Introduction to Statistical Significance and Hypothesis Testing

This flashcard set covers the fundamental concepts of descriptive and inferential statistics, hypothesis testing, error types, and the differences between parametric and nonparametric tests as discussed in the lecture.

Descriptive statistics

Statistics used to describe and summarize a sample's data, including measures such as the average, standard deviation, range, and variance.

Inferential statistics

Statistics that use descriptive data to make conclusions about how a population would perform if a study were replicated.

Statistical significance

A term in research meaning "probably true," indicating that a difference is real and not due to chance, measurement error, or sampling error.

Clinical significance

Also referred to as practical significance or effect size, it describes the magnitude of a difference and whether that difference is large enough to justify implementing an intervention.

Systematic variance

Patterns of change in a study that can be attributed to the independent variable (IV) or known extraneous variables.

Unsystematic variance

Random fluctuations in data that cannot be pinpointed to a specific cause, often resulting from sampling error, measurement error, or individual differences.

Sampling error

Unsystematic variance caused by problems with how the sample was selected from the population.

Measurement error

Unsystematic variance caused by factors such as imperfect test instruments, assessment methods, scoring mistakes, or observation errors.

Hypothesis

A statement developed before a study begins that describes the predicted relationship between the independent variable (IV) and the dependent variable (DV).

Null hypothesis ($H_0$)

A statistical statement representing zero change or no difference, used as a baseline to compare data against during statistical testing.

Alternative hypothesis ($H_a$)

A statement describing the relationship or change the researcher expects to occur between the independent and dependent variables.

One-tailed (directional) hypothesis

A type of alternative hypothesis where the researcher predicts a specific direction of change, such as an improvement or a decrease in behaviors.

Two-tailed (non-directional) hypothesis

A type of alternative hypothesis where the researcher predicts there will be a difference between groups but does not specify the direction of that difference.

Alpha level ($p$-value)

The acceptable level of risk or confidence level set by a researcher, such as $.05$ ($95\%$ confidence) or $.01$ ($99\%$ confidence).

Type I error

A "false positive" that occurs when the null hypothesis is rejected even though it is true, erroneously claiming an intervention is effective.

Type II error

A "false negative" that occurs when the null hypothesis is accepted even though it is false, erroneously claiming an intervention is not effective.

$$n$$

The symbol used in research articles to represent the sample size of the study.

Critical value

A cutoff number or range used to determine whether a test statistic is significant enough to reject the null hypothesis.

Parametric tests

Powerful inferential statistics, such as t-tests and ANOVAs, used when data are normally distributed and measured on ratio or interval scales.

Nonparametric tests

Inferential statistics, such as Chi square ($X$), Wilcoxon, or Mann-Whitney U, used when data are not normally distributed, have many outliers, or are ordinal in nature.