Scribd
Upload
14 views4 pages

Methodology

The document describes the methodology used to examine the relationship between IQ (independent variable) and CGPA (dependent variable) for final year college students. It involves collecting IQ and CGPA data from students and using regression analysis in Microsoft Excel. Linear, polynomial, and exponential regression models are discussed. The R-squared and adjusted R-squared metrics for assessing the predictive power of regression models are also introduced.

Uploaded by

Nurul Najla
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
14 views4 pages

Methodology

The document describes the methodology used to examine the relationship between IQ (independent variable) and CGPA (dependent variable) for final year college students. It involves collecting IQ and CGPA data from students and using regression analysis in Microsoft Excel. Linear, polynomial, and exponential regression models are discussed. The R-squared and adjusted R-squared metrics for assessing the predictive power of regression models are also introduced.

Uploaded by

Nurul Najla
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 4

3.

0 METHODOLOGY

The dataset that contains IQ and CGPA that was collected from the final year college students. Table 3.
1 shows IQ as an independent variable, x, and CGPA as a dependent variable , y. Regression analysis is used to
examine the relationship of IQ and CGPA among the final year college student. The intelligence quotient, or IQ,
is a measurement of your reasoning and problem-solving abilities (Cherry, 2022). It essentially indicates how
well you performed on a particular test when compared to other persons your age (Cherry, 2022). The CGPA is a
measure for evaluating your academic success (Bursaries et al., 2022). The CGPA is used to determine a
student's current standing overall (including all degree-related courses) and, if relevant, in a minor (Bursaries et
al., 2022). The CGPA is determined by dividing the total credit value of courses attempted by the sum of grade
points received (Bursaries et al., 2022). The data will be analyzed using Microsoft Excel.

TABLE 3.1 List of Data IQ and CGPA

Independent
variable , x 91 93 94 95 96 98 116 118 119 120
(IQ)

Dependent
variable , y 7.87 7.91 8.04 8.29 8.34 8.46 8.71 8.86 8.97 9.06
(CGPA)

3.1 LINEAR REGRESSION

Sir Francis Galton first proposed the notion of linear regression in 1894 (Kumar K et al., 2018).The
least squares method was introduced by Legendre in 1805 and Gauss in 1809 (Yan et al., 2009). That was the
very first form of linear regression (Yan et al., 2009).Linear regression is a statistical test used to establish and
quantify the relationship between the variables in a data set (Kumar K et al., 2018).

The simple linear regression model is often written as the following form :

Where y is the dependent variable, is y intercept, is the gradient or the slope of the regression
line,
is the independent variable, and is the random error (Yan et al., 2009) .

3.2 POLYNOMIAL REGRESSION

Diagnostics based on residuals can be used to assess the adequacy of polynomial models, which are specific
examples of multiple linear regression (Rawlings et al., 2005). Another option is to fit a higher-order polynomial
that is deemed adequate and then use statistical tests to obtain a low-order polynomial that is adequate (Rawlings
et al., 2005). It is possible to fit a polynomial of degree when one measurement is observed at each of the k
different values of the input variable (k-1) (Rawlings et al., 2005).

The polynomial’s formula is written as the following equation :


Here, , is the values of independent variables and, are constants (Shobhit et al., 2021)

3.3 EXPONENTIAL REGRESSION

Exponential regression is used to model situations in which growth begins slowly and then accelerates rapidly
without bound, or where decay begins rapidly and then slows down to get closer and closer to zero ("Exponential
Regression | College Algebra", 2022). We use the command “ExpReg” on a graphing utility to fit an exponential
function to a set of data points ("Exponential Regression | College Algebra", 2022). The exponential function
has form of

or

When performing regression analysis, we use the form which most commonly used on graphing utilities ,

("Exponential Regression | College Algebra", 2022). Note that b must be non-negative and for ,

it is an exponential growth model while for , it is an exponential decay model ("Exponential


Regression | College Algebra", 2022).

R-SQUARED

R-squared (also known as coefficient of determination) is a metric that quantifies how much variation a
regression model can explain (Muralidhar, 2022). The R-squared of a regression model is positive if the model's
prediction is better than the mean of the already available 'y' values, and negative otherwise. R-theoretical
squared's formula is shown below (Muralidhar, 2022).
The formula above is true in theory, but only when R-squared is positive (Muralidhar, 2022). The formula does
not provide a negative R-squared because we compute the sum of squares in both the numerator and
denominator, which always results in a positive R-squared (Muralidhar, 2022). From the aforementioned
formula, we can get the correct formula (the one that is utilized in practice and also returns negative R-squared),
as shown below (Muralidhar, 2022).

ADJUSTED R-SQUARED

If the new predictors are independent of the target variable and provide no value to the model's predicting power,
R-squared grows or stays the same when we add more predictors to a multiple regression model (Muralidhar,
2022). R-disadvantage is overcome with adjusted R-squared. It will only increase if the new predictor increases
the model's accuracy (Muralidhar, 2022). As more independent and irrelevant parameters are added to a
regression model, the adjusted R-squared declines (Muralidhar, 2022).

You might also like