Goss-Sampson (2020), Statistical Analysis in JASP - pp. 25-33

Question 1

• Population Parameters vs. Sample Statistics:

Answer 1

• A parameter is a measurable characteristic of a population (e.g., mean, standard deviation).
• A statistic is a measure derived from a sample used to estimate population parameters.

Question 2

Types of Bias

Answer 2

Bias refers to systematic errors that can distort results. Common types include:
- Participant Selection Bias
- Participant Exclusion Bias
- Analytical Bias

Question 3

• Participant Selection Bias:

Answer 3

Some participants are more likely to be selected.

Question 4

• Participant Exclusion Bias:

Answer 4

Systematic exclusion of certain individuals.

Question 5

• Analytical Bias:

Answer 5

Errors in the evaluation of results.

Question 6

Handling Outliers

Answer 6

• Approaches to address outliers:
• Correcting Errors: Check for data input or measurement errors.
• Keeping Outliers: Retain them if they represent valid observations.
• Deleting Data Points: Justifiable only if errors are confirmed.
• Replacing Values (Winsorizing): Replace outliers with nearest valid values.

Question 7

• Shapiro-Wilk Test:

Answer 7

Tests the null hypothesis that the data are normally distributed.

Question 8

Dealing with Non-Normal Data

Answer 8

1. Data Transformation: Apply logarithmic, square root, or other transformations to normalize
   data.
2. Non-Parametric Tests: Use these alternatives as they do not assume normality.

Question 9

Testing Homogeneity of Variance

Answer 9

• Levene’s Test: Examines the null hypothesis that variances are equal.
• Results interpretation:
• p>0.05: Equal variances assumed (homoscedasticity).
• p<0.05: Unequal variances (heteroscedasticity).

Question 10

Homoscedasticity

Answer 10

Residuals are uniformly distributed.

Question 11

Heteroscedasticity

Answer 11

Residuals show patterns or funnels