week 4.1 IRM

True or false: Parametric data is the same as normal data.

True or false: Non-parametric data can be skewed or kurtotic.

True or false: Transformations of data are always suitable for psychological data.

True or false: The decision to run a non-parametric test is commonly made when there is a 'true' value, indicating that the participant understood task instructions but scored very low?

True or false: Removing a value and replacing it with the mean is a commonly used approach in psychological studies?

True or false: The Shapiro-Wilk test takes into account both skew and kurtosis of the data?

Which of the following is NOT a characteristic of normal data?

What is the main reason for caring about normality in statistical models?

When should a non-parametric test be used?

Which of the following is a common approach for handling outliers in big data sets?

What does the Shapiro-Wilk test measure?

What does a p-value of 1 for the Shapiro-Wilk test indicate?

Data Types and Characteristics

Parametric data is not the same as normal data; parametric assumptions are based on the characteristics of the data distribution.
Non-parametric data can indeed be skewed or kurtotic, indicating variability in distribution shape.
Transformations of data are not always suitable for psychological data; caution must be taken to ensure that the transformation does not distort interpretations.

Statistical Tests

The decision to run a non-parametric test often arises when there is a presence of a 'true' value, which signifies that participants understood the task but scored low.
Removing a value and replacing it with the mean is a common but potentially misleading approach in psychological studies, as it can mask the true variance.

Shapiro-Wilk Test

The Shapiro-Wilk test assesses the normality of a dataset by evaluating both skewness and kurtosis.
A p-value of 1 for the Shapiro-Wilk test indicates perfect normality in the data, suggesting no deviation from the normal distribution.

Understanding Normality

Characteristics of normal data include symmetry, a single peak (unimodal), and a bell-shaped curve; any deviation from these traits suggests non-normal data.
Normality in statistical models is crucial because many statistical tests assume normal distribution to draw valid conclusions.

Non-parametric Tests Usage

Non-parametric tests should be used when data does not adhere to normality assumptions or when dealing with ordinal data or small sample sizes.

Handling Outliers

A common approach for managing outliers in big data sets includes robust statistical methods or transformations to reduce their impact without altering the overall dataset significantly.

Test your knowledge on determining parametric and non-parametric data in research methods with this quiz. Learn about the characteristics of parametric data, how to identify it, and the differences between parametric and non-parametric tests. Professor Hannah Keage provides valuable insights in this quiz based on the BEHL 2005 and BEHL 2019 (UO) Introductory Research Methods lecture.