JARGON An introduction to Statistics NMH Upward Bound
JARGON: An introduction to Statistics NMH Upward Bound Summer Academy 2014
Statistics: a definition àTry to define it? à“A collection of methods for planning experiments, obtaining data, and then organizing, summarizing, presenting, analyzing, interpreting, and drawing conclusions based on the data. ” à“A bunch of tools for figuring out, talking about, and determining the significance of, numbers of the real world. ”
OK, what are data? à You first! à “Observations that have been collected. ” à Some examples of data? à Test scores à Answers to a written or spoken survey à Amounts of natural occurrences like rainfall à Other ideas?
Whose data? Three terms to note: àPopulation – a complete collection of all elements to be studied. It includes all the subjects to be studied. à USA Example àCensus – the data from the whole population àSample – a subcollection of members selection from a population. à Samples have to be chosen randomly – and tested to make sure it was really random!
Parameter v. Statistic A parameter is a numerical measurement describing some characteristic of a population A statistic is a numerical measurement describing some characteristic of a sample - 52% of US votes cast for Barack Obama
Quantitative v. Qualitative Data àDo you already know? àQuantitative data consist of numbers representing counts or measurements àQualitative data are observations that do not have a quantitative nature à Plant example
Discrete v. Continuous Data àDiscrete data are measurements in finite and countable numbers à Chicken eggs, number of students in a class àContinuous data are measurements with infinitely possible values without interruptions or jumps à The weight of a professional athlete
Correlation and Causation Correlation: As event A happens, event B happens too. Causation: A causes B, or B is a direct result of A. Example from the pretest: Owners of luxury cars live longer than owners of standard cars. It might be true that luxury car owners do live longer, but the luxury car is not responsible for the longer life. There is another cause, which could or could not be related to the luxury car.
Levels of Measurement There are 5 levels of measurement àNominal àOrdinal àInterval àRatio àDichotomous
Nominal level of measurement àData that consist of names, labels, or categories only. The data cannot be arranged in an ordering scheme (such as low to high or large to small) àYes/No/Undecided, colors of cars driven by college students àOther examples?
Ordinal level of measurement àData that can be arranged in some order, but differences between data values cannot be determined numerically or are meaningless à Full size, mid-size, compact cars. The parking space?
Interval level of measurement à Can be arranged in some order, and the difference between values is meaningful. However, data do not have a natural zero starting point (where none of the quantity is present) àTemperatures: 98. 2°F and 98. 6°F are 0. 4° apart and 98. 6° is hotter than 98. 2°F, but because 0° is arbitrarily chosen, you can’t say that 50° is twice as hot as 25°.
Ratio level of measurement à The interval level with the additional property that there is also a natural zero starting point (where zero indicates that none of the quantity is present) àWeights, prices
Practice for your homework à#1 -23 odd and no worksheet! 1) Statistic or parameter? 31% of Americans live in multi-family housing (apartments). 2) Statistic or parameter? In a survey of 402 students at Holyoke High School, 63% said they were not fans of Justin Bieber. 3) Did this come from a discrete or continuous data set? When 19, 218 gas masks from a branch of the US Military were tested, it was found that 10, 322 of them were defective. 4) What level of measurement? Ratings of “fantastic, ” “good, ” “average, ” “poor” or “unacceptable” for blind dates 5) What level of measurement? Numbers on the jerseys of female basketball players in the WNBA
- Slides: 14