Independent t-test Procedure in R

Step 0: Set the working directory. This sets the folder location for the data file in the computer system.
Step 1: Activate the library. This activates the required analysis functions in the software (R).
Step 2: Set the significance level (alpha). This sets the threshold (often 0.05) for determining if results are statistically significant.
Step 3: Check data requirements
- a. Load the data. The data should be saved in the folder that's set as working directory.
- b. Independent variable. The variable affecting the dependent data can be numerical or text-based format.
- c. Check the data requirements (Assumptions).
  - i. Normality. Verify data distribution for each group using descriptive statistics (mean, standard deviation, skew, kurtosis). Skew and kurtosis values within +/-2 generally indicate a normal distribution.
  - ii. Variance equality. Test for equal variances using Levene's test. A p-value greater than 0.05 suggests variances are equal, allowing use of the more common equal variances independent t-test implementation.
  - iii. Independence assumes answers of individuals are not correlated.

Step 4: Run the Independent t-test. Use the data, significance level and whether variances are equal or not, as required.
Step 5: Run the Analysis (output). Examine the results (t-value, degrees of freedom, p-value, mean values, confidence intervals). Decide on statistical significance (reject or fail to reject null hypothesis) based on the p-value compared to the alpha value.

Step 1: Set the working directory. This sets the folder location for the data file in the computer system.
Step 2: Load the required libraries.
Step 3: Import the data file into the software (R). The file should be saved in the same folder as the R script.
Step 4: Test the data assumptions.
- Normality: Verify the distribution of data for each group using descriptive statistics (mean, standard deviation, trimmed mean, minimum, maximum, standard error). Skewness and kurtosis values within ±2 typically indicate a normal distribution. Use descriptive analysis functions within the software to verify normality.
Step 5: Run ANOVA.
Step 6: Conduct Post Hoc Analysis. If ANOVA is significant, conduct post hoc tests (e.g., Tukey HSD) to identify specific group differences.
Interpreting Results: Examine the ANOVA results for significant effects and follow up with appropriate post hoc analyses if necessary.

Step 1: Set working directory
Step 2: Load the required libraries
Step 3: Import data to software.
Step 4: Test assumptions (normality, level of measurement, independence)
Step 5: Perform correlation analysis.
- Determine which correlation test to use based on the level of measurement of each variable.
- Pearson for interval/ratio data, point biserial for one nominal and one interval/ratio variable.

Step 1: Set the working directory.
Step 2: Load necessary libraries.
Step 3: Import the data.
Step 4: Perform the Mann-Whitney U test. This compares two groups to see if they're from the same or different distributions.
Step 5: Interpret the results (p-value). A low p-value typically means a significant difference.

Step 1: Set the working directory.
Step 2: Load and attach necessary libraries.
Step 3: Import the data into the software (R)
Step 4: Fit the linear regression model.
Step 5: Test Model Assumptions and Interpretation.
- Independence. Assumes that the response of one subject or participant isn't affected by their other responses (i.e., subjects in the study do not interact with one another and each is independent of others' answers).
- Linearity. Assess whether a linear relationship fits the data. Examined via a scatterplot matrix of the predictor variables to the response variable. Visual check for a linear trend.
- Homoscedasticity of errors. Assesses whether the variability of errors is consistent across the entire range of predicted values. Examine via a scatterplot of residuals versus fitted values.
- Normality of error terms. Check the residuals (errors) for normality using a normal probability plot (e.g., using qqPlot()) or a histogram.
- Multicollinearity. Assess if the predictor variables are highly correlated with each other. Use Variance Inflation Factor (VIF). Values below 10 generally indicate a lack of multicollinearity.
- Interpret the results, including coefficients, p-values and R-squared.