Exam Review

Flashcards about Type I and II Errors, Paired-Samples T-test, Statistical Power, Issues with T-tests, Alternatives to T-tests, Testing Proportions, Correlations, ANOVA, and Regression

What are the four steps for answering Type I and II Error questions?

1) Identify the “true state of the world”. 2) See if the output shows a significant or non-significant result. 3) Does this align with what you identified in step 1. 4) If an error was made was this a type I or II.

If a group of researchers run an experiment and obtain a p-value of .045 and apply an alpha level of .05, and a significant result is found where one should not exist, what type of error has been committed?

Type 1 error (false positive)

What kind of studies is a Paired-Samples t test used for?

Within – subjects' studies

What kind of differences can a Paired-Samples t-test identify?

Statistical difference between two time points, two conditions, or between a matched pair

What are the three things that affect statistical power?

α – level, Effect size, Sample size

How do α, N, and effect size affect statistical power?

Power increases as α increases, Power increases as N increases, Power increases as effect size increases

What are some issues with t-tests?

Data are not at least on an interval scale, strongly skewed data, extreme unequal variance and sample sizes are also unequal

What are two rank-based tests used as alternatives to t-tests, and what kind of samples do they apply to?

U –test for independent samples and Wilcoxon signed-ranks test for paired samples

What analysis is used when a DV is about category membership (so is on a nominal scale)?

Chi-square (χ2)

What is the formula to work out estimate point of proportions?

divide the number in the group by total number and x 100

What is indicated if there was no relationship between variables in the population, and it would be highly unlikely to find a correlation of .36 (or stronger) in this sample?

Null hypothesis (H0) 0.4% chance is unlikely (less than alpha level)

What does ANOVA compare?

Variability between groups (difference between group means) to the variability within groups (variance/SD for each group mean)

What is needed to help reject H0 in ANOVA?

Large variance between sample means, small variance within groups, large sample size (for adequate power)

What is used to predict an outcome variable (DV) from a predictor variable (IV)?

Regression

What does each unit in X expect Y to do on average?

Increase/decrease on average by the slope.

What does a r = 0.2 mean in regression?

For every 1 SD increase in one variable, the other variable will increase on average by 0.2 SD