

The quantity of variables, types of data, and focus of the statistical inquiry all influence the data analysis
technique chosen.

Bivariate analysis allows us to investigate the relationship between two variables. This is quite useful in
real life. It aids in determining whether there is a relationship between the variables and, if so, what the
strength of the association is.

The bivariate analysis aids in the testing of the causality and association hypotheses. It aids in predicting
the value of a dependent variable based on changes in an independent variable.

BIVARIATE ANALYSIS:
Bivariate analysis is a method of determining the associations between two variables. They've been found
in studies on quality of life. One of the most basic types of quantitative (statistical) analysis is this. It
entails the examination of two variables (commonly referred to as X and Y) in order to determine their
empirical relationship. Bivariate analysis is particularly useful for evaluating simple association
hypotheses. It's a great way to figure out how much easier it is to know and anticipate a value for one
variable (perhaps a dependent variable) when the value of the other variable (maybe the independent
variable) is known. Bivariate analysis and univariate analysis, in which only one variable is examined,
can be contrasted. Both univariate and bivariate analysis can be used to make descriptive or inferential
conclusions. It is possible to describe it as a study of the relationship between the two variables. Bivariate
analysis is a simple (two-variable) and special instance of multivariate analysis (where several
relationships between variables are examined simultaneously).

The analysis of bivariate data is known as bivariate analysis. It is one of the most basic types of statistical
analysis, and it is used to determine whether two sets of values have a relationship. The variables X and Y
are usually involved.
 A single (‘uni') variable is examined in a univariate analysis.
 Bivariate analysis entails the examination of only two variables.
 A multivariate analysis is one in which more than two variables are examined.

A two-column data table can be used to hold the bivariate analysis findings. For example, we might be
curious about the connection between caloric consumption and weight (of course, the two are related very
strongly). Our independent variable, X, will be calorie consumption, and our dependent variable, Y, will
be weight.

Caloric Intake, Weight, Y

X
3500 250lbs
2000 225lbs
1500 110lbs
2250 145lbs
4500 380lbs

Two sample data analysis and bivariate analysis are not the same thing. In a two-sample data analysis
(similar to a two-sample Excel test), X and Y are not directly related, and each sample will have a distinct
number of data values. For each X in bivariate analysis, there is a Y value. Consider the following
scenario: you consume 3000 calories per day and weigh 300 pounds. You must write the following with
the x-variable first, then the y-variable: (3000, 300).

Here are a couple of examples of data analysis.

100, 45, 88, 99, 100, 45, 88, 99, 100, 45, 88, 99, 100, 45,

Sample number two is 44, 33, and 101.

Bivariate Analysis: (X, Y) = (100, 56), (23, 84), (398, 63), (56, 42).

Here are several examples: –

1) Each individual in a sample is measured for height (X) and weight (Y).

2) For each company in the sample, stock market valuation (X) and quarterly corporate earnings (Y) are
recorded.

3) At a collection of weather stations, temperature (X) and precipitation (Y) are measured on a given day.

A bivariate simple random sample (SRS) can be written (X 1, Y1), (X2, Y2)… (Xn, Yn).
Each observation is made up of two values; for example, the third observation is (X3, Y3). The
observations in a bivariate SRS are independent of one another, however the two measurements inside an
observation may not be (taller people are heavier, profitable corporations have higher stock market
valuation etc.).

HOW DO CONDUCT A BIVARIATE ANALYSIS?

Here is how the bivariate analysis is carried out.

SCATTER PLOT: Scatter plot shows the relationship between two variables. For numerous variables, a
scatter plot matrix displays all pairwise scatter plots. This illustrates the patterns that can be created by
combining the two variables.

The relationship is positive if the variables tend to rise and fall together. The relationship is negative if
one variable tends to increase while the other decreases. The association is zero if there is no pattern.

The relationship between the variables is said to be linear when a straight line describes the relationship.
The relationship is monotonic when it is described by a consistently increasing or decreasing nonlinear
function. Nonlinear or non-monotonic relationships are possible.

The statistical measures and association tests that are applicable are determined by the type of
relationship.

A bivariate normal density ellipse summarizes the correlation between variables if the relationship is
linear. The greater the connection between the variables, the narrower the ellipse. The variables are more
uncorrelated the wider and rounder it is. If the relationship is nonlinear, it is typically worth attempting to
modify the data to make the relationship linear because linear relationships include more statistics and are
easier to analyze than nonlinear relationships.

An outlier is an observation that stands out from the rest of the group and requires further inquiry. An
individual observation on each of the variables may appear to be entirely plausible on its own, but when
plotted on a scatter plot, it appears to be an outlier. Outliers can have a big impact on the product-moment
correlation coefficient, while other correlation values are more resilient.
REGRESSION ANALYSIS: Regression analysis is used for estimating the relationships between two
different variables. When the focus is on the link between a dependent variable and one or more
independent variables, it covers strategies for modelling and analyzing multiple variables. It aids in
comprehending how the dependent variable's value varies when any of the independent variables is
altered. For advanced data modelling objectives such as prediction and forecasting, regression analysis is
applied. This makes use of a variety of methods to figure out how the data post is related. The post may
follow an exponential curve. Regression analysis not only provides you an equation for that curve or line
but also gives you the correlation coefficient. Depending on the nature of the variable and the type of
analysis requested by the research, a variety of regression approaches are used. For example,

 Linear regression
 Simple regression
 Polynomial regression
 General linear model
 Discrete choice
 Binomial regression
 Binary regression
 Logistic regression

CORRELATION COEFFICIENT: A correlation coefficient measures the association between two

variables. The correlation between numerous pairs of variables is measured using a correlation matrix.

The type of relationship between the variables determines the best measure of association.

 When the relationship between the variables is linear, the product-moment correlation coefficient
represents the strength of the linear relationship.

The correlation coefficient can be anything between -1 and +1. A perfect positive linear relationship is
indicated by +1, whereas a perfect negative linear relationship is indicated by -1. The value zero denotes
that the variables are uncorrelated and that there is no linear relationship between them. In most cases, the
correlation coefficient is somewhere in the middle of these two numbers.

 When the relationship between the variables isn't linear, the strength of the relationship is described
by a rank correlation coefficient.

The range of rank correlation coefficients is -1 to +1. A positive rank correlation coefficient describes the
extent to which as one variable increases the other variable also tends to increase, without requiring that
increase to be linear. If one variable increases, as the other tends to decrease, the rank correlation
coefficient is negative.

It's preferable to start with a scatter plot to figure out what kind of relationship exists between the
variables, then apply an appropriate measure of association for the relationship. Don't be swayed by the
desire to find the highest correlation coefficient.

A correlation matrix measures the correlation between many variables. It is equivalent to a covariance
matrix of the standardized variables.
TYPES OF BIVARIATE CORRELATIONS:
The type of bivariate analysis is determined by the qualities and variables employed in the data analysis.
The variables can be numeric, ordinal, or categorical. The independent variable, like a pen brand, is
categorical. Probit regression or logit regression are employed in this scenario.

The ordered probit or ordered logit is used if the dependent attribute is ordinal. It's possible that the
dependent property is internal or a ratio, such as the temperature scale. This is the point at which
regression is calculated. This is how we refer to the various types of bivariate data correlation.

 Numerical and Numerical:

Both the dependent and independent variables of bivariate data, which include the dependent and
independent variables, have a numerical value in this type of variable.

 Categorical and Categorical:

The data is analyzed and statements and predictions are made about it when both variables in bivariate
data are in the static form. During the research, the analysis will help to determine the cause and impact to
conclude that the given variable is categorical.

 Categorical and numerical

This is when one of the variables is numerical and the other is categorical.

Bivariate analysis is a type of statistical analysis in which two variables are compared. One of the
variables will be a dependent variable, while the other will be an independent variable. The variables are
represented by X and Y. To determine the magnitude of the change, the differences between the two
variables are evaluated.

BIVARIATE DATA ANALYSIS EXAMPLE:

The investigation of any concurrent relation between two variables or attributes is known as bivariate
analysis. The study will look into the relationship that exists between the two variables as well as the
extent to which it exists. It aids in determining whether there are any inconsistencies between the
variables and, if so, what the causes are.

The bivariate analysis examples are used is to study the relationship between two variables. Let us
understand the example of studying the relationship between systolic blood pressure and age. Here you
take a sample of people in a particular age group. Say you take the sample of 10 workers.

The worker's age will be recorded in the first column, and their systolic blood pressure will be recorded in
the second column.

The table must subsequently be displayed in a graphical fashion in order to draw any conclusions. A
scatter plot is commonly used to represent bivariate data. The plots are made on grid paper with the y-axis
against the x-axis, which aids in determining the relationship between the data sets.
A scatter plot aids in the formation of a relationship between the variables and attempts to explain the
link. When you plot the age on the y-axis and the systolic blood pressure on the x-axis, you'll probably
see that they have a linear relationship.
How to understand the relationship?

The graph will demonstrate that there is a substantial and positive association between age and blood
pressure. Because the graph has a positive correlation, this is the case. As a result, as one gets older, the
systolic blood pressure rises. The line of best fit also aids in determining the correlation's strength. If there
is little space between the points then the correlation is strong.

The correlation coefficient, R, is a numerical value ranging from -1 to 1. This is a measure of how strong
a linear relationship between two variables is. To describe a linear regression the coefficient is called
Pearson’s correlation coefficient. When the correlation coefficient is close to one, it indicates that there is
a significant positive relationship. When the correlation coefficient is close to -1, then this shows a strong
negative correlation. When the correlation coefficient is equal to 0 then this shows no relationship at all.

CONCLUSION:

The above example lets you understand what bivariate analysis is. Inferential statistics and calculations
sometimes include analyzing two variables. The relationship between two continuous variables is the
focus of many scientific and business inquiries. The main question that the bivariate analysis answers are
if there is a correlation between the two variables, if the relationship is negative or positive and what is
the degree or strength of the correlation.