Descriptive statistics Flashcards
Cognitive biases
Ways of thinking that predispose one to favour a certain viewpoint over others
Confirmation bias
Tendency to search for or interpret information in a way that confirms one’s preconceptions while ignoring information that doesn’t support them
Selection bias
Distortion of evidence or data that arises from the way that the data is collected or the way that samples are selected to study
Supervisorship bias
Ignores the unsuccessful outcomes of a selection process
Publication bias
Scientific journal editors and publishers more likely to publish studies with positive results over those with negative results
Descriptive statistics
Communicate results without generalising beyond the sample to a larger group
What are the 4 levels of measurements
- Nominal
- Ordinal
- Interval
- Ratio
Nominal scale
Variables which have no numerical value and are in categories
Ordinal scale
Rank order categories from highest to lowest
Interval scale
When dealing with an interval scale, the difference between any 2 values can be calculated by using subtraction
Ratio scale
Quantitative data with true 0, Can add, subtract, multiply and divide
Why is distribution important (3)
- Determines which measure of central tendency to use
- Determines which measure if variability to use
- Determines further statistical analysis
Normal distribution
Even bell shaped curve
Non-normal distribution
Either…
- Negative skew (R)
- Positive skew (L)
Objective assessment of normality
Most variables follow a distribution that is not entirely normal