What are the assumptions of ANOVA?

The assumptions of Analysis of Variance (ANOVA) are as follows:

a) Normality: The data for each group or condition must be normally distributed. This can be checked
using normality tests or visual inspection of the data distribution.

b) Homogeneity of Variance: The variances of the populations must be equal across the groups or
conditions. This assumption can be tested using Levene's test or Bartlett's test.

c) Independence: The observations within each group must be independent of each other. This means
that the value of one observation does not depend on the value of another observation within the same
group.

d) Absence of Multicollinearity: In the case of multiple independent variables, there should be no strong
correlation between the predictor variables.

If these assumptions are violated, the results of the ANOVA may not be reliable, and alternative
statistical tests or data transformations may be necessary.

1.

Describe the logic of Chi-Square?

The logic of the Chi-Square test is based on the comparison of the observed frequencies (or values) in a
dataset with the expected frequencies (or values) under a particular hypothesis or model.

The Chi-Square test is used to assess the statistical significance of the difference between the observed
and expected frequencies. The test statistic, denoted as χ^2 (chi-square), is calculated by summing the
squared differences between the observed and expected frequencies, divided by the expected
frequencies.

The logic of the Chi-Square test is as follows:

a) Formulate a null hypothesis (H0) and an alternative hypothesis (H1) based on the research question or
the model being tested.
b) Calculate the expected frequencies based on the null hypothesis.
c) Calculate the Chi-Square test statistic using the formula: χ^2 = Σ(observed - expected)^2 / expected.
d) Determine the degrees of freedom (df) based on the number of rows and columns in the contingency
table.
e) Compare the calculated Chi-Square value with the critical value from the Chi-Square distribution
table, given the specified level of significance and the degrees of freedom.
f) If the calculated Chi-Square value is greater than the critical value, reject the null hypothesis, indicating
a statistically significant difference between the observed and expected frequencies.
 2.

Differentiate between Non Parametric and Parametric test?

The key differences between Parametric and Non-Parametric tests are:

Parametric tests:

3.
○ Assume that the data follows a particular probability distribution, such as the normal
distribution.
○ Rely on parameters of the distribution, such as the mean and standard deviation, to
make inferences.
○ Examples include t-test, ANOVA, Pearson's correlation, and linear regression.

Non-Parametric tests:

4.
○ Do not make any assumptions about the underlying probability distribution of the data.
○ Rely on the ranking or ordering of the data, rather than the actual values.
○ Examples include Mann-Whitney U test, Wilcoxon signed-rank test, Kruskal-Wallis test,
and Spearman's rank correlation.

The primary advantages of non-parametric tests are:

5.
○ They are more robust to violations of assumptions, such as non-normal distribution or
unequal variances.
○ They are suitable for ordinal or ranked data, where the exact values are not as important
as the relative ordering.
○ They are generally more appropriate for small sample sizes or when the distribution of
the data is unknown.

The advantage of parametric tests is that they are generally more powerful (i.e., have higher statistical
power) than non-parametric tests when the assumptions are met.

6.

What are the Advantages of Confidence Interval?

The main advantages of using confidence intervals are:

7.

Quantifying the uncertainty: Confidence intervals provide a range of plausible values for the parameter of
interest, rather than just a single point estimate. This allows for a better understanding of the precision
and reliability of the estimate.

Facilitating interpretation: Confidence intervals make it easier to interpret the practical significance of the
results, as they indicate the likely range of values for the parameter.

Informing decision-making: Confidence intervals can help in making informed decisions, as they provide
a more complete picture of the uncertainty associated with the estimate.

Comparing groups: Confidence intervals can be used to compare the estimates between different
groups or conditions, as overlapping confidence intervals suggest no statistically significant difference.

Hypothesis testing: Confidence intervals can be used to perform hypothesis testing, as the confidence
interval can be used to determine whether the null hypothesis value falls within the interval.

Flexibility: Confidence intervals can be calculated for a wide range of statistical estimates, including
means, proportions, regression coefficients, and more.

○
Overall, confidence intervals provide a more informative and nuanced way of presenting and interpreting
the results of a study, compared to relying solely on p-values or point estimates.

8.

What is the Difference between Confidence Interval and hypothesis testing?

The key differences between confidence intervals and hypothesis testing are:

Confidence Interval:

9.
○ Provides a range of plausible values for a parameter of interest, based on the observed
data and a specified level of confidence (e.g., 95% confidence interval).
○ Focuses on estimating the parameter and quantifying the uncertainty around the
estimate.
○ Allows for the interpretation of the practical significance of the results.
○ Can be used to compare estimates between groups or conditions.

Hypothesis Testing:

10.
○ Involves making a decision to either reject or fail to reject a null hypothesis based on the
observed data and a specified level of significance (e.g., α = 0.05).
○ Focuses on determining the statistical significance of the difference between the
observed data and the null hypothesis.
○ Provides a dichotomous decision (reject or fail to reject the null hypothesis) based on
the p-value.
○ Does not directly provide information about the magnitude or practical importance of the
effect.

11. The key distinction is that confidence intervals provide a range of plausible values for the
parameter, while hypothesis testing makes a binary decision about the statistical significance of
the observed data. Confidence intervals and hypothesis testing can be used in conjunction to
provide a more comprehensive understanding of the research findings.