Chapter 2: Modeling Distributions of Quantitative Data Flashcards
Define Percentiles
The percent of data “less than or equal to” a certain data value
Interpretation of Percentiles
__Percentile__ % of __context__ are less than or equal to __value__.
ex. 1 = __75__% of __high school student SAT scores__ are less than or equal to __1200__.
ex. 2 = The SAT score of __1200__ is at __75__th percentile of __high school student SAT scores__.
Define Standardized Scores (Z-scores)
measures how many “standard deviations” a data point is “above/below” the mean.
(data point - mean)/(standard deviation)
Define Standardization
a point’s location in the distribution depends of “both” distance from the “center” and the distribution’s “spread” or “variation”
Interpretation of Z-Score
__specific value with context__ is __z-score__ standard deviations __above/below__ the mean.
- above when z-score is positive
- below when z-score is negative
ex. __A quiz score of 71__ is __1.43__ standard deviations __below__ the mean. (z = -1.43)
Description of Cumulative Relative Frequency Graph
y- axis = cumulative proportion (0-1)
x- axis = variable
can find IQR or percentiles by drawing dotted lines and following the graph.
Transforming Data (addition/subtraction)
Mean = add/subtract
S.D. = X change
Shape = X change
Transforming Data (multiply/divide)
Mean = multiply/divide
S.D. = multiply/divide
Shape = X change
Define Density Curve
a curve that
1. is always on or above the horizontal axis
2. has area exactly 1 underneath it
describes the overall pattern of a distribution. The area under the curve and above any interval of values on the horizontal axis is the proportion of all observations that fall in that interval.
Identifying mean and median of a density curve
mean = balancing point
median = equal area point
Define Normal Curve
- symmetric and bell shaped
- the mean=median, both located at the exact center
Define Empirical Rule
a.k.a the 68-95-99.7 Rule
The normal curves below shows markings 1, 2, and 3 standard deviations away from the mean. For each, shade the middle area and write the percent of data covered in that middle area.
1 SD away on both sides = 68 %
2 SD away on both sides = 95 %
3 SD away on both sides = 99.7 %
Strategy of Normal Curve Problems
- Draw and Label curve (include mean, SD, tick marks, shade area/point in question)
- perform calculations (show work)
- answer the question with context
Using standard normal table
ex. z<2.45
–> draw a normal curve with N(0,1)
–> since it is “<”, draw a line of 2.45 above the mean of 0 and shade left side of the line –> find the area
Normal Distribution and the Calculator
- when finding percentages from values, use [Normal cdf]
- when finding values from percentages, use [inv norm]
Assessing Normality
checking normality using norm cdf or inv norm
- find X bar and S of given data set
- find IQR and calculate IQR dance values for 1 sd, 2 sd, and 3 sd.
- find proportion of data in the interval of 1 sd, 2 sd, and 3 sd. Compare with 68, 95, 99.7 %
- When Normal Probability Plot is checked on calculator, the graph should be linear.
Normal Probability Plot
scatter plot of the ordered pair (data value, expected z-score) for each of the individuals in a quantitative data set.
- enter values in sheets
- plot them
- menu - 1 - 4 and check the linearity