Research and Design 5 - Statistics Review

Descriptive Statistics

Summarizes data using things such as mean, median, and standard deviation

Infeerntial Statistics

Drawing conclusions about the population based on t-tests, ANOVA, correlation and regression

Mean

The average of something, you add all the values and divide by the number of values. This is very sensitive to outliers

Median

Middle score when all scores are in numerical order, it is not affected by outliers

Mode

It is the most frequent and useful for categorical data

Variance

Average squared distance from the mean

Standard Deviation

Square root of variance, will tell you how far apart the scores are

Z-score

How far apart scores are from the mean in standard deviation units

Frequency Histograms

Graphs that show the distribution of data by showing the shape and outliers

Null Hypothesis Significance Testing

State hypothesis, usually a null one or alternative, collect the data, calculate statistics, fine the p-value, compare to the alpha level which is usually going to be .5, and then either reject or retain the null

P-Value

Probability results occurred by chance

Alpha

Cut-off for significance

How to reduce type 1 error? (A false positive)

Lower the alpha level

How to reduce type 2 error? (A false negative)

Increase the power, usually by increasing sample size

Power

Probability of detecting a real effect if there is an effect, usually increased by sample size and effect size

Confidence Intervals

Range of likely value that are true for the population, being 95% confident that the true value is between certain numbers

Correlation

Measures relationship between two variables, no causation, two continuous variables

T-test

Compares two groups to each other

ANOVA

Compared three or more groups to each other and it is reported as F

Regression

Predicts one variable from another, reported as beta

Meta-analysis

Combines results from multiple studies

Effect size

The strength of a result, such as r, cohen’s d, or beta. But effect size is usually cohen’s d

Cronbach’s alpha

Used for scale correlations

Spearman correlation

Tests for strength of the association between two ordinal variables

Chi-Square

Tests for the strength of the association between two categorical variables

Independent t-test

Tests for the difference between two independent variables

Simple Regression

Tests how change in the predictor variable will predict the change in the outcome variable