Lecture 3 Flashcards
descriptive statistics focus on…
summarising & presenting data
tables & data visualisation portray…
data
2 types of descriptive statistics?
measures of central tendency
measures of variability
measures of central tendency?
median, mean, mode
focuses on the average or middle values
measures of variability?
standard deviation, range, variance, quartiles
focuses on the dispersion of data
how to calculate standard deviation?
Step 1: Find the mean.
Step 2: Subtract the mean from each score.
Step 3: Square each deviation.
Step 4: Add the squared deviations.
Step 5: Divide the sum by the number of scores.
Step 6: Take the square root of the result from Step 5.
how to calculate variance?
Calculate the mean of the data.
Find each data point’s difference from the mean value.
Square each of these values.
Add up all of the squared values.
Divide this sum of squares by n – 1 (for a sample) or N (for the population).
variables = ?
factors that can take on more than one value
can vary in value
distribution = ?
refers to the different values that can be assumed and their frequency
how often each value occur
for discrete data, we care especially about….
commonly occurring values (mode)
unusual values (e.g., outliers)
discrete data = ?
data that is counted, not measured
(e.g., the number of dogs, population)
continuous data = ?
can take on any value in an interval
(e.g., 1<x<10, can be 1.3, 5.6, 7.2 etc)
univariate descriptive statistics = ?
describing one variable at a time
the benefits of good graphics?
presents ideas with clarity, efficiency and precision
exhaustive = ?
fully comprehensive
considers all elements of interest in a dataset
what do statistics involve?
the collection, description, analysis and inference of conclusions based on data
categorical data = ?
data that reflects qualitative characteristics
AKA Qualitative data
assigns numbers values to qualitative data (e.g., 1=female, 2=male)
numerical data = ?
data that are naturally numbers based
e.g., age, height, weight
what levels of measurement come under categorical data?
nominal & ordinal
what levels of measurement come under numerical data?
interval & ratio
nominal level data?
describes qualitative characteristics or groups
no inherent numerical order/rank
e.g., gender, ethnicity, colours
often used in surveys
ordinal level data?
the same as nominal data as it describes categories
but ordinal level data can be ranked in order
e.g., level of income, level of satisfaction, level of agreement
interval level data?
numerical data that involves quantitative data
has an order and equal spaces between points
e.g., range of test score 0-30 = C, 30-60 = B, 60-100 = A
arbitrary 0 point
ratio level data?
ordered/ranked quantitative data
equal distance between points
e.g., length of time, weight, height, length
non-arbitrary zero point
arbitrary zero point?
zero doesn’t mean it doesn’t exist
e.g., 0 degrees isn’t an absolute be-all-end-all, as degrees can be -1
height, however, cannot be below zero
frequency = ?
the number of times something occurs
what are the distributional features of interest on a frequency table?
unusual and commonly occurring features of interest
patterns
must variable names be meaningful?
yes
good practice with tables?
- avoid vertical lines
- minimal horizontal lines to avoid confusion
- separate table contents from title & notes
- separate headings from the rest of the data
pie charts are good at presenting data when…
- discrete variable
- exhaustive
- small number of categories
- real variation across categories
benefits of bar charts?
enables direct comparison
purpose of graphs & tables?
powerful way of communicating complex information clearly & accurately
