Descriptive statistics

what is a measured variable
assigning numbers to objects, events or concepts according to a set of rules, allowing them to be categorised
1 of 28
list the levels of measurement
nominal, ordinal, interval, ratio
2 of 28
nominal scale
identity or category, must be mutually exclusive, good for quantifying qualitative data
3 of 28
ordinal scale
a measure of order e.g. small/medium/large
4 of 28
interval scale
a measure of order and quantity, difference between values can be calculated but can't establish X-fold increases e.g. shoe size/ year of birth
5 of 28
ratio scale
interval scale WITH AN ABSOLUTE ZERO, and therefore X-fold increases. e.g. time to run a race
6 of 28
what is the symbol for luminance
cd (candela)
7 of 28
symbol for current
A (ampere)
8 of 28
what are the benefits of using the mode as an ICT
quick to find, unaffected by extreme scores, can be used with any LoM
9 of 28
what are the disadvantages of using the mode
terminal statistic (can't be used further), a sub group (e.g. very poor people) can make it unrepresentative
10 of 28
benefits of using the median as an ICT
unaffected by extreme scores, can be used at all levels above nominal
11 of 28
disadvantages of using the median
ignores value/magnitude, only considers order
12 of 28
benefits of using the mean as an ICT
very sensitive measure, accounts for all the data, can be combined with other data to find an overall mean
13 of 28
disadvantages of using the mean
very sensitive, can only be used with interval/ratio data, can only be used when scores are symmetrical about the mean (bell curve)
14 of 28
who first described and who applied the normal distribution curve
De Moivre, Frederick Gauss
15 of 28
what are the 5 characteristics of a Normal Distribution curve
naturally occurring, theoretically asymptotic, symmetrical about ICT, point of inflection is the first SD, the Z score is equal to the SD
16 of 28
what is the name given to a normal ND curve
mesokurtic
17 of 28
to a pointy curve
leptokurtic
18 of 28
to a flat curve
platykurtic
19 of 28
generally, for a negatively skewed curve, from R to L what order will the ICTs be
mean, median, mode at the highest point
20 of 28
give 4 reasons why distribution is important. 1)
determines which ICT to use
21 of 28
2)
determines which measure of variability to use
22 of 28
3)
provides a Z score for standardised comparisons
23 of 28
4)
determines parametric (powerful, uses normal ND curves) or non-parametric stats analysis
24 of 28
give the 5 measures of variability
SD, SEM, range, IQR and nCI
25 of 28
where should you use SD and where should you use SEM. what if results are non-normal
SD in the method and SEM in the results. use IQR and median if results are non normal
26 of 28
what is SEM
a measure of sampling error - how accurate our mean is
27 of 28
what is nCI
normalised Confidence Interval represents the difference between means
28 of 28

Other cards in this set

Card 2

Front

list the levels of measurement

Back

nominal, ordinal, interval, ratio

Card 3

nominal scale

Back Card 4

ordinal scale

Back Card 5

interval scale

Back 