# SCU4 Statistical Thinking for Researchers

### This is an online course offered by the SCU.

'Statistical Thinking for Researchers' is SCU's new course.

It is an introduction to concepts in data analysis presented as a series of short videos on Echo360 with Powerpoint slides.

Topics include graphical summaries of data, hypothesis testing and confidence intervals, and regression models.

The content is evolving as more videos become available - current content includes:

Introductory Data Analysis

• Exploratory Data Analysis
• EDA 1: One categorical variable, includes bar graphs and pie charts.
• EDA 2: One quantitative variable, includes histograms and density cures.
• EDA 3: Two quantitative variables, includes scatter plots.
• EDA 4: Three quantitative variables, includes scatter plot matrices.
• EDA 5: Graphics, includes principles of good graphic design.
• EDA 6: Conclusion, wraps up the day.
• Statistical Inference
• Inference 1: introduction, includes the objectives for the day.
• Inference 2: Probability, includes classical rules of probability like multiplying and Bayes Theorem.
• Inference 3A: Normal distribution, introduces the normal distribution.
• Inference 3B: Normal distribution, shows how to calculate probabilities for normal distributions.
• Inference 3C: Checking normality, includes a couple of ways to check if variable follows a normal distribution.
• Inference 4A: Sampling distributions, introduces the idea that a sample mean comes from a distribution too.
• Inference 4B: Confidence intervals, shows how to derive confidence intervals for a population mean.
• Inference 5: Two sample confidence intervals, shows how to derive confidence intervals for the difference in two population means.
• Inference 6: Hypothesis testing, introduces the concepts of hypothesis testing.
• Inference 7: Issues, talks about some of the issues associated with hypothesis testing.
• Inference 8: Binomial, introduces the Binomial distribution for binary data.
• Inference 9: Two-way tables, includes hypothesis testing for data stored in a two-way table.
• Linear Modelling
• Modelling 1: Correlation, introduces the concept of correlation between two quantitative variables.
• Modelling 2: Correlation issues, discusses some of the issues associated with correlation.
• Modelling 3: Regression, introduces the concepts around fitting a straight line through a scatter of points.
• Modelling 4: Regression output, looks at Genstat and R output for a regression.
• Modelling 5: Regression inference, brings together hypothesis testing and regression.
• Modelling 6: Diagnostics, shows how to check whether a regression line is a good fit to a scatter of points.
• ANOVA
• ANOVA 1: completely randomised design, intrdocues the simplest design of experiment.
• ANOVA 2: randomised complete block design, extends the simple design to includes groups of subjects.
• ANOVA 3: hypothesis tests, applies the concepts of hypothesis tests from Day II to the designed experiment.
• ANOVA 4: factorial designs, extends the simple design in a different way to include a second treaatment.
• ANOVA 5: split plot, extends the complete block design to apply two treatments to two different groupings within the design.