Proportions

Observer bias may pseudodata interpretation toward expecting an association due to pre-held beliefs about games' impact on aggression.

Desensitization theory proves narrative forcing researchers to only look for confirming evidence of video game effects.

Data mining bias incorporates isolated statistical anomalies as though they were representative of significant trends in normal player behavior.

Violent content rating system guarantees objective assessment, removing any risk of subjectivity in measuring game aggression levels.

Probability of NOT making a Type II error

Probability of NOT making a Type I error

Probability of making a Type II error

Probability of making a Type I error

The probability of rejecting a true alternative hypothesis.

The probability of correctly rejecting a false null hypothesis.

The probability of failing to reject a false null hypothesis (Type II error).

The probability of correctly accepting a true null hypothesis.

The population mean is known prior to collecting sample data.

Sample sizes must exceed thirty regardless of underlying distributions.

The populations follow any distribution as long as it’s symmetric.

The population standard deviations are equal across groups.

A Type I error occurs.

A sampling error occurs.

A Type II error occurs.

An experimental design error occurs.

The test with lower powers, since that implies higher sample variability requiring wider confidence intervals.

The tests with lower powers, less prone to type II errors due to stricter criteria for rejecting the null hypothesis.

The tests with higher powers.

Both tests will have equal width confidence intervals since they share the same sample sizes and significance levels.

Significance level

Power of the test

Type I error

Type II error

The data occurred by random chance with a probability equal to the p-value.

There is strong evidence against the null hypothesis in favor of the alternative.

The probability that the null hypothesis is true is equal to the p-value.

The null hypothesis is definitively proven false.

Since everyone uses social media differently, it's impossible to have any valid correlations related to its usage affecting happiness levels at all.

Social media algorithms ensure unbiased sampling which guarantees causal relationships can be inferred from correlations observed.

Self-reported happiness levels are always accurate reflections of emotions making them reliable indicators for establishing causation with social media use.

The relationship might be influenced by confounding factors like personality traits or life events that aren’t accounted for in the analysis.

Preference for one outcome over another by researchers

Population size alone without regard to sample characteristics

Convenience of obtaining samples rather than randomness

Sample size, variability within data, and effect size