Hypothesis testing

Definition of Null Hypothesis: A null hypothesis states that there is no effect or no relationship between two measured phenomena or for a causal relation between them.
Alternative Hypothesis: It proposes that there is an effect or a relationship.
Example of Hypothesis:
- Research Hypothesis: Students who study more will have higher test scores.
- Null Hypothesis (H0): There is no relationship between the number of hours studying and test scores.
- Alternative Hypothesis (H1): There is a relationship between the number of hours studying and test scores.

p-Value Definition: A p-value is a measure that helps to determine the significance of results from a statistical hypothesis test.
- A smaller p-value indicates stronger evidence against the null hypothesis.
Example: If a study shows a p-value of 0.12:
- It means that there is a 12% chance of observing the data if the null hypothesis is true.
- As a standard, if p < 0.05, it is considered statistically significant.
Interpretation of Results:
- If p-value > 0.05, the findings are not significant and do not support the research hypothesis. State findings as inconclusive.
- Keeping scientific language cautious is essential: results do not prove hypotheses true, but may support them based on probability.

Formulate Hypothesis: Start with a clear and specific hypothesis guided by theory and previous observations.
Collect Data: Gather data, though you may not know the quality until later analysis.
Specify Null and Alternative Hypotheses: Define both hypotheses clearly in order to test.
Statistical Modeling: Select the appropriate statistical model based on your data type.
Calculate Test Statistic: Utilize respective statistical tests for data interpretation.

Types of Statistical Tests:
- Chi-Square Test: Used when dealing with categorical variables to look at differences between categories.
- t-Test: Applied for comparing the outcomes of two groups for continuous data.
- ANOVA (Analysis of Variance): Employed when comparing three or more groups regarding continuous outcomes.
Importance of Inferential Statistics: Since studies often examine samples rather than entire populations, it is crucial to utilize inferential statistics to accurately make generalizations.

Chapter Five Discussion: Detailed focus on the comparison of two means, specifically using the z-test.
z-Test Use: Ideal when population parameters (mean and standard deviation) are known.
- Practically, the z-test is rare as full population data is seldom available.
- Hypothetical scenarios may be used to rule out these calculations.

Normal Curve: Z-scores plotted on a normal distribution to locate p-values which are found in the tails of the curve.
Tail Testing:
- One-Tailed Test: Hypothesizes direction (greater than or less than).
- Two-Tailed Test: States there will be a difference without specifying direction.

Statistical Cutoff Values: Critical z-scores, such as those at a 0.05 significance level, which are used to assess whether the test statistic falls into the critical region for rejecting the null hypothesis.
- For one-tailed: z > 1.64 or z < -1.64.
- For two-tailed: z > 1.96 or z < -1.96.

Use examples to practice writing testable research hypotheses:
- Select a pair of variables (e.g., social media usage and self-esteem).
- Operationally define the variables, e.g., hours per day for social media usage and using self-esteem rating scales for measuring self-esteem.
Experimental Hypothesis: Formulate based on manipulation of variables.
Correlational Hypothesis: Draft based on observation of existing relationships.

Operational Definition: Clearly define what and how you measure variables, like the amount of time on social media or self-esteem metrics.
Example Definitions:
- Social media: Hours spent per day.
- Self-esteem: Scored on a rating scale from 0-10.

Statistical Tests vs. Hypothesis Testing: Understand the need to distinguish between drawing conclusions from statistical tests and hypotheses being true or false.
Theoretical Distributions: Recognizing the importance of theoretical distributions in hypothesis testing that help visualize how sample means derive from the population.

Sampling Distribution Definition: The distribution of a statistic (e.g., mean) obtained through repeated sampling from a population.
Central Limit Theorem: Given a large enough sample size, the sampling distribution will tend to be normally distributed, which underpins many statistical tests.
Standard Error Calculation: Standard deviation divided by the square root of the sample size n.

Confidence Interval Definition: Provides a range in which the population parameter lies with a specified level of confidence (e.g., 95%).
Calculations:
- Sample Mean +/- (Z-Value) * (Standard Error).
Interpreting Confidence Intervals: If a 95% confidence interval is calculated, it means if you were to take many samples, approximately 95% of the time, the true population parameter would fall within that interval.

Familiarize yourself with APA styles for reporting statistical findings in academic writing. Apply precise language and proper format to present statistical data in papers.