Stat 263 Flashcards

Front 1

Leptokurtic Curve

Back 1

More values are in the centre than a normal curve or less values in the tail

Front 2

Platykurtic Curve

Back 2

Fewer values in the centre than a normal curve or more values in the tails

Front 3

Formula for approximate median or any othre fractile in grouped data

Back 3

x = L + (j/f)c

L = lower boundary of the class

f = frequency of this class

j = index value - frequency of previous class

c = class interval

Front 4

index value of median or any percentage

Back 4

index (med) = (1/2) (n + 1)

index (%) = (x/100) (n + 1)

Front 5

Postively Skewed

Back 5

Positive: median is to the left of the mean (the tail is longer on the right)

Front 6

Negatively Skewed

Back 6

the median is right of the mean (longer tail on the left)

Front 7

Simple Box Plot

Back 7

1. Left box end = Q1
2. Right box end = Q3.
3. Line in box = mean
4. Length of box = Q3-Q1
5. Step Size (1.5 * IQR)
6. Upper and lower "whiskers" correspond to the largest value within one step
7. Outlier symbols (dot for 1 or 2 steps out. Blank dot with X if >2 steps out)

Info:
1. Center of data
2. Variation of spread
3. Skewness
4. Presence of outliers

Front 8

Five Number Summary

Back 8

1. Min.
2. Q1
3. Median
4. Q3
5. Max.

Front 9

Statistic vs Parameter

Back 9

Stat: what the experimenter knows. DATA

Parameter: what the experimenter wants to know.

Front 10

Confirmation Bias (Myside Bias)

Back 10

Tendency for people to favour information that confirms their pre-conceptions or hypotheses

Front 11

Optimism Bias (positivity illusion)

Back 11

People judge that their chances of experiencing a good outcome is higher than average

Front 12

Marginal and Joint Frequences

Back 12

Marginal: row and column totals in a 2 way freq. diagram.

Joint: cell frequencies in the table

Pg.8.

Front 13

Frequency Ogive

Back 13

S-shaped curve for cumulative or cumulative relative frequency histograms!

Front 14

Pareto Diagram.
The Pareto Relationship.

Back 14

bar charts with the categories listed in the same order as their frequencies.

Pareto relationship: 80-20 rule. 80% of height in the first bar. Where 80% of outcome is from 20% of problem (assuming 5 categories).

Front 15

Formula for approx. # of bins and for bin width for data set (200 values or lower)

Back 15

# bin = (sq.root(n))

bin width = range/(sq.root(n))

Front 16

Nominal Data

Back 16

Categorical Data.

Data that gives a name or a number to individuals or mutually exclusive categories.

(coded numbers, but no arithmetic)

Front 17

Ordinal Data

Back 17

Mutually exclusive categories with fixed order.

Letter grades. First, second, third.

Front 18

Interval vs Ratio Data

Back 18

Mutually exclusive, with fixed order, and equal spacing between categories.

Interval: 0 is meaningless.
Ratio: 0 means absence of.

Wont be tested on exam.

Front 19

Lurking Variable

(confounding variable or factor)

Back 19

variable that has an important effect on the relationship among the variables in the study, but is not included.

Front 20

Operational Definition

Back 20

A clear, concise, detailed defintion of the measure or data set

Front 21

Categorical vs Numerical Data

Back 21

Qualitative vs Quantitative.

Only Nominal data is numerical and categorical