Statistics Chapters 1-3
1.1
Statistical significance: Achieved if the chances of an event occurring by chance is 5% or less.
Sample Data Reported instead of measured: When collecting data, it’s better to take measurements yourself rather than asking the subject to self report.
1.2
Quantitative, or numerical: Data consisting of #’s representing counts or measurements. (Numbers represent counts or measures, meaningful calculations can be performed).
Ex. Weight of supermodels.
Categorical, or qualitative: Data consisting of names or labels.
Ex. Shirt numbers on jerseys.
Discrete Data: Data values are quantitative and the number of values is finite, or ‘countable’. (Whole numbers).
Continuous Data: Result from infinitely many possible quantitative values, collection of data not countable. (Usually Decimals.)
characterized by data that consist of names,
Nominal: labels, or categories only, and the data cannot be arranged in some
natural order. (Style of beer: pilsner, lager, IPA, Saison, stout)
Ordinal: involves data that can be arranged in some
order, but differences between data values either cannot be determined
or are meaningless. (Rating of beer: scale of 1 to 5)
Interval: involves data that can be ordered, and the
differences between data values can be found and are meaningful.
However, there is no natural zero starting point at which none of the
quantity is present. (Fahrenheit temperature of beer: 44°F, 49°F, 52°F)
Ratio: ________________ level - data can be ordered, differences can be
found and are meaningful, and there is a natural zero starting point
(where none of the quantity is present). Differences and ratios are both
meaningful. (Size of beer: 4 oz., 12 oz., 16 oz., 20 oz.)
1.3
Simple Random Sample: Random.
Systematic Sampling: Every kth term.
Stratified Sample: Subdivide populations, take a little out of everything. Some out of all.
Cluster Sample: Pick a few out of a cluster, take all of the selected. All from some.
Cross-sectional: Data collected all at one point in time.
Retrospective: Data collected from the past.
Prospective: Data collected in the future from groups sharing common factors.
Midpoint: Add lower class limit to upper class limit and divide by two.
Class width: Subtract two consecutive lower class limits.
Class width = Max value - min value / # of classes
Round up for a convenient number.
Relative Frequency = Frequency / Totaled frequencies = Relative ( * 100 to make percentage).
Cumulative Frequency = Frequency for each class sum of all frequencies for that class + previous classes.
2.2
Skewed to the right: Mean > Median.
Skewed to the left: Median < Mean.
Skewness and the mean both follow the tail.
Mean not resistant to outliers.
Median resistant to outliers.
Quantitative Graphs:
Frequency Polygon, Time-series Graph, Relative frequency polygon, scatterplot, dot plot, stem plot.
Qualitative Graphs:
Pie chart, Bar chart, pictographs.
Non-zero vertical axis: Bottom point does not start at 0.
Measures of Center: Value at the center of a data set.
Resistant: Will not change in the presence of extreme outliers.
Mean: Sum of all values / Totaled # of values
Median: Middle value; sort all values,
If odd → Median = exact middle.
If even → Mean of the two middle numbers.
Mode: Value that occurs most often.
Unimodal = 1 mode.
Bimodal = 2 modes.
Multimodal = Multiple modes.
Midrange: max + min / 2. [Not resistant].
3.2
Range = max - min.
Standard deviation: Set of sample values, measure of how much data deviates from mean. Never negative.
Variance [Sample]: s², set of values equal to square of standard deviation.
Significantly low: Mean - 2(Standard Deviation).
Significantly high: Mean + 2(Standard Deviation).
3.3
Z-Score: Number of standard deviations a value is above or below the mean.
Significantly low if equal to -2 or >.
Significantly high if greater or = to 2.
If data less than mean → Z-score = (-) Negative.
Percentiles are measures of location.
Data Value → Percentile
Percentile of value x = # of values less than x / total number of values * 100 [Remember to sort data!]
Round percentile to nearest whole number.
Percentile → Data
L = (k / 100) * n
k = percentile (40th percentile → k = 40)
n = size of data (30 values, n = 30)
L = Locator, if L rounds to 12, find 12th value in ordered data.
If whole number, (Lth) + (L+1) / 2
Quartiles: Measures of location.
Q1 = 1st quartile = P25
Q2 = 2nd Quartile = P50 [Median]
Q3 = 3rd Quartile = P75
Interquartile Range: IQR = Q3 - Q1
5-Number Summary: Min, Q1, Median, Q3, Max.