QUIZ 2 Flashcards
discrete variables
(3)
dichotomous = binary (female/male)
categorical = nominal ( bus, bike, walk)
ordinal ( very satisfied –> neutral –> dissatisfied)
interval = numerical data
(i.e. differences are meaningful)
3
count
ratio ( height)
continuous (time)
measurement
precision vs accuracy
significant figures
describing distributions
univariate or bivariate
univariate
5 characteristics
normal
uniform
bimodal
U-shaped
skewed
descriptive stats
3
measures of central tendency
measures of variation
shape ( higher moments)
measures of central tendency
3
mean, median, mode
measures of variation
5
range, variance and standard deviation, quantiles, percentiles, inter-quartile
shape (higher moments)
3
variance, skewness, kurtosis
confidence
shown by 3
standard error
confidence interval ( credible interval)
statistical significance
scientific/ economic/ clinical significance
likert scale example
strongly disagree–> disagree–> neutral –> agree –> strongly agree
( this data is categorical, ordinal, dichotomous, continuous)
accuracy
true. consistent with the truth or objective (unbiased)
precise
detailed, specific, having low uncertainty, highly resolved
measurement e.g. the flood water is 35 cm deep, plus or minus half a cm
when we measure something in science, we always provide an estimate of our confidence along with the measurement
estimation e.g. ospreys live on average 38.4 +/- 5 years
when we statistically estimate something that was not measured directly, or infer something about the world through our research methods, we must also calculate a measure of confidence for what we report
communicating precision = probabilistic statements
between 31% and 37% with 95 % confidence ( i.e., e [31%,37%] 19 times out of 20)
displaying univariate qualitative (categorical) data
3
tables, bar charts, pie charts
displaying univariate continuous data
3
histogram, box plot, kernel density estimation (carpet plot)
bivariate plots: two categorical variable
1
paired bar plots
how to visualize –> bivariate plots : one categorical, one continuous variable
2
multiple histograms, box plots
visualize –> bivariate plots: two continuous variables
1
scatter plots
univariate measures of dispersion
5
range
standard deviation: average distance from the mean, coefficient of variation, index of dispersion, interquartile range
measures of shape
4
variance, skewness, kurtossis, L-moments
Skewness Left (negative), Right (positive)
(a) negative direction ( mean - median - mode)
(b) positive direction ( mode -median-mean)