Week 2 descriptives Flashcards
Nominal data
Categorical data eg gender
Ordinal data
Orders people, objects or events along some continuum. No information is given about the differences between the points in the scale
Interval scales
Equal intervals between objects represent equal differences.
No absolute/true zero point
Cannot make ratio comparisons
Eg time, temp and IQ
What does no true zero point mean
It’s not nothing.
E.g 0 Celsius is not as lack of temperature, it is a degree and means something
Ratio scales
True zero point
Higher level of measurement - more detailed
What is the goal of Descriptive statistics
Characterize a numerical dataset efficiently
Condense meaningful information
Minimize the inevitable error involved during condensation
What is the goal of inferential statistics
infer the characteristics of the whole population are from a sample
Going beyond the information given to make likely assertions rather than certain ones
Sample statistics (English letters) are used to estimate population parameters ( Greek letters)
Key ideas: theoretical sampling distributions composed of innumerable random samples
key indices - p-value and confidence intervals
measures of central tendency
mean, median and mode
mean
average - add all and divide by how many present
cons of using the mean
inaccurate description and extreme scores influence mean
what does a histogram tell us
if the data is symmetrical and if the mean is appropriate to describe the sample
median
the score in the middle when are scores are arranged from smallest to largest
pro of using median
not affected by extreme scores
more accurate representation of data
mode
the most frequent score
what central tendency can nominal data only use
mode
what does measure of variability describe
the degree to which values vary
what are the measures of variability
range, interquartile range, variance and SD
range
difference between max and min scores
pros of using range
straightforward to calculate and easy to interpret
cons of using range
unstable across different samples
easily distorted by outliers
what does interquartile range use
percentiles
what are percentiles
cut off point that divides the data into percentage chunks
variance
a measure of how much the scores vary in terms of distance from the mean
the average of each scores squared deviation from the mean score
when do you use population formula for variance
when you have a whole population and don’t want to generalize scores
when do you use sample variance formula
when you want to estimate variance for population and generalize scores
standard deviation
the square root of the average of each scores squared deviation of the mean score = the square root of variance
bigger value - more spread out
