NAMPALLY

Introduction to
Parametric Tests
Parametric tests are statistical methods used to analyze data that meet certain
assumptions. These assumptions include the normality of the data distribution,
homogeneity of variance, and independence of observations.
Assumptions of Parametric Tests

Normality Homogeneity of Variance

The data should follow a normal distribution. This can be The variances of the groups being compared should be
checked using histograms, Q-Q plots, and statistical tests. equal. This can be tested using Levene's test or Bartlett's
test.

Independence of Observations Interval or Ratio Scale

The observations should be independent of each other. This The data should be measured on an interval or ratio scale.
means that the value of one observation should not This means that the differences between values are
influence the value of another observation. meaningful and can be compared.
Applications of Parametric Tests

1 Quantitative Data 2 Continuous Variables

Parametric tests are designed for analyzing These tests are suitable for variables that can
numerical data that can be measured on a take on any value within a range, such as height,
continuous scale. weight, or temperature.

3 Interval or Ratio Scales 4 Normal Distribution

Parametric tests work well with data measured A key assumption of parametric tests is that
on scales where differences between values are the data follows a normal distribution.
meaningful and have a true zero point.
One-Sample t-Test

Define Hypothesis
1
Formulate null and alternative hypotheses.

Calculate t-statistic
2
Compute the t-statistic using sample mean, population mean, and standard error.

Determine p-value
3
Find the p-value corresponding to the calculated t-statistic.

Interpret Results
4 Compare the p-value to the significance level to reject or
accept the null hypothesis.

The one-sample t-test is used to compare the mean of a single sample to a known population mean. It helps determine if there is a
significant difference between the sample mean and the hypothesized population mean. This test is useful for examining if a treatment
or intervention has a noticeable effect on a group.
Two-Sample t-Test

Define Hypothesis
1 Formulate null and alternative hypotheses about the means of two groups.

Calculate t-statistic
2 Compute the t-statistic using sample means, standard deviations, and sample
sizes.

Determine p-value
3
Find the p-value corresponding to the calculated t-statistic.

Interpret Results
4 Compare the p-value to the significance level to reject or
accept the null hypothesis.

The two-sample t-test is used to compare the means of two independent groups. It helps determine if there is a significant difference
between the average values of two populations. This test is useful when comparing the effectiveness of two different treatments,
interventions, or approaches.
Paired t-Test

Define Hypothesis
Formulate null and alternative hypotheses about the means of two related groups.

Calculate t-statistic
Compute the t-statistic using the differences between paired observations, mean difference, and standard error.

Determine p-value
Find the p-value corresponding to the calculated t-statistic.

Interpret Results
Compare the p-value to the significance level to reject or accept the null hypothesis.
ANOVA (Analysis of Variance)

Compares Multiple Groups One-Way ANOVA Two-Way ANOVA

ANOVA is used to analyze One-way ANOVA compares Two-way ANOVA analyzes the
differences between means of two unmatched groups with data effect of two nominal predictor
or more groups simultaneously. categorized by one factor. variables on a continuous outcome
variable.
One-Way ANOVA

1 Define Hypothesis
Formulate the null hypothesis stating no difference in means across groups, and the alternative
hypothesis stating at least one group mean differs.

2 Calculate F-statistic
Compute the F-statistic using the between-group variability and within-group variability.

3 Determine p-value
Find the p-value corresponding to the calculated F-statistic based on degrees of freedom.

4 Interpret Results
Compare the p-value to the significance level to reject or accept the null hypothesis, concluding
whether there is a significant difference in means between groups.
Two-Way ANOVA

Define Hypothesis
Formulate null and alternative hypotheses regarding the interaction effects between two
independent variables on a dependent variable.

Calculate F-statistic
Compute the F-statistic based on the variance between groups and within groups.

Determine p-value
Find the p-value corresponding to the calculated F-statistic.

Interpret Results
Compare the p-value to the significance level to determine if there is a significant interaction effect.
Assumptions of ANOVA

1 Independent Samples 2 Normal Distribution

The samples are independent and selected The parent population from which the samples
randomly. are taken follows a normal distribution.

3 Additive Effects 4 Normal Error Distribution

Treatment and environmental effects are Experimental errors are normally distributed
additive in nature. with a mean of zero and variance σ2.
Pearson's Correlation

Define Hypothesis
1
Formulate null and alternative hypotheses about the linear relationship between two variables.

Calculate Correlation Coefficient

2
Compute Pearson's r using the covariance and standard deviations of the variables.

Determine p-value
3 Find the p-value corresponding to the calculated r, assessing the
significance of the correlation.

Interpret Results
4 Compare the p-value to the significance level to
reject or accept the null hypothesis.

Pearson's correlation coefficient, denoted by r, measures the strength and direction of the linear relationship
between two continuous variables. The value of r ranges from -1 to 1, where a positive value indicates a positive
linear association, a negative value indicates a negative linear association, and 0 indicates no linear relationship. It is
important to note that correlation does not imply causation, meaning a strong correlation between two variables
does not necessarily mean that one variable causes the other.
Pearson's 'R' Correlation

Correlation Coefficient Type of Correlation

r = +1 Perfect positive correlation

0 < r < +1 Partial positive correlation

r=0 No correlation

0 > r > -1 Partial negative correlation

r = -1 Perfect negative correlation

Regression Analysis

1 Define Model
Establish a linear or non-linear relationship between a dependent variable and independent variables.

2 Estimate Coefficients
Determine the slope and intercept of the regression line, quantifying the relationship.

3 Assess Model Fit

Evaluate the accuracy of the model using statistics like R-squared and p-values.

4 Make Predictions
Use the regression equation to forecast the dependent variable based on values of the independent
variables.
Z-Test

Assumptions
1
The population is normally distributed, and the sample is drawn at random.

Conditions
2
The population standard deviation (σ) is known, and the sample size is large (n >
30).

Null Hypothesis
3
The population mean (μ) is equal to a specified value (μ0).
Choosing the Right Parametric Test

Type of Group Parametric Test

Comparison of two paired groups Paired t-test

Comparison of two unpaired groups Unpaired two-sample t-test

Comparison of population and sample drawn from One-sample t-test

the same population

Comparison of three or more matched groups varied Two-way ANOVA

in two factors

Comparison of three or more matched groups varied One-way ANOVA

in one factor

Correlation between two variables Pearson Correlation

Conclusion and
Summary
Parametric tests are powerful statistical tools for analyzing data that
meets certain assumptions. They provide insights into differences
between groups, relationships between variables, and the effects of
interventions. Understanding the assumptions and choosing the
appropriate test is crucial for drawing valid conclusions.
THANK YOU
• Name : M.ROHIT
• H.no : 2129-23-672-170
MBA 1st year
A section