Inferential Statistics

INFERENTIAL
STATISTICS FOR
ENVIRONMENTAL
DATA
GROUP 4
INTRODUCTION
Inferential statistics plays a crucial role in analyzing and interpreting
environmental data by drawing meaningful conclusions from a sample to
make broader inferences about a larger population. It helps environmental
scientists and researchers extrapolate findings from collected data to
understand and predict trends, correlations, and impacts on ecosystems,
biodiversity, climate patterns, and more. Through various methods like
hypothesis testing, regression analysis, and confidence intervals, inferential
statistics allows us to derive deeper insights, make informed decisions, and
address critical environmental challenges.
WHAT IS
Inferential Statistics uses
INFERESTIAL statistical techniques to
STATISTICS extrapolate information from a
smaller sample to make
predictions and draw
conclusions about a larger
population.
It uses probability theory and

statistical models
Primary uses of Inferential Statistic
• Providing population estimation
• Testing theories to make conclusions about populations
TYPES OF
INFERENTIAL
STATISTICS
HYPOTHESIS TESTING
• A formal process of statistical analysis, making use of inferential statistics
• The goal is tocompare and contrast popu;ations or analyse the relationships that
exist between variables making use of samples
• You can test hypotheses by conducting a careful and critical use of statistical
tests. Creating a null hypothesis and alternative hypothesis, then performing a
statistical test of significance are required.
• A hypothesis test can have left-, right-, or two-tailed distributions
Some Hypothesis tests employed in inferential statistics:
Z Test
• When data has a normal distribution and a sample size of at least 30, the z test is
applied to the data. When the population variance is known, it determines if the
sample and population means are equal.
• Applying a z-test involves several steps:
• Formulate Hypothesis
• Collect Data
• Calculate the Z-score
• Determine Critical Value or P-value
• Make a decision
• Interpret results
T Test
• When the sample size is less than 30,and the data has a student distribution, a t-
test is utilized. The sample and population mean are compared when the
population variance is unknown
• Applying a t-test involves several steps:
• Collect Data
• Choose the right T-test
• Calculate the T-score
• Determine the Critical Value or P-value
• Make a decision
F Test
• When comparing the variances of two sample populations, an f-test is used to see
if there is a difference
• Applying an f-test involves several steps:
• Collect Data
• Choose the right f-test
• Calculate the F-statistics
• Determine the Critical Value or P-value
• Make a decision
CONFIDENCE INTERVAL
• Confidence intervals make use of a statistic's variability to come up with an
interval estimate for unique parameters.
• Often come in handy when estimating parameters, as they consider sampling
errors in their estimates.
• A confidence interval can also be used to determine the crucial value in
hypothesis testing.
• For instance, a 95% confidence interval means that 95 out of 100 tests with fresh
samples performed under identical conditions will result in the estimate falling
within the specified range.
• Confidence intervals involve several steps:
• Collect data
• Compute sample statistical
• Determine the confidence level
• Identify the critical value
• Calculate the Confidence Interval
• Interpret the Interval
REGRESSION ANALYSIS
• a powerful statistical method that allows you to examine the relationship between two or
more variables of interest
• is done to calculate how one variable will change in relation to another
• helps determine the strength and direction of the relationship between variables and to
make predictions about the value of the dependent variable based on the values of the
independent variables
• commonly used in various fields such as economics, finance, marketing and social
sciences to understanding the underlying factors that influence a particular outcome.
• most common type of regression analysis is linear regression
REGRESSION ANALYSIS
Linear Regression
• a statistical technique used to model the linear relation between dependent variable and
one or more independent variables.
• finding the best-fit line that represents the relationship between the dependent and
independent variable.
• regression line
• widely used in various fields, such as economics, finance, engineering, and social
sciences to make predictions, forecast future values and test hypothesis
• determines the strength and significance of the relationship between the variables.
Applying a Linear regression inloves several steps:
• Understand the problem

• Gather the data
• Prepare the data
• Split the data
• Fit the model
• Evaluate the model
• Interpret the results
• Communicate the results
Analysis of Variance (ANOVA)
• Analysis of variance, or ANOVA, is a statistical method that separates observed variance
data into different components to use for additional tests.
• helps researchers understand whether there are statistically significant differences among
the group means and if those differences are due to random chance or actual effects.
• decomposes the total variance in the data into two components: variance between groups
and variance within groups.
• widely employed in various fields, including psychology, biology, economics, and
engineering, to name a few.
One-Way ANOVA
A two-way ANOVA is an extension of the
Versus Two-Way one-way ANOVA. With a one-way, you
ANOVA have one independent variable affecting a

dependent variable. With a two-way
A one-way ANOVA evaluates the impact of a ANOVA, there are two independents. For
sole factor on a sole response variable. It example, a two-way ANOVA allows a
determines whether all the samples are the company to compare worker productivity
same. The one-way ANOVA is used to based on two independent variables, such as
determine whether there are any statistically salary and skill set.
significant differences between the means of
three or more independent (unrelated)
groups.
NON-PARAMETRIC METHODS
• also known as “distribution-free”
• statistical techniques that do not make assumptions about the distribution of the data.
• particularly useful when the underlying distribution is unknown or uncertain, or when the
data do not follow a specific distribution. They are also useful when the data are discrete,
skewed, or heavily tailed, as they do not require the data to follow a specific shape
• one of the most commonly used nonparametric methods includes Rank-based methods
Some commonly used rank-based nonparametric methods include:
1. Wilcoxon rank-sum test (also known as the Mann-Whitney U test): This test is used to
compare two independent samples and does not assume a normal distribution. It is based on
the ranks of the data rather than their actual values.
2. Kruskal-Wallis test: This test is used to compare more than two independent samples and
does not assume a normal distribution. It is based on the ranks of the data rather than their
actual values.
3. Friedman test: This test is used to compare related samples (such as multiple treatments on
the same subjects) and does not assume a normal distribution. It is based on the ranks of the
data rather than their actual values.
THANK YOU
FOR YOUR
ATTENTION

Inferential Statistics

Uploaded by

Inferential Statistics

Uploaded by

INFERENTIAL

It uses probability theory and

• Understand the problem

ANOVA have one independent variable affecting a

You might also like