• The manipulation of one independent variable and measurement of one dependent variable, while other variables are kept constant. Controlled setup
• Thus studies like correlational studies, quasi-experiments, and natural experiments (that is, any research undertaken without control over the independent variable and without a controlled sampling procedure) are not acceptable for the simple experimental study.

• title page
• appendices
• graphs
• tables
• section headings
•  works cited page (references) 

Parametric Tests

• Used in Quantitative Research: Parametric tests are designed for analyzing numerical (quantitative) data that meet certain assumptions.

• Assumptions: The most common assumptions for parametric tests include the normal distribution of data and equal variances among groups. These tests are also typically used with interval or ratio data, where the measurements have a known and consistent relationship.

• Examples: T-tests, ANOVA, and linear regression are examples of parametric tests. They are used when you're confident your data meets the necessary assumptions, allowing for more powerful and precise statistical analysis.

Non-Parametric Tests

• Also Used in Quantitative Research: Non-parametric tests are used for quantitative data that do not meet the assumptions required for parametric tests. They can handle ordinal data (which have a ranked order but not a consistent interval between ranks) or nominal data (categorical data without any order), as well as interval or ratio data that does not follow a normal distribution.

• Fewer Assumptions: These tests do not assume a normal distribution of the data and are more flexible in terms of the types of data they can analyze.

• Examples: Chi-square tests, Mann-Whitney U tests, and Kruskal-Wallis tests are common non-parametric tests. They are useful for analyzing data that are skewed, ranked, or otherwise not suitable for parametric testing.

Application to Qualitative Research

While parametric and non-parametric tests are tools of quantitative research for analyzing numerical data, qualitative research focuses on non-numerical data such as words, images, or objects. Qualitative data analysis involves categorizing and thematically analyzing data, rather than using statistical tests. However, qualitative researchers may use quantitative methods, including non-parametric tests, in mixed-methods research to analyze survey results or other quantifiable elements within a primarily qualitative study.

Summary

• Parametric and non-parametric tests are both used in quantitative research depending on the nature of the data and whether it meets certain statistical assumptions.

• Qualitative research primarily involves non-numerical data and does not typically use parametric or non-parametric tests. However, mixed-methods approaches can incorporate quantitative analyses, including non-parametric tests, alongside qualitative analyses.

You reject a true null hypothesis. (Null hypothesis is true, but you reject it.)

e.g- I detect danger (when the danger is not there=Null hyp present)

Type II Error (False Negative):

You fail to reject (accept) a false null hypothesis. (Null hypothesis is false, but you fail to reject/ accept it.)

e.g. I fail to detect danger (when the danger is there+Null hyp absent)

Remember for null hypothsis,, e do not use the word accept fo null hypothesis but use as fail to reject

• 1% Level of Significance is ideal for research that cannot afford false positives, such as health-related studies or legal decisions. Requires larger sample sizes to mitigate the risk of missing real effects (Type II error).

• 5% Level of Significance is the default choice for balanced exploratory and confirmatory studies. Works well across small to moderate sample sizes and diverse fields.

• 10% Level of Significance is suitable for exploratory research or noisy data sets. It detects weak effects but comes at the cost of more false positives (Type I error). Ideal for small sample sizes when detecting weak relationships is more important than being strictly accurate.