Ch 2 pt 4 - Variability Flashcards
Spread or Dispersion
Measures of variability or dispersion are quantitative measures of how spread out scores are in a distribution.
More spread=greater variability
less spread=lower variability
3 important points
- Variability provides an indication of how accurately the mean describes the distribution of scores
The mean more accurately describes the distribution when the scores are tightly clustered, as most of the scores will fall in and around the mean.
- Variability provides an indication of how well any individual score represents the entire distribution. This is because with less spread, any score is going to be more representative of the typical scores in the distribution
- When comparing different distributions, the variability of scores and the extent to which the distributions do or do not overlap determines whether the distributions are reliably different.
Common measures of variability: Range
Involves finding the difference between the upper limit of a data set and the lower limit of a data set.
Advantage: easy, simple way to describe the variability in a data set
Disadvantages: completely determined by 2 single values in the data set. It is subject to fluctuations and doesn’t give as much info as other measures of variability. (It neglects the rest of the data)
Interquartile & Semi interquartile
Variance & SD
Uses the mean as a reference point and measure how far, on average, each individual score is from the mean.
Variance
The average of each score’s squared difference from the mean
Variance is the square of SD
Variance is always calculated before SD, because deviations scores will always sum to 0.
Variance values are inflated because of squaring procedures, so we must find the SD to bring the value back in line with the original data.
Standard Deviation
The average difference of each score from the mean
The most common way of describing the spread of a group of scores
The square root of a variance
To find SD, we must compute variance first
