Lesson 7: Descriptive statistics Flashcards
What is descriptive data?
Descriptive statistics summarises and describes features of a dataset, like as the range, variation, data distribution and central tendency
What are the different types of quantitative measures and give examples
- nominal: eye colour, named
- ordinal: level of satisfaction, named and natural order
- interval: temperature, named, natural order, equal interval between variables
- ratio: height, named, natural order, equal interval between variables, has a ‘true zero’ value thus ratio between values can be calculated
why is it important to know what type of data you have?
- can know how to handle different types of data correctly
- what you can calculate with different types
- which describe stats and visualisations are appropriate
- which statistical hypothesis test you can use and cannot use
How can qualitative data be altered into quantitative data
ways of measuring qualitative answers can be ‘transformed’ into the number like a written response can be transformed into an ‘intensity score’
Are longitudinal or analytical surveys more analytical than descriptive and why
yes as they analyse events at more than one point in time or test a hypothesis so the possibility of suggesting the direction of cause and effect
What do analytical surveys do?
collect data that can be statistically compared to analysed for associations/ patterning
What are the different types of agreement scales you could use?
- frequency: never, rarely, sometimes
- importance: unimportant, insignificant, important
- liking: dont like at all, like slightly etc
What can you calculate using nominal and ordinal measurement scales?
frequency
proportion
percentage
central point
How is nominal and ordinal data represented?
- pie charts
- bar charts
What can you calculate with interval or ratio data?
- range: minimum and maximum values
- spread: inter-quartile range, SD or confidence intervals
- central point: mean, medium mode
definition of SD
measure of how dispersed the data set is from the mean
What is one SD of the mean means
68% of the data is wihtin one SD of the mean
what does it means when Sd if 1.96
95% of values are within 1.96 Sd of the mean
what does it mean when SD is 2.58
99% of the values are within 2.58 Sd of the mean
What does it mean when SD is within 3
99.7% of values are within 3 SD of the mean
how could interval and ratio data be represented graphically
box plot
histogram
how are outliers represented in a box plot
on the sides of the box plot so not included in the box plot