Experimental vs Nonexperimental Studies
In experiments, researchers give treatments and observe to see if they cause changes in behavior. A simple experiment is one in which we form two groups at random and give each group a different treatment. When the participants are divided at random, the experiment is called a true experiment. In non experimental studies, the researcher does not give treatments. An experiment in which treatments are given in order to observe their effects. However, sometimes it is impossible or impractical to conduct a true experiment. There are many examples of non-experimental studies.
- Causal-Comparative Study (ex-post facto study) - 1) We observe and describe some current condition and 2) we look to the past to try to identify the possible cause(s) of the condition.
- Survey - The purpose is to describe the attitudes, beliefs, and behavior of a population.
- Case Study - The emphasis is on obtaining thorough knowledge of an individual
- Field Research - A thorough study of a group of people.
- Longitudinal Research - Repeatedly measuring trait(s) of participants over a period of time in order to trace developmental trends.
- Correlational Research - When we are interested in the degree of relationship between two or more quantitative variables.
This shiny app allows you to conduct meta analysis as well as inter-rater reliability, publication bias and calculate effect sizes.
Free internet sources for data
Interpreting random effects in linear mixed-effect models
This blog helps explain a difficult statistical method.
Averages and Likert Scale Data
This book presents better ways to analyze Likert scale data than averages.
Frequnetism and Bayesianism: A practical Introduction
A 5 part series on two approaches to statistics.
MorePower 6.0 is a flexible freeware statistical calculator that computes sample size, effect size, and power statistics for factorial ANOVA designs.
Power ANalysis for GEneral Anova designs
A quick experiment displaying the impact of sample size on estimates we make from the data.
Test of Insufficient Variance - TIVA
A tool to detect questionable research practice
Interactive tools for meta-analysis, power analysis and experimental planning.
An interactive visualization for interpreting Cohen's d