Research Methods (3): Hypothesis Testing and Statistical Significance

Kurtosis

Describes how pointy or flat the distribution is

Sample

A group of people, objects, items or events taken from a larger population for measurement

Sampling error

The difference between the sample statistics and the unknown population parameter

Probability

The likelihood of an event occurring

Exploratory data analysis

Analyzing data to understand its distribution and variation

Histogram

A visual representation of the distribution of data, divided into intervals called 'bins'

Outliers

Data points that differ significantly from other observations

Normal distribution

A bell-shaped curve with properties such as mean = mode = median

- symmetric

- unimodal curve

- never touches the X axis

- divides data in half using mean

Bimodal/multimodal distributions

Distributions with two or more peaks

Skewed distributions

Distributions that are not symmetric

Sampling

The process of selecting a sample from a larger population for measurement

Unbiased sampling

Sampling where every member of the population has an equal chance of selection

Inferential statistics

Calculations based on sampling distribution to make inferences about the population

Central limit theorem

The sampling distribution of sample means will be normal or nearly normal

Standard error

The standard deviation of the sampling distribution, indicating variability in sample means

SE= SD/|N

Confidence intervals

Ranges within which the population mean is likely to fall

Parametric tests

Tests that assume normal distribution and require interval or ratio scale data

Non-parametric tests

Tests suitable for nominal or ordinal scale data

Statistical significance

The probability that findings are not due to chance

statistical tests

allow researchers to work put the probability that their results could've occured by chance

NHST

null hypothesis significance testing

goal of NHST

to collect enough evidence and reject the H0 (null hypothesis) if it appears unlikely to be true

Null hypothesis

The default position in hypothesis testing

e.g physical exercise doesn't increase mood

Alternative hypothesis

The researcher's prediction in hypothesis testing

e.g physical exercise increases mood

P-value

The probability of obtaining results as extreme as the observed results, given that the null hypothesis is true

Critical region

The range of values that would lead to rejection of the null hypothesis

problem created by NHST

- all or nothing thinking

p

One-tailed hypothesis

A hypothesis that predicts a specific direction of difference or relationship

Two-tailed hypothesis

A hypothesis that predicts a difference or relationship without specifying the direction

Types of inferential tests

Correlation, chi-square, t-test, ANOVA, regression

correlation

tests if theres a relationship between 2 continious variables

chi square

tests if there is a relationship between nominal variables

t test

Tests if there is a difference between means of 2 groups or conditions (experimental designs - within or between subjects)