Guh

Type I error

False positive; rejecting true null hypothesis.

Type II error

False negative; accepting false null hypothesis.

Alpha level

Probability threshold for Type I error.

Beta level

Probability threshold for Type II error.

Power of a test

Probability of correctly rejecting a false null hypothesis.

Effect size

Magnitude of result independent of sample size.

Sample size (n)

Number of observations in a study.

Statistical significance

Determines if results are likely due to chance.

Critical value

Threshold for rejecting the null hypothesis.

One-tailed test

Tests for effect in one direction only.

Two-tailed test

Tests for effect in both directions.

Variance of scores

Measure of score dispersion in a population.

Standard error (SE)

Expected difference by chance if null is true.

Homogeneous sample

Sample with similar characteristics to reduce variance.

Increasing alpha

Raises power but increases Type I error risk.

Increasing sample size

Enhances power and reduces standard error.

Population level effect

Real-world impact of an intervention or treatment.

Balancing risk

Managing Type I and Type II error probabilities.

Statistical probability

Likelihood of an event occurring.

Magnitude of results

Size or importance of an observed effect.

Calculating power

Determining likelihood of finding an effect.

Power

Probability of correctly rejecting the null hypothesis.

Effect Size (d)

Standardized measure of difference between groups.

Cohen's d

Effect size convention: 0.2 (small), 0.5 (medium), 0.8 (large).

Statistical Effect (delta)

Difference in standard errors between populations.

Power Tables

Reference for delta and corresponding power values.

Independent Measures

Different participants in each group of study.

Familywise Error Rate

Probability of one or more Type I errors in comparisons.

Omnibus Technique

Tests overall differences among multiple groups.

ANOVA

Analysis of variance for comparing means across groups.

Independent Groups ANOVA

One-way ANOVA with different subjects in each group.

Null Hypothesis (H0)

Assumes no difference between group means.

Variance Reduction

Decreasing score variability increases statistical power.

Sample Size Formula

N=2n for total participants in two-sample tests.

Power Calculation

Estimate needed power before conducting experiments.

Effect Size Formula

Different for single sample and between groups.

Interpret Results

Analyze findings and determine statistical significance.

Increasing Power

Achieved by increasing alpha, N, or effect size.

Comparative Error Rate

Increases with the number of statistical comparisons.

μk

Represents levels of treatment in ANOVA.

Total Variance

Sum of between-groups and within-groups variance.

Between-Groups Variance

Variance due to treatment effects among groups.

Within-Groups Variance

Variance due to individual differences within groups.

MStreat

Mean square treatment; estimates between-groups variability.

MSerror

Mean square error; estimates within-groups variability.

Null Hypothesis

Assumes no treatment effect; F should be around 1.

F-Ratio

Ratio of MStreat to MSerror; indicates treatment effect.

F Distribution

Family of distributions based on degrees of freedom.

Critical F-Value

Threshold to reject null hypothesis in ANOVA.

Omnibus Test

Initial test indicating presence of differences among means.

Degrees of Freedom

Number of independent values in a calculation.

Positive Skewness

F-distribution is always positively skewed.

Sum of Squares

Total variance calculated for all participants.

Significant Result

Indicates at least one group mean differs significantly.

Follow-Up Analysis

Examines specific differences after significant ANOVA result.

Linear Contrast

Method to compare specific group means post-ANOVA.

Experimental Error

Variability due to confounding factors in data collection.

Treatment Effect

Difference in means attributable to specific treatments.

Reject Null Hypothesis

Decision made if calculated F exceeds critical F.

t-test

Statistical test comparing means of two groups.

CBT

Cognitive Behavioral Therapy for anxiety management.

ACT

Acceptance and Commitment Therapy for anxiety management.

Control Group

Group not receiving treatment for comparison.

Weights Assignment

Process of assigning values to groups in contrasts.

Sum of Weights

Total weight assigned must equal zero.

Whole Numbers

Only integers used for weights in contrasts.

MS Error

Mean Square Error from ANOVA table.

n

Number of participants in each group.

Critical t'

Threshold t-value adjusted for multiple comparisons.

Bonferroni Correction

Adjusts critical value for multiple hypothesis testing.

Degrees of Freedom (df)

Calculated as N - k in ANOVA.

MCBT

Mean score for Cognitive Behavioral Therapy group.

MACT

Mean score for Acceptance and Commitment Therapy group.

p-value

Probability measure for statistical significance.

Dunn's Test

Post-hoc test for multiple comparisons after ANOVA.

Comparisons

Number of statistical tests conducted.

Squared Weights

Sum of weights squared for calculations.

Orthogonal Contrasts

Independent questions assessing non-redundant data.

Maximum Orthogonal Contrasts

Calculated as k-1, where k is treatments.

Non-Orthogonal Contrasts

Contrasts with overlapping information, reducing power.

Post Hoc Comparisons

Analysis after data examination to control error rates.

Tukey Test

Adjusts alpha for all pairwise comparisons.

Scheffe Test

Adjusts alpha for all pairwise and non-pairwise comparisons.

Bonferroni Adjustment

Corrects alpha for multiple comparisons to control error.

Adjusted Alpha

Calculated as .05 divided by number of comparisons.

t Obtained

Calculated t-value from hypothesis testing.

Reject Null

Decision when t obtained exceeds critical t'.

Hypothesis Testing

Procedure to determine if a hypothesis is supported.

Power of Test

Probability of correctly rejecting a false null hypothesis.

Omega squared

Effect size measure for variance explained by treatment.

Experimental effect

Importance of IV in explaining DV variance.

Significance of F

Probability that observed treatment differences are due to chance.

Cohen's rule of thumb

Guidelines for interpreting effect sizes.

Small effect

Cohen's d of 0.01 or less.

Medium effect

Cohen's d of 0.06.

Large effect

Cohen's d of 0.15 or more.

t statistic

Difference between means divided by chance difference.

F statistic

Variability between treatments divided by within treatment variability.

Repeated measures ANOVA

Participants serve in each treatment condition.

Homogeneity of variances

Equal spread of data across treatment groups.